home *** CD-ROM | disk | FTP | other *** search
Text File | 1993-12-31 | 227.2 KB | 5,200 lines |
- 1-Nov-87 02:09:00-EST,2578;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 1 Nov 87 02:09-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA22690@EDDIE.MIT.EDU>; Sat, 31 Oct 87 22:18:37 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA22671@EDDIE.MIT.EDU>; Sat, 31 Oct 87 22:18:17 EST
- Received: from Taurus.Tau.Ac.IL by wiscvm.wisc.edu ; Sat, 31 Oct 87 13:34:27 CDT
- Received: by Taurus.Tau.Ac.IL (5.51/ta.1.4)
- id AA01885; Sat, 31 Oct 87 14:02:20 +0200
- Comments: You can reach this host both as TAURUS.TAU.AC.IL
- and TAURUS.BITNET The first form is preferred.
- Return-Path: <ofer@Taurus.Tau.Ac.IL>
- Date: Sat, 31 Oct 87 14:02:20 +0200
- From: Ofer Lapid 4X6OJ <ofer@Taurus.Tau.Ac.IL>
- Message-Id: <8710311202.AA01885@Taurus.Tau.Ac.IL>
- To: info-hams@simtel20.arpa, packet-radio@eddie.mit.edu
- Subject: Packet Request from 4XNET
-
- From: Ofer Lapid 4X6OJ <ofer@taurus.BITNET> || <ofer@taurus.TAU.AC.IL>
- To: TCP/IP operators and wizards.
- Subject: TCP/IP over packet radio in Israel's 4XNET
-
- Hello netpeople,
- I call you with an urgent request.
- We in Israel are quite new to the subject of Packet Radio, nevertheless
- we have been improving our network rapidly since the summer of 86 when we
- built the first PacCom TNC-200 kits.
- Today we have more than 70 users all around the country and the number
- is increasing everyday.
- We are successfully operating some 6 WA7MBL PBBS software version 320
- and running some servers around it to aid the distribution of binary
- files accross the country, we even have some ham from Egypt who uses our
- PBBS's frequently.
- We have been following the advances of the group lead by KA9Q implementing
- the TCP/IP on packet and decided to adopt it to our networking as well.
- Until last spring SIMTEL20 was responsive to bitnet mail requests of files
- and we got the recent versions from there, now that the SIMTEL20 is *NOT*
- responding to mail requests we *CANNOT* be updated and cannot advance with
- the rest of the packet community.
-
- I am calling for every ham in big USA who can help us solving this
- matter and establish a strong, long lasting link through which software
- updates would flow to us from the large continent.
-
- We have tried sending self addressed diskettes over the mail but it
- came out very *SLOW* *UNRELIABLE* and *SQUASHED* ( the media not the software).
-
- WE CANNOT FTP TO ANY ARPA HOST ONLY PLAIN E-MAIL...
-
- I will thank and appreciate any mail reply to this call
- Ofer Lapid 4X6OJ .
-
- r mylogo
- 2-Nov-87 19:16:57-EST,2816;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 2 Nov 87 19:16-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA21053@EDDIE.MIT.EDU>; Mon, 2 Nov 87 14:21:39 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA21046@EDDIE.MIT.EDU>; Mon, 2 Nov 87 14:21:28 EST
- Message-Id: <8711021921.AA21046@EDDIE.MIT.EDU>
- To: packet-radio@EDDIE.MIT.EDU
- Cc: clements@LF-SERVER-2.BBN.COM
- Subject: Re: PBBSes and TCP/IP
- Date: Mon, 02 Nov 87 14:23:38 -0500
- From: clements@LF-SERVER-2.BBN.COM
-
-
- < In article <8710282317.AA03565@june.cs.washington.edu>,
- < Gary Delong <linus!raybed2!cvbnet!gdelong@EDDIE.MIT.EDU> writes:
- < In article <4286@well.UUCP>, tenney@well.UUCP (Glenn S. Tenney) writes:
- < >
- < > Now the questions:
- < > 1. W0RLI software *sucks* compared to FIDO/OPUS. Is there any
- < > decent PBBS software out there? Where?
- < > 2. Is there any existing software to run with the TCP/IP code
- < > that would let people use it as a BBS (ie. conferencing)?
- < > Ideally, any readnews software (I mean, netnews is better than
- < > W0RLI by a looong shot!)?
- < >
- < > Were these stupid questions or what?
- < > tnx
- < > Glenn Tenney
- < > AA6ER
- <
- < No, Glenn, they aren't stupid questions, just phrased in such a way that anyone
- < who spent thousands of hours of his own time developing code that is given away
- < would have to be pissed.
- < ...
- < Remember, W0RLI is a person, not a product.
- <
- < 73s
- < _____
- < / \ / All spelling errors | Gary A. Delong, N1BIP
-
- I gave Hank, W0RLI, copies of the above two messages at dinner Friday night.
- He was not pissed. After all, he has been in the PBBS business for a LONG
- time now, and has received a LOT of feedback of all kinds.
-
- He did comment that from Glenn's location it would be impossible to
- get any decent throughput, whether AX.25 or TCP. Serious hidden
- terminal problems, congested channels and digipeater disasters.
- He also mentioned that Glenn seemed to be unwilling to type H<return> for
- help, no matter how many times the system suggested he do so. I will
- certainly agree that the system's command interface leaves a lot to be
- desired, though, and I've pushed hard on Hank to improve it over
- the years, with only partial success...
-
- Hank did mention that he HAS become ticked off at two people, though,
- one JA op and one 2-lander named Brian something, who have
- taken Hank's code, modified it slightly and claimed it as their own,
- with copyright statements, even.
- This may be the straw that finally causes Hank to stop work on the
- system. [But he has been claiming he was through developing the BBS
- for years...]
-
- He gave me a copy of version 4.1, hot off the presses. I'll get it
- posted to SIMTEL shortly.
-
- 73, Bob, K1BC clements@bbn.com
- 3-Nov-87 04:28:11-EST,1057;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 3 Nov 87 04:28-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA04351@EDDIE.MIT.EDU>; Tue, 3 Nov 87 00:43:26 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA04331@EDDIE.MIT.EDU>; Tue, 3 Nov 87 00:41:41 EST
- Date: Mon, 2 Nov 1987 22:41 MST
- Message-Id: <KPETERSEN.12347568901.BABYL@SIMTEL20.ARPA>
- Sender: KPETERSEN@SIMTEL20.ARPA
- From: Keith Petersen <W8SDZ@SIMTEL20.ARPA>
- To: packet-radio@EDDIE.MIT.EDU
- Subject: W0RLI C PBBS V 4.1 now available from SIMTEL20 (correction)
-
- The directory name in the previous announcement was incorrect.
- It showed device PS:. It should have read PD: That was my fault.
- the corrected list is below.
-
- --Keith W8SDZ
-
- Filename Type Bytes CRC
-
- Directory PD:<MSDOS.PACKET>
- IO41.ARC BINARY 9793 3224H
- IOSRC41.ARC BINARY 36076 110CH
- MB41.ARC BINARY 140294 56D8H
- MB41READ.ME ASCII 2344 7552H
- MBSRC41.ARC BINARY 91852 FAA2H
-
- Again, MBBIOS.ARC is unchanged.
-
- MBBIOS.ARC BINARY 32256 E594H
- 3-Nov-87 05:56:42-EST,3583;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 3 Nov 87 05:56-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA04148@EDDIE.MIT.EDU>; Tue, 3 Nov 87 00:30:49 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA04124@EDDIE.MIT.EDU>; Tue, 3 Nov 87 00:30:03 EST
- Resent-Message-Id: <8711030530.AA04124@EDDIE.MIT.EDU>
- Date: Monday, 2 November 1987 21:31-MST
- Message-Id: <KPETERSEN.12347563981.BABYL@SIMTEL20.ARPA>
- Sender: clements@BBN.COM
- From: clements@BBN.COM
- To: w8sdz@SIMTEL20.ARPA
- Subject: W0RLI C PBBS V 4.1 now available from SIMTEL20
- Resent-From: KPETERSEN@SIMTEL20.ARPA
- Resent-To: packet-radio@EDDIE.MIT.EDU
- Resent-Date: Mon 2 Nov 1987 22:14-MST
-
- I have just uploaded version 4.1 of the W0RLI C PBBS to SIMTEL20.
-
- Filename Type Bytes CRC
-
- Directory PS:<MSDOS.PACKET>
- IO41.ARC BINARY 9793 3224H
- IOSRC41.ARC BINARY 36076 110CH
- MB41.ARC BINARY 140294 56D8H
- MB41READ.ME ASCII 2344 7552H
- MBSRC41.ARC BINARY 91852 FAA2H
-
- Again, MBBIOS.ARC is unchanged.
-
- MBBIOS.ARC BINARY 32256 E594H
-
- Sorry to ship you a new one right after you announced V4.0, but Hank
- handed me a disk on Friday and it's my duty to spread the wealth...
-
- 73, Bob, K1BC
-
- [Extract from the README file in MB41.ARC]
-
- The W0RLI / VE3GYQ C BBS
-
- Release notes for C BBS Version 4.1 - 10/15/87
-
- See the file CHANGES.MB for details on new features included
- and bugs fixed in this release.
-
-
- *** Warning ***
-
- If you are now running a version prior to 4.0 you must convert
- your existing message and user files to version 4 format.
-
- First, backup all files in the directory used by the MailBox.
- In case something goes wrong, you will be able to restore
- everything to where it was without loss of messages.
-
- The conversion from earlier versions to version 4 is done by MBCONV.
- Set the default directory to the directory containing the
- MAIL.DAT and USER.DAT files. Then run MBCONV.
-
- If you are running a version prior to 3.0 you must first make
- certain that the message text has been moved to individual files.
- Do this by executing the GF command.
-
- *** Note ***
-
- COMXBIOS has been replaced by PIPE. The old COMXBIOS device
- driver will not work with versions of the MailBox beyond 4.0
-
- Check that any sub-directory you use in config.mb exists,
- the existance of most directory / device paths is NOT checked.
-
- ... Hank
-
- [Extract from the CHANGES.MB file]
-
- Changes from V4.0 to V4.1 Starting 10/1/87
-
- Removed the "max # users in user file" from config.mb
- Separate times for "old" for NTS, user, bulletins.
- Would not rev fwd if target system had script in fwd.mb
- Removed /EX as end-of-msg, it was a bad idea.
- COMXBIOS replaced by PIPE. Generalizes the function to 4 ports.
- "monitor" "watch user" "see TNC commands" per port.
- Added DOS escape: ! command and ! list in fwd.mb.
- Got tired of seeing all the commands to the tnc, now not echo to screen.
- Zip code was not justified correctly in user record.
- Repaired "defered remote sysop command.
-
- Changes from V3.5 to V4.0 Starting 9/19/87
-
- Fixed bug: download to a tnc did not return to terminal mode properly.
- Removed 256 character limit on greeting message.
- Added message state "H" (held), hold list, LH command.
- Much better handling of disconnect of "unhappy" tnc (from ve3gyq).
- Various bugs that showed up in 3.5 are fixed.
- Yet another change to DD timeslice handling.
- Message and User files version 8
- ----------------------------------------
- 3-Nov-87 21:29:26-EST,2148;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 3 Nov 87 21:29-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA22192@EDDIE.MIT.EDU>; Tue, 3 Nov 87 18:31:53 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA22169@EDDIE.MIT.EDU>; Tue, 3 Nov 87 18:30:58 EST
- Received: by june.cs.washington.edu (5.52.1/6.10)
- id AA05145; Tue, 3 Nov 87 15:32:05 PST
- Return-Path: <uunet!mcvax!nikhefk!henkp@eddie.mit.edu>
- Message-Id: <8711032332.AA05145@june.cs.washington.edu>
- Date: 31 Oct 87 20:00:03 GMT
- From: uunet!mcvax!nikhefk!henkp@EDDIE.MIT.edu (Henk Peek)
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: (Belgium) Packet Radio Seminar
-
-
- (Belgium) PACKET RADIO SEMINAR
- ------------------------------
-
- The Packet Work Group organize on 7 november a packet radio
- semainar at the Antwerp University Clininc Auditorium (AZU).
-
- 9h45 10h00 Welcom address
-
- 10h00 11h00 Networking via TCP/IP
- G.J. van der Grinten, PA0GRI
-
- 11h30 12h30 Rhein-Main Network controller
- Klaus Dieter Friedrich, DH1FAB
- Rhein-Main Packet Radio Group
-
- 14h00 15h00 Packet radio experiment on bord of phase IIIc
- Dipl-Ing Hans Peter Kuhlen, DK1YQ
- Rudak group
-
- 15h30 16h30 Packet message forwarding in the UK. packet network protocols
- Jeff Ward, G0/K8KA
- Uosat Spacecraft Engineering Unit
- Universitery of Surrey
-
- 17h00 18h00 Debate and conclusion
-
- Free admittance to the surronding demonstration and meeting area.
- We have to charge 250 BEF covering the admittance to the auditorium
- (lectures), the complete seminar documentation and coffee breaks at
- regulair intervals. Luncheon can be provided at 250 BEF. Payment
- is required in advance for lectures and/or lunch:
- acct. 441-9610359-38 PWZ vzw, Teanisstraat 30, 9920 Lovendegem, Belgium
-
- Information: Willy Wittseale, ON1AWU, Ter Borchtlaan 32, B2520 Edegem, Belgium.
-
- ===============================================================================
-
- Henk Peek PA0HZP
-
- Mail: NIKHEF-K, postbox 4395, 1000 AJ Amsterdam, Holland
- UUCP: uunet!mcvax!nikhefk!henkp.UUCP
- BITNET: U00234@HASARA5
-
-
- 3-Nov-87 22:18:06-EST,1807;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 3 Nov 87 22:18-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA22121@EDDIE.MIT.EDU>; Tue, 3 Nov 87 18:28:20 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA22117@EDDIE.MIT.EDU>; Tue, 3 Nov 87 18:28:00 EST
- Received: by june.cs.washington.edu (5.52.1/6.10)
- id AA04977; Tue, 3 Nov 87 15:29:24 PST
- From: sun!imagen!atari!portal!cup.portal.com!Patrick_J_Fagan@EDDIE.MIT.edu
- Return-Path: <sun!imagen!atari!portal!cup.portal.com!Patrick_J_Fagan@eddie.mit.edu>
- Message-Id: <8711032329.AA04977@june.cs.washington.edu>
- Date: 31 Oct 87 21:40:21 GMT
- Newsgroups: rec.ham-radio.packet
- Subject: Re: Information exchange
- References: <7283@eddie.MIT.EDU>
- Xportal-User-Id: 1.1001.2475
- Apparently-To: packet-dist@EDDIE.MIT.EDU
-
- GateWay was made available in ELECTRONIC FORM by Jon Bloom, KE3Z
- who is the former editor of this newsletter. Since he is no longer
- charged with that responsibility, the files have not been made read-
- ily available to him. This is why the ARRL Letter and GateWay have
- been hard to find on Packet/LL BBS's around the country. There have
- been a few industrious souls that have taken on the monumental task
- of manually keying in the text from the printed edition. Most have
- given up after a few issues when learning of the amount of time re-
- quired.
-
- These periodicals are very reasonably priced. So much so, the ARRL
- doesn't even cover the cost of production with revenue from paid
- subscriptions. This may be another reason why it hasn't shown up
- in electronic form on PORTAL and elsewhere as of late. Giving it
- away on BBS's hinders the incentive to subscribe. Bob, I hope this
- will help answer your question.
-
- 73, Pat - WA5DVV @ WA5DVV
- sun!cup.portal.com!WA5DVV
-
- 3-Nov-87 22:48:00-EST,2878;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 3 Nov 87 22:47-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA22042@EDDIE.MIT.EDU>; Tue, 3 Nov 87 18:23:05 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA22037@EDDIE.MIT.EDU>; Tue, 3 Nov 87 18:22:52 EST
- Received: by june.cs.washington.edu (5.52.1/6.10)
- id AA04629; Tue, 3 Nov 87 15:24:17 PST
- Return-Path: <husc6!bbn!rochester!udel!princeton!idacrd!mac@eddie.mit.edu>
- Message-Id: <8711032324.AA04629@june.cs.washington.edu>
- Date: 3 Nov 87 18:29:58 GMT
- From: husc6!bbn!rochester!udel!princeton!idacrd!mac@EDDIE.MIT.edu (Bob McGwier)
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: Re: PBBSes and TCP/IP
- References: <7306@eddie.MIT.EDU>
-
- > I gave Hank, W0RLI, copies of the above two messages at dinner Friday night.
- > He was not pissed. After all, he has been in the PBBS business for a LONG
- > time now, and has received a LOT of feedback of all kinds.
- >
- >
- > Hank did mention that he HAS become ticked off at two people, though,
- > one JA op and one 2-lander named Brian something, who have
- > taken Hank's code, modified it slightly and claimed it as their own,
- > with copyright statements, even.
-
- I would think that Bob would agree that I have known Hank for a long
- time and I do know Brian and I admire Hank as much as anybody who has
- ever met him but this is simply going too far. "Modified slightly" is
- simply not at all accurate. I would say that the similarities are
- outnumbered by the differences now. ON EVERY occassion I have EVER
- heard Brian talk about the code he says, "PRMBS is based on the CBBS by
- W0RLI and with many additions made by me". He in fact has EVERY right
- to copyright ANY code that is his own work so long as credit is given
- where credit is due. Brian does not get enough credit as far as I am
- concerned for ideas contributed to the "other one" either.
- Brian and Hank are "warring" over those damn headers again and this
- should not be allowed to spill over into other things. Frankly, Brian's
- additions to that code have made it the most full featured BBS for
- packet radio of any I know.
-
- BBS's are and should be only a stop gap to a more general object that is
- capable of running tasks at the behest of the user and sending files,
- mail, etc at request. We need a better network for this and better
- networking software. I believe that we have the bedrock for this in
- KA9Q's TCP/IP. Brian is making a valuable contribution to the existing
- networking arrangements and it behooves us not to have statements like
- this made.
-
- I am sure Bob feels loyalty to our old friend from New England but care
- should be taken when making flat statements like this one.
-
- NN2Z, Dave has also made contributions to Brian KA2BQE's efforts. I for
- one wish all this childish bickering would cease.
-
- Bob McGwier N4HY
-
-
-
- 3-Nov-87 23:46:29-EST,2213;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 3 Nov 87 23:46-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA22229@EDDIE.MIT.EDU>; Tue, 3 Nov 87 18:34:35 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA22224@EDDIE.MIT.EDU>; Tue, 3 Nov 87 18:34:21 EST
- Received: by june.cs.washington.edu (5.52.1/6.10)
- id AA05273; Tue, 3 Nov 87 15:35:46 PST
- Return-Path: <cornell!batcomputer!mitch@eddie.mit.edu>
- Message-Id: <8711032335.AA05273@june.cs.washington.edu>
- Date: 29 Oct 87 12:46:38 GMT
- From: cornell!batcomputer!mitch@EDDIE.MIT.edu (Mitch Collinsworth)
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: Hurtin' TNC
- Keywords: GLB TNC-2A, TNC2 clone, doesn't transmit
-
- Hi all. After being quite inactive on packet during the summer months,
- I returned to the rig a few weeks ago and found that I could not coerce
- my GLB TNC-2A to transmit with any of the normal commands. The only
- command that would cause it to transmit is the one for adjusting the
- tones that puts it key-down and lets you alternate between the tones
- until you shut it off (I forget the name of the command). Other than
- that it seems to be functioning "normally." When you C W2XYZ, it
- does nothing for approximately the right period of time and then gives
- the normal messages you would see if it had actually transmitted the
- connect packets, but got no response. It receives fine and if someone
- tries to connect to me, the CON light comes on and I get the connected
- message, but it never acknowledges the connect.
-
- I first thought it might be seeing something on the receive line that
- would look like another transmitter and keep it from interrupting, but
- I put a MFJ TNC in its place and it works fine.
-
- It was working just fine until I ignored it for the summer while
- persuing outdoor activities. Anyone have any ideas what might cause a
- problem like this. I thought it was simply a bad connection somewhere
- until I found that it actually would transmit with the above-mentioned
- test command. Could this be simply a parameter that needs to be changed?
- Don't assume I haven't overlooked the obvious, as that could very well
- have happened.
-
- -Mitch Collinsworth, K2VD
-
-
- 4-Nov-87 00:39:03-EST,1899;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 4 Nov 87 00:39-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA22891@EDDIE.MIT.EDU>; Tue, 3 Nov 87 19:20:25 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA22872@EDDIE.MIT.EDU>; Tue, 3 Nov 87 19:20:03 EST
- Message-Id: <8711040020.AA22872@EDDIE.MIT.EDU>
- To: packet-radio@EDDIE.MIT.EDU
- Cc: clements@LF-SERVER-2.BBN.COM
- Subject: Oh, noooo!!! Get the fire extinguisher!!!
- Date: Tue, 03 Nov 87 19:18:57 -0500
- From: clements@LF-SERVER-2.BBN.COM
-
- In message <325@idacrd.UUCP> mac@idacrd.UUCP (Bob McGwier) writes:
-
- < I am sure Bob feels loyalty to our old friend from New England but care
- < should be taken when making flat statements like this one.
- < ...
- < I for
- < one wish all this childish bickering would cease.
-
- < Bob McGwier N4HY
-
- Good grief! I seem to have hit a sore point! My apologies.
-
- First, I said clearly that I was repeating a comment from Hank,
- not making a flat statement of my own. I cannot swear that he used
- the word "slightly", so I may have misrepresented him to that extent,
- but that was the impression I got.
-
- Second, this is the first I had heard of ANY complaints along
- these lines, so I think that the "all this childish bickering"
- comment is out of place, certainly for this newsgroup. There
- hasn't been any "all" except for one message.
-
- It is certainly true that I know nothing about Brian's system. I
- don't think there are any copies of it running in New England.
- I, for one, have stopped running a PBBS at all. It ain't worth
- it.
-
- In any case, forget I ever mentioned it. I had no intention of
- starting (or continuing) a war. Apparently I stepped on a land
- mine that I didn't know was there.
-
- 73, (running to hide in a hole somewhere)
- Bob, K1BC, clements@bbn.com
- 4-Nov-87 01:23:37-EST,2878;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 4 Nov 87 01:23-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA22042@EDDIE.MIT.EDU>; Tue, 3 Nov 87 18:23:05 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA22037@EDDIE.MIT.EDU>; Tue, 3 Nov 87 18:22:52 EST
- Received: by june.cs.washington.edu (5.52.1/6.10)
- id AA04629; Tue, 3 Nov 87 15:24:17 PST
- Return-Path: <husc6!bbn!rochester!udel!princeton!idacrd!mac@eddie.mit.edu>
- Message-Id: <8711032324.AA04629@june.cs.washington.edu>
- Date: 3 Nov 87 18:29:58 GMT
- From: husc6!bbn!rochester!udel!princeton!idacrd!mac@EDDIE.MIT.edu (Bob McGwier)
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: Re: PBBSes and TCP/IP
- References: <7306@eddie.MIT.EDU>
-
- > I gave Hank, W0RLI, copies of the above two messages at dinner Friday night.
- > He was not pissed. After all, he has been in the PBBS business for a LONG
- > time now, and has received a LOT of feedback of all kinds.
- >
- >
- > Hank did mention that he HAS become ticked off at two people, though,
- > one JA op and one 2-lander named Brian something, who have
- > taken Hank's code, modified it slightly and claimed it as their own,
- > with copyright statements, even.
-
- I would think that Bob would agree that I have known Hank for a long
- time and I do know Brian and I admire Hank as much as anybody who has
- ever met him but this is simply going too far. "Modified slightly" is
- simply not at all accurate. I would say that the similarities are
- outnumbered by the differences now. ON EVERY occassion I have EVER
- heard Brian talk about the code he says, "PRMBS is based on the CBBS by
- W0RLI and with many additions made by me". He in fact has EVERY right
- to copyright ANY code that is his own work so long as credit is given
- where credit is due. Brian does not get enough credit as far as I am
- concerned for ideas contributed to the "other one" either.
- Brian and Hank are "warring" over those damn headers again and this
- should not be allowed to spill over into other things. Frankly, Brian's
- additions to that code have made it the most full featured BBS for
- packet radio of any I know.
-
- BBS's are and should be only a stop gap to a more general object that is
- capable of running tasks at the behest of the user and sending files,
- mail, etc at request. We need a better network for this and better
- networking software. I believe that we have the bedrock for this in
- KA9Q's TCP/IP. Brian is making a valuable contribution to the existing
- networking arrangements and it behooves us not to have statements like
- this made.
-
- I am sure Bob feels loyalty to our old friend from New England but care
- should be taken when making flat statements like this one.
-
- NN2Z, Dave has also made contributions to Brian KA2BQE's efforts. I for
- one wish all this childish bickering would cease.
-
- Bob McGwier N4HY
-
-
-
- 6-Nov-87 15:18:23-EST,1061;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 6 Nov 87 15:18-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA14469@EDDIE.MIT.EDU>; Fri, 6 Nov 87 09:52:17 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA14433@EDDIE.MIT.EDU>; Fri, 6 Nov 87 09:50:53 EST
- Date: Fri 6 Nov 87 09:43:50-EST
- From: Jeff Markel <HAZELTINE@A.ISI.EDU>
- Subject: Removal from distribution
- To: packet-radio@EDDIE.MIT.EDU
- Cc: hazeltine@A.ISI.EDU
- Message-Id: <12348454018.31.HAZELTINE@A.ISI.EDU>
-
- Gentlemen:
-
- Please remove me immediately from the Packet Radio (amateur)
-
- distribution. Your discussion are becoming too voluminous
-
- for me to keep up with.
-
-
-
- I will attempt to reinstate myself in the future. Right now I am
-
- picking up Program Manager responsiblity from J.L. Markel
-
- and it is time consuming enough to keep up with the Program's
-
- Arpanet mail. Thanks for your help.
-
-
-
- Donna Linke-Klein
-
- -------
- 7-Nov-87 14:56:59-EST,1468;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 7 Nov 87 14:56-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA18888@EDDIE.MIT.EDU>; Sat, 7 Nov 87 13:14:27 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA18880@EDDIE.MIT.EDU>; Sat, 7 Nov 87 13:14:16 EST
- Received: from ROOK.SCRC.Symbolics.COM by VALLECITO.SCRC.Symbolics.COM via CHAOS with CHAOS-MAIL id 175685; Sat 7-Nov-87 13:14:50 EST
- Date: Sat, 7 Nov 87 13:14 EST
- From: Henry Minsky <hqm@VALLECITO.SCRC.Symbolics.COM>
- Subject: Repeater advice anyone?
- To: packet-radio@eddie.mit.edu
- Message-Id: <871107131436.4.HQM@ROOK.SCRC.Symbolics.COM>
-
-
- It happens that there is a 440 Mhz repeater ready and waiting to be
- installed here at MIT, which Prof. Steve Ward once got from Motorola
- for the express purpose of being a digital communications repeater.
- Some people from the MIT w1mx radio club have experience with an
- identical piece of equiptment, and are willing to help set up a packet
- only repeater. I know almost nothing about repeaters.
-
- I saw some postings awhile back about someone who had set up a repeater
- for packet, and was very pleased with it. I seem to have lost those
- messages! I'd like to ask if anyone has any advice on setting up a
- machine, which could be used for low and high speed packet repeating.
- I'm interested in nitty gritty details, like what kind of controller to
- use, or antenna placement or ...
-
- thanks,
- Henry N1EZP
-
- 8-Nov-87 12:48:28-EST,1204;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 8 Nov 87 12:48-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA05832@EDDIE.MIT.EDU>; Sun, 8 Nov 87 11:11:09 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA05826@EDDIE.MIT.EDU>; Sun, 8 Nov 87 11:10:56 EST
- Received: by june.cs.washington.edu (5.52.1/6.10)
- id AA20196; Sun, 8 Nov 87 08:11:33 PST
- Return-Path: <oliveb!pyramid!voder!randy@eddie.mit.edu>
- Message-Id: <8711081611.AA20196@june.cs.washington.edu>
- Date: 7 Nov 87 06:38:33 GMT
- From: oliveb!pyramid!voder!randy@EDDIE.MIT.edu (Randy Flatness)
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: pk232 software (also ts440).
- Keywords: pk232, ts440, software
-
- Recently I arquired an AEA pk232 all mode terminal unit for use
- with my ts440. First of all I think it is an outstanding piece
- of gear! Second, I would like to get even more out of it by using
- a control program for interfacing to both the radio and pk232.
-
- Does anyone have any such program, or is there anything in public
- domain close to this. (I have seen one ad for a commercial product -
- and am interested in any user response also).
-
-
- Thanks in advance Randy Flatness
-
-
-
- 10-Nov-87 18:26:17-EST,1288;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 10 Nov 87 18:26-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA00598@EDDIE.MIT.EDU>; Tue, 10 Nov 87 15:17:34 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA00587@EDDIE.MIT.EDU>; Tue, 10 Nov 87 15:17:14 EST
- Received: by june.cs.washington.edu (5.52.1/6.10)
- id AA12553; Tue, 10 Nov 87 12:18:19 PST
- Return-Path: <ucdavis!uop!mrapple@ucbvax.berkeley.edu>
- Message-Id: <8711102018.AA12553@june.cs.washington.edu>
- Date: 6 Nov 87 05:45:27 GMT
- From: ucdavis!uop!mrapple@ucbvax.berkeley.edu (Nick Sayer)
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: Packet freqs
-
- I know that on 2 meters the frequencies of choice are
- 5.01-5.09 every .02 Mhz (and to some extent 4.91-4.99),
- but can someone tell me what the 220 Mhz freqs. are?
- How about for 10 meters?
-
- The band plans mention 'digital and experimental FM,'
- but in pretty vague terms.
-
- -----
- Nick Sayer | Packet Radio: N6QQQ @ WA6RDH | CMS: SYSOP@STOKTON%STOCKTON
- uucp: ...!sdcsvax!ucbvax!ucdavis!uop!mrapple | Fido: 161/31
- Disclaimer: You didn't REALLY believe that, did you?
- Kirk: Analysis, Mr. Spock? Spock: It appears to be of external origin.
- Kirk: Mr. Sulu, go to pass two. Sulu: Aye aye, sir, going to pass two.
-
-
- 10-Nov-87 18:37:42-EST,1130;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 10 Nov 87 18:37-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA00545@EDDIE.MIT.EDU>; Tue, 10 Nov 87 15:16:06 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA00533@EDDIE.MIT.EDU>; Tue, 10 Nov 87 15:15:41 EST
- Received: by june.cs.washington.edu (5.52.1/6.10)
- id AA12523; Tue, 10 Nov 87 12:16:46 PST
- Return-Path: <bellcore!faline!karn@eddie.mit.edu>
- Message-Id: <8711102016.AA12523@june.cs.washington.edu>
- Date: 6 Nov 87 16:25:51 GMT
- From: bellcore!faline!karn@EDDIE.MIT.edu (Phil R. Karn)
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: Re: Packet freqs
- Summary: 220 mhz packet freqs
- Keywords: Frequencies
- References: <698@uop.UUCP>
-
- The band plan recommendation is to put conventional (i.e., AFSK/FM) packet
- on the frequencies 221.01, .03, ... with 20 khz spacing. High speed
- (100 Khz) channels are 220.55, .65, .75, .85 and .95.
-
- These recommendations are only that; you must still clear them with your
- local frequency coordination committee. And, of course, all this assumes
- we keep that portion of the band.
-
- Phil
-
-
- 10-Nov-87 18:39:33-EST,3885;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 10 Nov 87 18:39-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA00436@EDDIE.MIT.EDU>; Tue, 10 Nov 87 15:12:06 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA00427@EDDIE.MIT.EDU>; Tue, 10 Nov 87 15:11:49 EST
- Received: by june.cs.washington.edu (5.52.1/6.10)
- id AA12336; Tue, 10 Nov 87 12:11:40 PST
- Return-Path: <ihnp4!homxb!hotps!ka2qhd!kc2wz@eddie.mit.edu>
- Message-Id: <8711102011.AA12336@june.cs.washington.edu>
- Date: 5 Nov 87 20:42:01 GMT
- From: ihnp4!homxb!hotps!ka2qhd!kc2wz@EDDIE.MIT.edu (Bob)
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: Level 3 protocols - some info?
- Keywords: TCPIP,COSI,NETROM
-
- Ok people first I want to make it clear I am searching for information. I am
- NOT trying to fuel the so-called protocol war. So please keep my terminal
- >from bursting into FLAMES. I'm opened to reasoned arguments not "name
- calling". That said, on to my questions.
-
- Though I have been active on packet since 1984, I have for a number of reasons
- not completely followed the growing of the various Level 3 protocols. I find
- the information confusing and in some cases conflicting. I'm certain there
- are others in hamdom who are similarly confused. If I can get together enough
- real information, maybe I'll write an article for one ham magazines.
-
- Please straighten out any misconceptions or errors I have about the following:
-
- 1. TCP/IP - which KA9Q, et al are proposing is the best way to go for
- Level 3. Why? What are its advantages over the other protocols? What
- are is disadvantages? (Come now we know nothing is perfect.)
-
- 2. COSI - I'm not certain who developed the idea. Again, why should this
- be adopted as a Level 3 standard? What are its advantages? What are
- its disadvantages?
-
- 3. NET/ROM - Is this the same as COSI? It SEEMS to be the way things are
- going in the New Jersey/New York area on 2 meters. Am I wrong? Why
- should it be adopted? What are its advantages? Its disadvantages?
-
- 4. OTHER PROTOCOLS? Are there others be proposed? These are the only
- ones I have heard mentioned. As above what are the advantages and
- disadvantages?
-
- 5. Much of the work seems to centered are the IBM-PC and clones.
- Obviously, these are very popular micros. But there are a variety of
- other makes which don't seem to be supported. My machine (Tandy Color
- Computer 3) is one example. Other 6809/68000 machines also SEEM to be
- left in the cold. Is this true or just my impression?
-
- 6. My final(?) question (for now), how to go about setting up the TNC
- and computer to use Level 3? Is it necessary that everyone has Level 3
- running in their TNC or is this a waste? I have a TAPR TNC-1 running
- WA8DED's software. Will I have to burn a new EPROM with the new code?
- How will my computer be interfaced with the TNC?
-
- I think that is enough for now. You can be certain I will have more to come.
- As I said please no FLAMES. I realize that there are emotions running high in
- support of various protocols. I am completely open agruments from all sides.
- What I am not open to is calling protocol X a piece of trash. Maybe it is.
- But convince me with good arguments not FLAMES.
-
- I guess all these questions are what results when one fails to keep up with
- the fast changing packett scene. So I am playing catch up before I fall too
- far behind.
-
- Thanks in advance for the information. 73...
-
- --
- Bob Billson, KC2WZ Packet: KC2WZ at NN2Z
- SnailMail: 837 Summit Avenue UUCP: ucbvax!ihnp4!hotps!ka2qhd!kc2wz
- Westfield, NJ 07090 Phone: 201-232-2805
- For those with maps: Lat. 40.640972 N, Long. 74.337833 W, Elev. 15.24 M
-
-
- 10-Nov-87 19:01:47-EST,1326;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 10 Nov 87 19:01-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA00775@EDDIE.MIT.EDU>; Tue, 10 Nov 87 15:22:30 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA00767@EDDIE.MIT.EDU>; Tue, 10 Nov 87 15:22:06 EST
- Received: by june.cs.washington.edu (5.52.1/6.10)
- id AA12725; Tue, 10 Nov 87 12:22:44 PST
- From: ihnp4!drutx!mtgzz!mtuxo!mtune!whuts!homxb!hotps!ka2qhd!kc2wz@EDDIE.MIT.edu
- Return-Path: <ihnp4!drutx!mtgzz!mtuxo!mtune!whuts!homxb!hotps!ka2qhd!kc2wz@eddie.mit.edu>
- Message-Id: <8711102022.AA12725@june.cs.washington.edu>
- Date: 3 Nov 87 20:06:59 GMT
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: Any OS-9 Users on packet?
- Keywords: OS9
-
- Is there anyone in this group who has (or is interested in) experimented(ing)
- with OS-9 Levels 1 and/or 2, BASIC09, or C running on a Tandy Color Computer
- or any other OS-9 based system on packet radio?
-
- If so, I would like to hear from you. I am interested in seeing how
- OS-9 can be adapted to packet. Since it is a multi-user, multi-tasking
- operating system, it should have many varied and INEXPENSIVE uses on packet.
- Anyone game?
-
- 73...Bob Billson, KC2WZ
-
- nn2z-4 U Snail: 837 Summit Ave.
- Westfield, NJ 07090
-
-
- 10-Nov-87 20:04:28-EST,2834;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 10 Nov 87 20:04-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA00836@EDDIE.MIT.EDU>; Tue, 10 Nov 87 15:25:20 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA00823@EDDIE.MIT.EDU>; Tue, 10 Nov 87 15:24:43 EST
- Received: by june.cs.washington.edu (5.52.1/6.10)
- id AA12864; Tue, 10 Nov 87 12:25:40 PST
- Return-Path: <cornell!batcomputer!mitch@eddie.mit.edu>
- Message-Id: <8711102025.AA12864@june.cs.washington.edu>
- Date: 5 Nov 87 12:59:34 GMT
- From: cornell!batcomputer!mitch@EDDIE.MIT.edu (Mitch Collinsworth)
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: Stolen from club newsletter. Enjoy.
-
- MORE FROM THE OLD TIMER
-
- Contrary to what was reported about me last month, I was not sleeping...
- I was only resting my eyes!
-
- As you know well, packet is all the rage these days. There's a lot of
- "hacking" activity on two meters. So I thought I would try it. Being
- rather thrifty (some call it cheap), I searched high and low for the
- least expensive TNC (terminal node controller) available. Finally I
- found one for $39.95 from some outfit in the midwestern desert named
- "Trapper". No sales tax and prepaid shipping. Boy, I thought "what
- a great deal!"
-
- Alas, when the TNC arrived, I found that the unit contained a send-only
- modem, and the resident firmware was installed on a write-only EEPROM
- CHIP. So, I am saving up again for a TNC that really works -- one with
- 4 ports and one starboard.
-
- All was not lost, because the instruction manual sent with the TNC was
- very informative, although slightly obscure. If any of you could help
- me understand their list of default commands and parameter settings,
- some of which are shown below, please write the editor.
-
- FRACK AFTER FRICK'N
- XENON GASSED
- MAXOUT 12
- TAXDELAY FEDERAL/STATE/COUNTY/CITY
- CONOK CONOK WHO'S THERE
- STREAMSWITCH ON GET MOP
- ALAMODE ON
- HAIRSHIRT OFF
- BACON EVERY 10
- EGGS EVERY 5
- MYRESPONSE NEGATIVE
- KILL EVERY CONTROL-G
- MALL CLOSED
- SHOP EARLY
- BED CHECK
- XMITOK AFTER XMITIK
- SENDPAC COOKIES
- BUG OFF
- SUPLISP THERTAINLY
- SWEDEDETECT ON
- DANE DETECT OFF
- RAGTIME AFTER 1
- PERM $20
- MANICUR $5
- UNPROTON NEUTRON
- MYDILEMM ON
- CSTAMP OUCH
- TEATIME OFF
- CANPAC CHOO CHOO
- ABAWD DIRTY
- HBAWD DIRTY
- AXDELAY EXECUTION OFF
- BTEXT BZZZZ
- BUDLIST ANHEUSER/BUSCH
- MONOTOR SUNK
- ENUF ALREADY
-
- If anyone can, please help! I know I should have stayed with twenty
- meter CW!!
-
-
- From November 1987 RaRa Rag
- (RaRa = Rochester Amateur Radio Association)
-
- Credit in RaRa Rag was given to RAGS Review
- (RAGS = Radio Amateurs of Greater Syracuse)
-
- ------------------------------------------------------------------------
-
- -Mitch
- mitch@tcgould.tn.cornell.edu
-
- "If your antenna didn't fall down last winter, it wasn't big enough."
-
-
- 10-Nov-87 21:08:45-EST,1718;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 10 Nov 87 21:08-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA00768@EDDIE.MIT.EDU>; Tue, 10 Nov 87 15:22:11 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA00744@EDDIE.MIT.EDU>; Tue, 10 Nov 87 15:21:16 EST
- Received: by june.cs.washington.edu (5.52.1/6.10)
- id AA12668; Tue, 10 Nov 87 12:21:30 PST
- From: ihnp4!drutx!mtgzz!mtuxo!mtune!whuts!homxb!hotps!ka2qhd!kc2wz@EDDIE.MIT.edu
- Return-Path: <ihnp4!drutx!mtgzz!mtuxo!mtune!whuts!homxb!hotps!ka2qhd!kc2wz@eddie.mit.edu>
- Message-Id: <8711102021.AA12668@june.cs.washington.edu>
- Date: 3 Nov 87 20:31:45 GMT
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: TCP/IP available for 6809/68000?
- Keywords: OS9,TCPIP
-
- Can anyone tell me where I might obtain KA9Q's TCP/IP files which will run
- on the 6809/680000 microprocessors? I run OS-9 Level 2 on a Tandy Color
- Computer 3. I would like to be able to use TCP/IP on my TNC1 (which is runn-
- ing WA8DED's software.)
-
- Unfortunately, all the files are apparenty only available for MS-DOS machines.
- I do have a utility which can unARC the files but I still have a few problems.
-
- 1. No way to transfer MS-DOS files to an OS-9 disk. Can anyone help
- here?
-
- 2. Once I get them unARCed what do I do? I do have a C compiler
- (Microware's) so I can recompile the code. However, not being very
- familiar with MS-DOS, I don't know how to handle the machine
- specific code. Can anyone help here?
-
- Thanks very much all for the assistance.
-
- 73...Bob Billson, KC2WZ
-
- nn2z-4 U Snail: 837 Summit Ave.
- Westfield, NJ 07090
-
-
- 10-Nov-87 21:12:18-EST,1992;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 10 Nov 87 21:12-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA07376@EDDIE.MIT.EDU>; Tue, 10 Nov 87 19:30:57 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA07353@EDDIE.MIT.EDU>; Tue, 10 Nov 87 19:29:19 EST
- Received: by june.cs.washington.edu (5.52.1/6.10)
- id AA22371; Tue, 10 Nov 87 16:30:07 PST
- Return-Path: <somewhere!David_G._Michelson%UBC.MAILNET@um.cc.umich.edu>
- Message-Id: <8711110030.AA22371@june.cs.washington.edu>
- Date: Tue, 10 Nov 87 11:32:08 PST
- From: David_G._Michelson%UBC.MAILNET@um.cc.umich.edu
- To: packet-radio@EDDIE.MIT.EDU
- Subject: Repeater advice anyone?
-
- Re: 440 MHz Repeater for Packet Radio
-
- One of the most interesting things about packet radio (in my opinion)
- is the different ways one can approach the problem of repeating.
-
- The simplex digipeater is the simplest but, as subscribers to this
- mailing list are aware, it has severe disadvantages due to contention
- between the orginal packet, all its copies, and the many acknowledge-
- ments that fly back and forth up and down the link.
-
- NET/ROM is a significant improvement to the simple digipeater concept
- which, as I understand, uses standard TNC-2 hardware but enhanced
- firmware.
-
- Cross-band and cross-channel packet radio repeaters offer significant
- advantages because they eliminate the possibility of collision between
- originator-repeater traffic and repeater-addressee traffic.
-
- And then there are the many people who use conventional duplex repeaters
- for packet traffic (i.e., no regeneration).
-
- Are these the issues you were concerned with or are you after the
- technical details of a specific implementation?
-
-
- David G. Michelson, VE7TSX
- University of British Columbia
- Amateur Radio Club
-
- BitNet Address: USERMDG@UBCMTSG
-
-
- 10-Nov-87 21:27:36-EST,1091;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 10 Nov 87 21:27-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA00353@EDDIE.MIT.EDU>; Tue, 10 Nov 87 15:09:30 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA00349@EDDIE.MIT.EDU>; Tue, 10 Nov 87 15:09:17 EST
- Received: by june.cs.washington.edu (5.52.1/6.10)
- id AA12262; Tue, 10 Nov 87 12:10:16 PST
- Return-Path: <ihnp4!cbosgd!gwspc!cbcsta!n8emr!gws@eddie.mit.edu>
- Message-Id: <8711102010.AA12262@june.cs.washington.edu>
- Date: 8 Nov 87 04:02:04 GMT
- From: ihnp4!cbosgd!gwspc!cbcsta!n8emr!gws@EDDIE.MIT.edu (Gary W. Sanders )
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: ham domain
-
- Has anyone ever looking into getting a domain registered for the
- amateur radio packet network? I know that arpa has a set up
- address setup, but what about on other networks. Can an single
- person register a domain or does an orgainazation (ARRL) have
- to do this??
-
- --
- Gary W. Sanders {ihnp4|cbosgd}!n8emr!gws
- (cis) 72277,1325 (packet) N8EMR @ W8CQK
- HAM/SWL BBS (HBBS) 614-457-4227.. 300/1200 bps
-
-
- 11-Nov-87 00:27:42-EST,40098;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 11 Nov 87 00:27-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA00500@EDDIE.MIT.EDU>; Tue, 10 Nov 87 15:14:54 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA00473@EDDIE.MIT.EDU>; Tue, 10 Nov 87 15:13:15 EST
- Received: by june.cs.washington.edu (5.52.1/6.10)
- id AA12415; Tue, 10 Nov 87 12:14:14 PST
- From: sun!imagen!atari!portal!cup.portal.com!Patrick_J_Fagan@DECWRL.DEC.COM
- Return-Path: <sun!imagen!atari!portal!cup.portal.com!Patrick_J_Fagan@decwrl.dec.com>
- Message-Id: <8711102014.AA12415@june.cs.washington.edu>
- Date: 6 Nov 87 03:22:58 GMT
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: MFJ-40 A/B/C Upgrade Kit - Software Release 1.1.5 for TNC 2
- Xportal-User-Id: 1.1001.2475
-
- [Via WA5DVV]
-
- *====================================================================*
- | This information was taken from MFJ documentation. Permission to |
- | electronically reproduce and distribute this data has been granted |
- | to the Mississippi Amateur Radio Digital Association {MARDA}. |
- *====================================================================*
-
-
- MFJ-40A/40B/40C
- UPDATE KIT FOR TNC 2 VERSION 2, RELEASE 1.1.5
-
- This kit contains one 27C256 EPROM with Release 1.1.5 software
- and one instruction sheet. MFJ-40A contains software
- release 1.1.5-16K for 16K RAM operation. MFJ-40B contains
- software release 1.1.5-32K EPROM and a 32K RAM. MFJ-40C contains
- software release 1.1.5-32K EPROM without 32K RAM.
-
- INSTALLATION OF MFJ-40A FOR 16K RAM OPERATION
- 1. Remove all power from the TNC 2.
- 2. Remove the cover by removing the mounting screws on the side
- of the cabinets.
- 3. Remove JMP 5 jumper. This disconnects the Lithium battery.
- 4. Carefully remove the EPROM (U23, 27256) from the TNC
- 2 board. Make note of the orientation of the IC.
- 5. Install the new EPROM containing 1.1.5-16K software on
- the TNC 2 board at U23. Make sure that no IC pins are
- bent under the IC. Make sure that the notch is pointed in
- the same direction as the old EPROM.
- 6. Install JMP5 jumper.
- 7. Replace the cover of the cabinet.
- 8. Apply power to the TNC 2.
-
- INSTALLATION OF MFJ-40B/40C FOR 32K RAM OPERATION
-
- 1. Remove all power from the TNC 2.
- 2. Remove the cover by removing the mounting screws on the
- side of the cabinets.
- 3. Remove JMP5 jumper. This disconnects the Lithium
- battery.
- 4. Remove 8K RAMs at U24 and U25.
- 5. FOR MFJ-1270: Remove circuit board from chassic. Cut the
- trace that is connected between the center pad and the pad
- closest to R29 of JMP 12. Connect a jumper wire between the
- center pad and the pad closest to C28 of JMP 12. Re-install the
- circuit board.
- FOR MFJ-1270B AND MFJ-1274: A removable jumper for JMP12 is
- provided. Remove the jumper from pin 2 and 3 of JMP12 and
- install it on pin 1 and 2 of JMP12.
- 6. Install 32K RAM chip at U25 (U24 is left unused). Some
- 32K RAMs which may be used are NEC 43256C-15L, Hitachi
- HM64456L-15 or HM62256L-15 and Toshiba 55257PL-15.
- NOTE: If you purchased the MFJ-40C update, you must supply
- your own 32K RAM. This EPROM will not operate with 16K RAM.
- 7. Carefully remove the EPROM (U23, 27256) from the TNC 2
- board. Make note of the orientation of the IC.
- 8. Install the new EPROM with 1.1.5-32K software on the TNC
- 2 board at U23. Make sure that no IC pins are bent
- under the IC. Make sure that the notch on the IC is
- pointed in the same direction as the old EPROM.
- 9. Install JMP5 jumper.
- 10. Replace the cover of the cabinet.
- 11. Apply power to the TNC2.
-
-
- The 1.1.5-16K and 1.1.5-32K software releases contain all
- FIXES, CHANGES AND ADDITIONS of the 1.1.4 and 1.1.3 releases.
-
-
- SOFTWARE RELEASE NOTES FOR RELEASE 1.1.3
-
- FIXES
-
- o- When AX25L2V2 is ON, the TNC now answers L2 UI
- frames with P and C set with either: RR if connected
- (regardless of rcvr flow control state), or DM if not
- connected.
-
- o- Path for SABM received while in link-setup state is
- not checked. This corrects earlier operation where a
- self-connect via an odd number of different digipeaters
- fails.
-
- o- T2 <RESPTIME> utilization corrected for the 2nd
- through nth link
-
- o- FULLDUP operation restored
-
- o- Channel capture during busy times improved
-
- o- NOMODE bug, where the 'connect mode' setting was not
- used on received SABMS has been corrected. This bug
- may have caused the TNC to enter a state where it
- seemed like the TNC died, when it actually was in
- TRANSparent mode.
-
- CHANGES
-
- o- BEACONS not sent if BTEXT is null
-
- o- RESPTIME default now at 5; i.e. 500ms
-
-
- ADDITIONS
-
- o- RXBLOCK command
-
- o- TNC "Health" feature group installed
-
- Fourteen counters have been added to TNC 2. All of them are
- 16 bits wide, and are ALWAYS initialized to 0000 on power up
- or "RESTART".
-
- o- ASYRXOVR: Increases when the software does not
- service the asynchronous receiver in
- time. Indicates data from the user to
- the TNC is being dropped. This error
- counter should never become non-zero
- under supported data rates.
-
- o- DIGISENT: Each frame digipeated by this TNC causes
- the counter to increase.
-
- o- HOVRERR: Increases when HDLC receiver is not
- serviced rapidly enough and data is
- lost. This counter should never incre-
- ment at any supported data rate.
-
- o- HUNDRERR: Increases when the HDLC transmitter is
- not serviced rapidly enough and frames
- are aborted. This counter should never
- be non-zero at any supported data rate.
-
- o- RCVDFRMR: Increases when Frame reject frames are
- received from a connected station.
-
- o- RCVDIFRA: Increases for each reception of an I
- frame from a connectee.
-
- o- RCVDREJ: Increases for each reception of an REJ-
- ect frame from a connectee.
-
- o- RCVDSABM: Each received SABM frame addressed to
- the TNC causes this counter to be inc-
- reased by one.
-
- o- RXCOUNT: Increases when any frame is received
- with good CRC (or any CRC if HGARBAGE is
- turned on).
-
- o- RXERRORS: Increments each time a received frame is
- thrown out due to it being too short,
- suffering overrun(s), or it having a
- bad CRC. Latter occurs only when CRC
- checking is enabled (i.e. HGARBAGE is
- OFF). This counter will often increment
- in the presence of noise.
-
- o- SENTFRMR: Increments each time a Frame reject
- frame is transmitted.
-
- o- SENTIFRA: Increases by one each time an I frame is
- sent.
-
- o- SENTREJ: Whenever a REJect frame is transmitted,
- this counter is incremented.
-
- o- TXCOUNT: Incremented whenever a frame is correct-
- ly transmitted.
-
-
-
-
-
- The following information should be inserted into the DISPLAY command.
-
- DISPLAY HEALTH
-
- The counters just described, and the setting of HEALLED are
- displayed in response to the health inquiry.
-
-
- New Commands
-
- AX25L2V2 ON|OFF Default: OFF
-
-
- Parameters:
-
- ON The TNC will use AX.25 Level 2 Version 2.0
- protocol.
-
- OFF The TNC will use AX.25 Level 2 Version 1.0
- protocol.
-
- Some implementations of the earlier version of AX.25 proto-
- col (e.g., TAPR's TNC 1) won't properly digipeat version 2.0
- AX.25 packets. This command exists to provide compatibility
- with these other TNCs until their software has been updated.
-
- During the protocol transition period, you should set
- AX25L2V2 OFF.
-
- After your local area TNCs are updated to the newer protocol
- version, you should set AX25L2V2 ON.
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
- RXBLOCK ON|OFF Default: OFF
-
-
- Parameters:
-
- ON The TNC will send data to the terminal in
- RXBLOCK format.
-
- OFF The TNC will send data to the terminal in
- standard format.
-
-
- RXBLOCK is designed for automated operations, such as packet
- bulletin board stations. It is intended to help such
- systems discriminate between data received from the
- connected station and TNC-generated messages.
-
- Correct operation of RXBLOCK is dependant on the AWLEN
- parameter getting set to 8 (bits) since the character FF hex
- marks the beginning of a recieved data unit header.
-
- When RXBLOCK is on, data from other stations will be sent
- >from the TNC in the following format:
-
- ------------------------------------------------------
- | $FF | L0 | L1 | PID | DATA |
- ------------------------------------------------------
-
- ( prefix )( length ) ( pid ) ( data )
-
- The fields above are defined as follows:
-
- prefix $FF ::= A character with all 8 bits set
- length L0 ::= The high order length of the data,
- length, and pid fields logically ORed
- with the value $F0
- L1 ::= The low order length of the data,
- length, and pid fields
- pid PID ::= The Protocol IDentifier byte received
- for the following data field
- data DATA ::= [Optional], variable length data
-
-
- For best operation it is suggested that parameters like
- AUTOLF, MFILTER etc. be set OFF in order to prevent uncer-
- tanties in the size of the data field.
-
-
-
-
-
-
-
-
-
- HEALLED ON|OFF Default: OFF
-
-
- Parameters:
-
- ON The TNC will "dither" the CON and STA LEDs.
-
- OFF The TNC will control the CON and STA LEDs in
- normal fashion.
-
-
- This command allows the user to redefine the functions of
- the two CPU controllable LEDs (i.e. the STAtus and CONnect
- LEDs).
-
- When HEALLED is set ON, the two LEDs flash in a seeming
- random fashion. At a glance, the user may make a judgement
- on whether the software has crashed, since the LEDs will
- probably not flash if the software fails catastrophically.
-
- With HEALLED set OFF, the LEDs function as before.
-
-
- LCSTREAM ON|OFF Default: ON
-
-
- Parameters:
-
- ON The TNC will translate the chareacter immedi-
- ately following the STREAMSWITCH character to
- upper case before processing it.
-
- OFF The TNC will process the character immediate-
- ly following the STREAMSWITCH character as it
- is entered.
-
-
- When operating multi-connect, the user must enter a stream
- identifier (default A through J) after the STREAMSWITCH
- character (default |) to select a new logical stream to send
- data. Normally, the stream identifier must be in upper
- case, or an error message will result.
-
- When LCSTREAM is ON, the character immediately following the
- streamswitch character is converted to upper case before
- being acted upon. Thus, the case (upper or lower) becomes
- insignificant. Use of LCSTREAM is useful if you are typing
- in lower case and don't want to be bothered with remembering
- to switch to upper case when changing streams.
-
-
-
-
-
-
- NOMODE ON|OFF Default: OFF
-
-
- Parameters:
-
- ON The TNC will only switch modes (command,
- converse or transparent) upon explicit com-
- mand.
-
- OFF The TNC will switch modes in accordance with
- the setting of NEWMODE.
-
-
- When NOMODE is ON, the TNC will never change between
- CONVERSE or TRANSPARENT mode to COMMAND mode (or vice-versa)
- on its own. Only user commands (CONV, TRANS, or ^C) may
- change the typein mode.
-
- If NOMODE is OFF, then automatic mode switching is handled
- according to the setting of the NEWMODE command.
-
-
- SOFTWARE RELEASE NOTES FOR RELEASE 1.1.4
-
-
- FIXES
-
- o - Transmitted I-frames under Level 2 Version 2 did not
- have their P bits set at the appropriate times. In
- fact, they never had their P bits set. This has now
- been fixed. The last I-frame of a multiple I-frame
- transmission has its P bit set.
-
- o - A mistake in the protocol state table was fixed.
-
- o - bbRAM scanning now checks all ten possible connection
- control structures (instead of just the first one).
-
-
- CHANGES
-
- o - AX25L2V2 defaults to the ON position.
-
- o - Major change made to AX25L2V2 handling. If retry limit
- is exceeded, or the TNC receives a "disconnected" respo-
- nse to a poll, the connection is ended
-
- The old method (and the one proscribed) is fraught with
- problems for automated stations that can not recover
- without an indication of loss of the connecion.
-
- The PERMCON control will replace the functionality of this
- aspect of AX25L2V2 which was removed.
-
-
- ENHANCEMENTS
-
- o - 32K of RAM is now expected. Virtually all of the new
- space is used to enlarge existing queues within the TNC,
- yielding greater performance especially at faster RF
- data rates, and making the on-board message buffer capa-
- bility a bit more useful.
-
- o - The MCOM command decodes all control fields (invalid
- ones are marked with ????).
-
- For I and S frames, sequence number informnation is also
- presented.
-
- Frames compatible with the AX.25 Level 2.0 standard are also
- decoded as to the state of the Command/Response (C/R) and
- Poll/Final (P/F) bits.
- Ex: WA7GXD>KV7B <I C SO RO>:
- Hi Dan,
-
- WA7GXD>KV7B <I C P S1 RO>:
- have you been on EIES lately?
-
- KV7B>WA7GXD <RR R F R2>
- KV7B>WA7GXD <I C P S1 R2>:
- I was just thinking about that. I heard that
- @(username) made some real unbelievable comment on
- it!
-
- WA7GXD>KV7B <RR R F R2>
- WB2SPE>KV7B <C>
- KV7B>WB2SPE <DM>
- KV7B>WA7GXD <I C P S2 R2>:
- Good conditions now...
-
- WA7GXD>KV7B <RR R F R3>
- WA7GXD>KV7B <I C P S2 R3>:
- Yes @(username) did. It was quite remarkable.
-
- And so on. See Chapter 9 Table 9-1 in your TNC 2 manual for
- a breakdown of the control field codes. For complete infor-
- mation on the AX.25 Level 2 Version 2.0 Protocol, please
- refer to the ARRL AX.25 Protocal Specification document,
- available from ARRL for $10.00.
-
-
-
-
-
-
-
-
-
-
-
- NEW COMMANDS
-
- CBELL ON:OFF Default: OFF
-
- Parameters:
-
- ON Connect bell enabled
-
- OFF Connect bell disabled
-
- This command is used to control whether an ASCII $O7 (BELL)
- character is sent as part of the connected message.
-
- When set ON, the bell character immediately procedes the
- asterisk portion of the connected message, e.g.:
-
- <BELL>*** Connected to: <callsign>
-
-
-
- CM5GDISC ON:OFF Default: OFF
-
- Parameters:
-
- ON Automatic disconnect enabled
-
- OFF Automatic disconnect disabled
-
- This command controls whether the TNC 2 will iniitiate a
- disconnect sequence after it is connected to.
-
- If CMSG is OFF, or CTEXT has no connected text, the TNC
- initiates a disconnect immediately upon receiving informa-
- tion or acknowledgement frames from the other station.
-
- If CMSG is ON end CTEXT contains some text information, the
- TNC initiates a disconnect after the packet containing con-
- nect text (CTEXT) is acknowledged.
-
- This command may be useful to bulletin board operators or
- others with a need to send a short message, confirm its
- receipt, and disconnect.
-
- NOTE: Use this command with care - If you find you're
- able to receive connects, yet never get data, it's
- possible CMSGDisc has been left on. It's also
- possible is that RS-232 DCD is holding the
- terminal off -- see the TNC 2 System Manual for
- details on hardware flow control.
-
-
-
-
-
-
- LFIGNORE ON:OFF Default: OFF
-
- Parameters:
-
- ON TNC will ignore <LF> characters.
-
- OFF TNC will respond to <LF> characters.
-
- This command controls whether TNC 2 responds to ASCII Lind
- Feed (<LF> $OA) characters or ignores them in command and
- conerse modes.
-
- When turned on, line feeds are totally ignored except in
- transparent mode.
-
- New HEALTH Counters
-
- BBfailed n : Counts number of times bbRAM checksum was in
- error.
-
- TXQovflw n : Counts how many times frames were discarded
- because the outgoing frame queue was too
- small.
-
- SOFTWARE RELEASE NOTES FOR RELEASE 1.1.5
-
- This document describes the differents between the latest
- release 1.1.5 and the previous (1.1.4) release of TAPR
- software for the TNC 2.
-
- The new object image occupies almost 0X45FF bytes of EPROM.
-
- NEW COMMAND
-
- BBSMSGS ON:OFF Default: OFF
-
- This command controls how the TNC displays certain messages
- in command and CONVERSE modes. The messages affected are
- described below:
-
- MESSAGE EFFECT WHEN BBSMSGS ON
-
- ***CONNECTED to xxxx -A newline is added just before"***"
-
- ***DISCONNECTED - " "
-
- ***retry limit exceeded- " "
-
- ***xxxx Busy - " "
-
- ***FRMR sent - " "
-
- ***FRMR rcvd - " "
-
- ***Connect request:xxxx-This message is omitted.
-
- The BBSMSGS command is primarily useful for host operation.
- Primarily with WORLI and like bulletin board systems that
- require link status messages to begin in the first output
- column.
-
- The connect request message is omitted during BBSMSGS mode.
- This should be most useful for preventing corruption of
- messages when forwarding with small frames.
-
-
- TXTMO : Counter Default: 0
-
- TXTMO is a new addition to the TNC health-group. This
- register may accumulate counts as the TNC succesfully reco-
- vers from HDLC transmitter timeouts. This is not a useful
- command for the majority of the users.
-
-
-
-
- FIXES
-
- o - Release 1.1.4 suffered from a spurious condition where
- the HDLC transmitter would time out. When this
- happened. TXQOVFLW would typically show a non-zero
- count. Release 1.1.5 incorporates an HDLC transmitter
- timeout feature to capture and recover from the
- timeout error.
-
- o - DWAIT operation has been fixed.
-
-
- ****************************************************************************
- | Above fix corrects the DIGI HANG problem when used in DIGI mode. When |
- | used as an unattended DIGI - BUG in 1.1.4 caused TNC not to respond. |
- | The only way to recover was to reset TNC. This has been fixed in 1.1.5. |
- ****************************************************************************
-
- [EOF]
-
-
-
-
-
-
-
-
-
-
-
-
-
-
- MFJ-40A/40B/40C
- UPDATE KIT FOR TNC 2 VERSION 2, RELEASE 1.1.5
-
- This kit contains one 27C256 EPROM with Release 1.1.5 software
- and one instruction sheets. MFJ-40A contains software
- release 1.1.5-16K for 16K RAM operation. MFJ-40B contains
- software release 1.1.5-32K EPROM and a 32K RAM. MFJ-40C contains
- software release 1.1.5-32K EPROM without 32K RAM.
-
- INSTALLATION OF MFJ-40A FOR 16K RAM OPERATION
- 1. Remove all power from the TNC 2.
- 2. Remove the cover by removing the mounting screws on the side
- of the cabinets.
- 3. Remove JMP 5 jumper. This disconnects the Lithium battery.
- 4. Carefully remove the EPROM (U23, 27256) from the TNC
- 2 board. Make note of the orientation of the IC.
- 5. Install the new EPROM containing 1.1.5-16K software on
- the TNC 2 board at U23. Make sure that no IC pins are
- bent under the IC. Make sure that the notch is pointed in
- the same direction as the old EPROM.
- 6. Install JMP5 jumper.
- 7. Replace the cover of the cabinet.
- 8. Apply power to the TNC 2.
-
- INSTALLATION OF MFJ-40B/40C FOR 32K RAM OPERATION
-
- 1. Remove all power from the TNC 2.
- 2. Remove the cover by removing the mounting screws on the
- side of the cabinets.
- 3. Remove JMP5 jumper. This disconnects the Lithium
- battery.
- 4. Remove 8K RAMs at U24 and U25.
- 5. FOR MFJ-1270: Remove circuit board from chassic. Cut the
- trace that is connected between the center pad and the pad
- closest to R29 of JMP 12. Connect a jumper wire between the
- center pad and the pad closest to C28 of JMP 12. Re-install the
- circuit board.
- FOR MFJ-1270B AND MFJ-1274: A removable jumper for JMP12 is
- provided. Remove the jumper from pin 2 and 3 of JMP12 and
- install it on pin 1 and 2 of JMP12.
- 6. Install 32K RAM chip at U25 (U24 is left unused). Some
- 32K RAMs which may be used are NEC 43256C-15L, Hitachi
- HM64456L-15 or HM62256L-15 and Toshiba 55257PL-15.
- NOTE: If you purchased the MFJ-40C update, you must supply
- your own 32K RAM. This EPROM will not operate with 16K RAM.
- 7. Carefully remove the EPROM (U23, 27256) from the TNC 2
- board. Make note of the orientation of the IC.
- 8. Install the new EPROM with 1.1.5-32K software on the TNC
- 2 board at U23. Make sure that no IC pins are bent
- under the IC. Make sure that the notch on the IC is
- pointed in the same direction as the old EPROM.
- 9. Install JMP5 jumper.
- 10. Replace the cover of the cabinet.
- 11. Apply power to the TNC2.
-
-
- The 1.1.5-16K and 1.1.5-32K software releases contain all
- FIXES, CHANGES AND ADDITIONS of the 1.1.4 and 1.1.3 releases.
-
-
- SOFTWARE RELEASE NOTES FOR RELEASE 1.1.3
-
- FIXES
-
- o- When AX25L2V2 is ON, the TNC now answers L2 UI
- frames with P and C set with either: RR if connected
- (regardless of rcvr flow control state), or DM if not
- connected.
-
- o- Path for SABM received while in link-setup state is
- not checked. This corrects earlier operation where a
- self-connect via an odd number of different digipeaters
- fails.
-
- o- T2 <RESPTIME> utilization corrected for the 2nd
- through nth link
-
- o- FULLDUP operation restored
-
- o- Channel capture during busy times improved
-
- o- NOMODE bug, where the 'connect mode' setting was not
- used on received SABMS has been corrected. This bug
- may have caused the TNC to enter a state where it
- seemed like the TNC died, when it actually was in
- TRANSparent mode.
-
- CHANGES
-
- o- BEACONS not sent if BTEXT is null
-
- o- RESPTIME default now at 5; i.e. 500ms
-
-
- ADDITIONS
-
- o- RXBLOCK command
-
- o- TNC "Health" feature group installed
-
- Fourteen counters have been added to TNC 2. All of them are
- 16 bits wide, and are ALWAYS initialized to 0000 on power up
- or "RESTART".
-
- o- ASYRXOVR: Increases when the software does not
- service the asynchronous receiver in
- time. Indicates data from the user to
- the TNC is being dropped. This error
- counter should never become non-zero
- under supported data rates.
-
- o- DIGISENT: Each frame digipeated by this TNC causes
- the counter to increase.
-
- o- HOVRERR: Increases when HDLC receiver is not
- serviced rapidly enough and data is
- lost. This counter should never incre-
- ment at any supported data rate.
-
- o- HUNDRERR: Increases when the HDLC transmitter is
- not serviced rapidly enough and frames
- are aborted. This counter should never
- be non-zero at any supported data rate.
-
- o- RCVDFRMR: Increases when Frame reject frames are
- received from a connected station.
-
- o- RCVDIFRA: Increases for each reception of an I
- frame from a connectee.
-
- o- RCVDREJ: Increases for each reception of an REJ-
- ect frame from a connectee.
-
- o- RCVDSABM: Each received SABM frame addressed to
- the TNC causes this counter to be inc-
- reased by one.
-
- o- RXCOUNT: Increases when any frame is received
- with good CRC (or any CRC if HGARBAGE is
- turned on).
-
- o- RXERRORS: Increments each time a received frame is
- thrown out due to it being too short,
- suffering overrun(s), or it having a
- bad CRC. Latter occurs only when CRC
- checking is enabled (i.e. HGARBAGE is
- OFF). This counter will often increment
- in the presence of noise.
-
- o- SENTFRMR: Increments each time a Frame reject
- frame is transmitted.
-
- o- SENTIFRA: Increases by one each time an I frame is
- sent.
-
- o- SENTREJ: Whenever a REJect frame is transmitted,
- this counter is incremented.
-
- o- TXCOUNT: Incremented whenever a frame is correct-
- ly transmitted.
-
-
-
-
-
- The following information should be inserted into the DISPLAY command.
-
- DISPLAY HEALTH
-
- The counters just described, and the setting of HEALLED are
- displayed in response to the health inquiry.
-
-
- New Commands
-
- AX25L2V2 ON|OFF Default: OFF
-
-
- Parameters:
-
- ON The TNC will use AX.25 Level 2 Version 2.0
- protocol.
-
- OFF The TNC will use AX.25 Level 2 Version 1.0
- protocol.
-
- Some implementations of the earlier version of AX.25 proto-
- col (e.g., TAPR's TNC 1) won't properly digipeat version 2.0
- AX.25 packets. This command exists to provide compatibility
- with these other TNCs until their software has been updated.
-
- During the protocol transition period, you should set
- AX25L2V2 OFF.
-
- After your local area TNCs are updated to the newer protocol
- version, you should set AX25L2V2 ON.
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
- RXBLOCK ON|OFF Default: OFF
-
-
- Parameters:
-
- ON The TNC will send data to the terminal in
- RXBLOCK format.
-
- OFF The TNC will send data to the terminal in
- standard format.
-
-
- RXBLOCK is designed for automated operations, such as packet
- bulletin board stations. It is intended to help such
- systems discriminate between data received from the
- connected station and TNC-generated messages.
-
- Correct operation of RXBLOCK is dependant on the AWLEN
- parameter getting set to 8 (bits) since the character FF hex
- marks the beginning of a recieved data unit header.
-
- When RXBLOCK is on, data from other stations will be sent
- >from the TNC in the following format:
-
- ------------------------------------------------------
- | $FF | L0 | L1 | PID | DATA |
- ------------------------------------------------------
-
- ( prefix )( length ) ( pid ) ( data )
-
- The fields above are defined as follows:
-
- prefix $FF ::= A character with all 8 bits set
- length L0 ::= The high order length of the data,
- length, and pid fields logically ORed
- with the value $F0
- L1 ::= The low order length of the data,
- length, and pid fields
- pid PID ::= The Protocol IDentifier byte received
- for the following data field
- data DATA ::= [Optional], variable length data
-
-
- For best operation it is suggested that parameters like
- AUTOLF, MFILTER etc. be set OFF in order to prevent uncer-
- tanties in the size of the data field.
-
-
-
-
-
-
-
-
-
- HEALLED ON|OFF Default: OFF
-
-
- Parameters:
-
- ON The TNC will "dither" the CON and STA LEDs.
-
- OFF The TNC will control the CON and STA LEDs in
- normal fashion.
-
-
- This command allows the user to redefine the functions of
- the two CPU controllable LEDs (i.e. the STAtus and CONnect
- LEDs).
-
- When HEALLED is set ON, the two LEDs flash in a seeming
- random fashion. At a glance, the user may make a judgement
- on whether the software has crashed, since the LEDs will
- probably not flash if the software fails catastrophically.
-
- With HEALLED set OFF, the LEDs function as before.
-
-
- LCSTREAM ON|OFF Default: ON
-
-
- Parameters:
-
- ON The TNC will translate the chareacter immedi-
- ately following the STREAMSWITCH character to
- upper case before processing it.
-
- OFF The TNC will process the character immediate-
- ly following the STREAMSWITCH character as it
- is entered.
-
-
- When operating multi-connect, the user must enter a stream
- identifier (default A through J) after the STREAMSWITCH
- character (default |) to select a new logical stream to send
- data. Normally, the stream identifier must be in upper
- case, or an error message will result.
-
- When LCSTREAM is ON, the character immediately following the
- streamswitch character is converted to upper case before
- being acted upon. Thus, the case (upper or lower) becomes
- insignificant. Use of LCSTREAM is useful if you are typing
- in lower case and don't want to be bothered with remembering
- to switch to upper case when changing streams.
-
-
-
-
-
-
- NOMODE ON|OFF Default: OFF
-
-
- Parameters:
-
- ON The TNC will only switch modes (command,
- converse or transparent) upon explicit com-
- mand.
-
- OFF The TNC will switch modes in accordance with
- the setting of NEWMODE.
-
-
- When NOMODE is ON, the TNC will never change between
- CONVERSE or TRANSPARENT mode to COMMAND mode (or vice-versa)
- on its own. Only user commands (CONV, TRANS, or ^C) may
- change the typein mode.
-
- If NOMODE is OFF, then automatic mode switching is handled
- according to the setting of the NEWMODE command.
-
-
- SOFTWARE RELEASE NOTES FOR RELEASE 1.1.4
-
-
- FIXES
-
- o - Transmitted I-frames under Level 2 Version 2 did not
- have their P bits set at the appropriate times. In
- fact, they never had their P bits set. This has now
- been fixed. The last I-frame of a multiple I-frame
- transmission has its P bit set.
-
- o - A mistake in the protocol state table was fixed.
-
- o - bbRAM scanning now checks all ten possible connection
- control structures (instead of just the first one).
-
-
- CHANGES
-
- o - AX25L2V2 defaults to the ON position.
-
- o - Major change made to AX25L2V2 handling. If retry limit
- is exceeded, or the TNC receives a "disconnected" respo-
- nse to a poll, the connection is ended
-
- The old method (and the one proscribed) is fraught with
- problems for automated stations that can not recover
- without an indication of loss of the connecion.
-
- The PERMCON control will replace the functionality of this
- aspect of AX25L2V2 which was removed.
-
-
- ENHANCEMENTS
-
- o - 32K of RAM is now expected. Virtually all of the new
- space is used to enlarge existing queues within the TNC,
- yielding greater performance especially at faster RF
- data rates, and making the on-board message buffer capa-
- bility a bit more useful.
-
- o - The MCOM command decodes all control fields (invalid
- ones are marked with ????).
-
- For I and S frames, sequence number informnation is also
- presented.
-
- Frames compatible with the AX.25 Level 2.0 standard are also
- decoded as to the state of the Command/Response (C/R) and
- Poll/Final (P/F) bits.
- Ex: WA7GXD>KV7B <I C SO RO>:
- Hi Dan,
-
- WA7GXD>KV7B <I C P S1 RO>:
- have you been on EIES lately?
-
- KV7B>WA7GXD <RR R F R2>
- KV7B>WA7GXD <I C P S1 R2>:
- I was just thinking about that. I heard that
- @(username) made some real unbelievable comment on
- it!
-
- WA7GXD>KV7B <RR R F R2>
- WB2SPE>KV7B <C>
- KV7B>WB2SPE <DM>
- KV7B>WA7GXD <I C P S2 R2>:
- Good conditions now...
-
- WA7GXD>KV7B <RR R F R3>
- WA7GXD>KV7B <I C P S2 R3>:
- Yes @(username) did. It was quite remarkable.
-
- And so on. See Chapter 9 Table 9-1 in your TNC 2 manual for
- a breakdown of the control field codes. For complete infor-
- mation on the AX.25 Level 2 Version 2.0 Protocol, please
- refer to the ARRL AX.25 Protocal Specification document,
- available from ARRL for $10.00.
-
-
-
-
-
-
-
-
-
-
-
- NEW COMMANDS
-
- CBELL ON:OFF Default: OFF
-
- Parameters:
-
- ON Connect bell enabled
-
- OFF Connect bell disabled
-
- This command is used to control whether an ASCII $O7 (BELL)
- character is sent as part of the connected message.
-
- When set ON, the bell character immediately procedes the
- asterisk portion of the connected message, e.g.:
-
- <BELL>*** Connected to: <callsign>
-
-
-
- CM5GDISC ON:OFF Default: OFF
-
- Parameters:
-
- ON Automatic disconnect enabled
-
- OFF Automatic disconnect disabled
-
- This command controls whether the TNC 2 will iniitiate a
- disconnect sequence after it is connected to.
-
- If CMSG is OFF, or CTEXT has no connected text, the TNC
- initiates a disconnect immediately upon receiving informa-
- tion or acknowledgement frames from the other station.
-
- If CMSG is ON end CTEXT contains some text information, the
- TNC initiates a disconnect after the packet containing con-
- nect text (CTEXT) is acknowledged.
-
- This command may be useful to bulletin board operators or
- others with a need to send a short message, confirm its
- receipt, and disconnect.
-
- NOTE: Use this command with care - If you find you're
- able to receive connects, yet never get data, it's
- possible CMSGDisc has been left on. It's also
- possible is that RS-232 DCD is holding the
- terminal off -- see the TNC 2 System Manual for
- details on hardware flow control.
-
-
-
-
-
-
- LFIGNORE ON:OFF Default: OFF
-
- Parameters:
-
- ON TNC will ignore <LF> characters.
-
- OFF TNC will respond to <LF> characters.
-
- This command controls whether TNC 2 responds to ASCII Lind
- Feed (<LF> $OA) characters or ignores them in command and
- conerse modes.
-
- When turned on, line feeds are totally ignored except in
- transparent mode.
-
- New HEALTH Counters
-
- BBfailed n : Counts number of times bbRAM checksum was in
- error.
-
- TXQovflw n : Counts how many times frames were discarded
- because the outgoing frame queue was too
- small.
-
- SOFTWARE RELEASE NOTES FOR RELEASE 1.1.5
-
- This document describes the differents between the latest
- release 1.1.5 and the previous (1.1.4) release of TAPR
- software for the TNC 2.
-
- The new object image occupies almost 0X45FF bytes of EPROM.
-
- NEW COMMAND
-
- BBSMSGS ON:OFF Default: OFF
-
- This command controls how the TNC displays certain messages
- in command and CONVERSE modes. The messages affected are
- described below:
-
- MESSAGE EFFECT WHEN BBSMSGS ON
-
- ***CONNECTED to xxxx -A newline is added just before"***"
-
- ***DISCONNECTED - " "
-
- ***retry limit exceeded- " "
-
- ***xxxx Busy - " "
-
- ***FRMR sent - " "
-
- ***FRMR rcvd - " "
-
- ***Connect request:xxxx-This message is omitted.
-
- The BBSMSGS command is primarily useful for host operation.
- Primarily with WORLI and like bulletin board systems that
- require link status messages to begin in the first output
- column.
-
- The connect request message is omitted during BBSMSGS mode.
- This should be most useful for preventing corruption of
- messages when forwarding with small frames.
-
-
- TXTMO : Counter Default: 0
-
- TXTMO is a new addition to the TNC health-group. This
- register may accumulate counts as the TNC succesfully reco-
- vers from HDLC transmitter timeouts. This is not a useful
- command for the majority of the users.
-
-
-
-
- FIXES
-
- o - Release 1.1.4 suffered from a spurious condition where
- the HDLC transmitter would time out. When this
- happened. TXQOVFLW would typically show a non-zero
- count. Release 1.1.5 incorporates an HDLC transmitter
- timeout feature to capture and recover from the
- timeout error.
-
- o - DWAIT operation has been fixed.
-
-
- *==========================================================================*
- | Above fix corrects the DIGI HANG problem experienced in DIGI mode. When |
- | used as an unattended DIGI - A BUG in 1.1.4 caused TNC not to respond. |
- | The only way to recover was to reset TNC. This has been fixed in 1.1.5. |
- | Version 1.1.5 will work in any 100% compatible TNC 2 unit. |
- *==========================================================================*
-
- [EOF]
-
-
- 11-Nov-87 01:18:32-EST,1080;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 11 Nov 87 01:18-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA09486@EDDIE.MIT.EDU>; Tue, 10 Nov 87 21:12:10 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA09472@EDDIE.MIT.EDU>; Tue, 10 Nov 87 21:11:42 EST
- Date: Tue, 10 Nov 1987 19:10 MST
- Message-Id: <KPETERSEN.12349627581.BABYL@SIMTEL20.ARPA>
- Sender: KPETERSEN@SIMTEL20.ARPA
- From: Keith Petersen <W8SDZ@SIMTEL20.ARPA>
- To: Info-Cpm@SIMTEL20.ARPA
- Cc: Info-Micro@BRL.ARPA, Info-HZ100@RADC-TOPS20.ARPA, Info-IBMPC@C.ISI.EDU,
- packet-radio@EDDIE.MIT.EDU, Info-Hams@SIMTEL20.ARPA
- Subject: SIMTEL20 device name change
-
- We have doubled the disk storage on SIMTEL20. In the process some
- changes had to be made. All directories, except those listed below,
- which were previously addressed as device PD: must now be addressed as
- PD1: These include <CPM*>, <MISC*> and <MSDOS*>.
-
- The following directories must be addressed as device PD2: <ADA*>,
- <CM*>, <VHDL*>, <UNIX*>, <MACINTOSH*>, and <PCNET>.
-
- --Keith Petersen
- 11-Nov-87 17:04:12-EST,977;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 11 Nov 87 17:04-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA29646@EDDIE.MIT.EDU>; Wed, 11 Nov 87 13:13:19 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA29617@EDDIE.MIT.EDU>; Wed, 11 Nov 87 13:12:25 EST
- Received: by june.cs.washington.edu (5.52.1/6.10)
- id AA10896; Wed, 11 Nov 87 10:13:24 PST
- Return-Path: <sdcsvax!brian@ucbvax.berkeley.edu>
- Message-Id: <8711111813.AA10896@june.cs.washington.edu>
- Date: 11 Nov 87 04:17:42 GMT
- From: sdcsvax!brian@ucbvax.berkeley.edu (Brian Kantor)
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: Re: ham domain
- References: <326@n8emr.UUCP>
- Reply-To: brian@sdcsvax.ucsd.edu (Brian Kantor)
-
- Domain registration for the amateur world is in progress. The people at
- the NIC and I have been having productive discussions, and we hope to
- have something to announce soon - perhaps before the new year.
-
- Brian Kantor WB6CYT UC San Diego
-
-
- 11-Nov-87 20:10:34-EST,1343;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 11 Nov 87 20:10-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA06584@EDDIE.MIT.EDU>; Wed, 11 Nov 87 18:31:23 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA06576@EDDIE.MIT.EDU>; Wed, 11 Nov 87 18:31:11 EST
- Received: by june.cs.washington.edu (5.52.1/6.10)
- id AA20918; Wed, 11 Nov 87 15:32:06 PST
- Return-Path: <hplabs!pyramid!voder!randy@eddie.mit.edu>
- Message-Id: <8711112332.AA20918@june.cs.washington.edu>
- Date: 11 Nov 87 02:15:25 GMT
- From: hplabs!pyramid!voder!randy@eddie.mit.edu (Randy Flatness)
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: Software control programs for pk232 and ts440.
- Keywords: pk232, ts440, software
-
-
- I have been experimenting with the AEA pk232 for a couple
- of months now, and find it to be quite impressive box!
- But now I would like to expand it's operation.
-
- What I am looking for is a control program which can control
- the pk232 (and the ts440 radio) on an a pc or mac with more
- features than a standard terminal emulator.
-
- So, if anyone has developed such a program or knows about one
- in public domain please let me know the details.
-
- I have seen one advertised in the mags and would also be interested
- in any user reports also.
-
- 73, and thanks in advance. Randy Flatness
-
-
- 11-Nov-87 20:55:19-EST,1343;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 11 Nov 87 20:55-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA06584@EDDIE.MIT.EDU>; Wed, 11 Nov 87 18:31:23 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA06576@EDDIE.MIT.EDU>; Wed, 11 Nov 87 18:31:11 EST
- Received: by june.cs.washington.edu (5.52.1/6.10)
- id AA20918; Wed, 11 Nov 87 15:32:06 PST
- Return-Path: <hplabs!pyramid!voder!randy@eddie.mit.edu>
- Message-Id: <8711112332.AA20918@june.cs.washington.edu>
- Date: 11 Nov 87 02:15:25 GMT
- From: hplabs!pyramid!voder!randy@eddie.mit.edu (Randy Flatness)
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: Software control programs for pk232 and ts440.
- Keywords: pk232, ts440, software
-
-
- I have been experimenting with the AEA pk232 for a couple
- of months now, and find it to be quite impressive box!
- But now I would like to expand it's operation.
-
- What I am looking for is a control program which can control
- the pk232 (and the ts440 radio) on an a pc or mac with more
- features than a standard terminal emulator.
-
- So, if anyone has developed such a program or knows about one
- in public domain please let me know the details.
-
- I have seen one advertised in the mags and would also be interested
- in any user reports also.
-
- 73, and thanks in advance. Randy Flatness
-
-
- 12-Nov-87 02:51:01-EST,5683;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 12 Nov 87 02:51-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA13073@EDDIE.MIT.EDU>; Wed, 11 Nov 87 23:33:52 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA13065@EDDIE.MIT.EDU>; Wed, 11 Nov 87 23:33:33 EST
- Received: by june.cs.washington.edu (5.52.1/6.10)
- id AA27437; Wed, 11 Nov 87 20:34:43 PST
- Return-Path: <delni.dec.com!goldstein@decwrl.dec.com>
- Message-Id: <8711120434.AA27437@june.cs.washington.edu>
- Date: 12 Nov 87 00:24:00 GMT
- From: delni.dec.com!goldstein@decwrl.dec.com (Fred R. Goldstein dtn226-7388)
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: re: Level 3 protocols, some comments
-
- In reply to KC2WZ's questions, here's my mildly opinionated answer.
- I'll try to avoid flamage.
-
- > 1. TCP/IP - which KA9Q, et al are proposing is the best way to go for
- > Level 3. Why? What are its advantages over the other protocols? What
- > are is disadvantages? (Come now we know nothing is perfect.)
-
- "TCP/IP" (DDN Protocol Suite) has a few advantages. It is a known,
- working technology. It is widely used outside hamdom, so there are
- existing implementations to check against and use if affordable. It's
- free with Berkeley Unix. It has the most popular applications
- (terminal, mail and file transfer) already defined. Performance is OK.
-
- It is connectionless, which is better or worse than
- connection-oriented based on your religion (CONS or CLNS). TCP (Layer
- 4) makes nice connections out of disjoint Layer 3 datagrams. This means
- that every packet must be routed, but there's no need for the router
- ("gateway") to have knowledge of connections; that's only in the end
- nodes. This could make for simpler gateways. UDP supports
- connectionless applications too.
-
- There's no real smart routing in the KA9Q version of IP, just
- some clever table-lookup, but some of us are looking at writing an
- automatic routing program for it. Addressing is a problem: It's 32 bits,
- so you can't map in a call sign or anything. Directories are thus needed.
-
- > 2. COSI - I'm not certain who developed the idea. Again, why should this
- > be adopted as a Level 3 standard? What are its advantages? What are
- > its disadvantages?
-
- This comes from a few guys in New Jersey (RATTS?). It's
- connection-oriented, recycling some X.25 Level 3 protocols. Again a
- religious matter; it doesn't need a smart transport layer since the
- network keeps track of packets, but routers must keep track of virtual
- circuits along the way. I don't think it's on the air yet.
-
- > 3. NET/ROM - Is this the same as COSI? It SEEMS to be the way things are
- > going in the New Jersey/New York area on 2 meters. Am I wrong? Why
- > should it be adopted? What are its advantages? Its disadvantages?
-
- This is a home-grown protocol which runs in TNC2s. It's rather
- limited in size and scope due to its ROM distribution, but that makes it
- easy to build little networks in a hurry. It uses datagrams internally
- but provides a connection-oriented interface to AX.25. Good side: It's
- expedient. Bad side: It's inflexible (burned in ROM) and thus won't
- "grow up" to easily support a big nationwide network, unless it's
- reimplemented, at which point it's not so expedient.
-
- NET/ROM does beat the pants off of digipeating.
-
- > 4. OTHER PROTOCOLS? Are there others be proposed? These are the only
- > ones I have heard mentioned. As above what are the advantages and
- > disadvantages?
-
- I've proposed "AmOSInet", which is an OSI-derived datagram network.
- Procedurally it's a lot like TCP/IP but the addressing is different,
- using a two-level scheme (regions and sections). The region is the
- Maidenhead and the station uses its call (or anything else) as its
- address within that section, so you don't need a directory. This is
- just an architectural strawman now, not a product, and I expect it will
- evolve slowly out of TCP/IP.
-
- Down in Florida they have their own X.25L3-based connection
- oriented network. Like COSI, it uses telephone area codes as the first
- part of the address. Texas also has a local "Texnet" protocol, I think.
-
- > 5. Much of the work seems to centered are the IBM-PC and clones.
- > Obviously, these are very popular micros. But there are a variety of
- > other makes which don't seem to be supported. My machine (Tandy Color
- > Computer 3) is one example. Other 6809/68000 machines also SEEM to be
- > left in the cold. Is this true or just my impression?
-
- Phil's TCP/IP is being ported to several machines (the Mac is
- out, and I think Amiga too) and its source is as portable as they can
- make it. Expect more 68k machines to be supported soon (where's the
- Atari ST version btw?)
-
- > 6. My final(?) question (for now), how to go about setting up the TNC
- > and computer to use Level 3? Is it necessary that everyone has Level 3
- > running in their TNC or is this a waste? I have a TAPR TNC-1 running
- > WA8DED's software. Will I have to burn a new EPROM with the new code?
- > How will my computer be interfaced with the TNC?
-
- L3 doesn't run in the TNC, typically; it runs in the end node
- and maybe the gateway. To use TCP/IP, it's best to run the TNC in KISS
- mode and let the PC handle the AX.25 procedures. But you typically can
- use "transparent" AX.25 on the TNC too. NET/ROM, btw, eats a whole TNC
- which no longer supports terminals, just routing. Stick it on a
- mountaiantop somewhere.
- fred k1io (FN42jk) goldstein@delni.dec.com
-
-
- 12-Nov-87 21:38:02-EST,3039;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 12 Nov 87 21:38-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA05614@EDDIE.MIT.EDU>; Thu, 12 Nov 87 18:59:40 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA05587@EDDIE.MIT.EDU>; Thu, 12 Nov 87 18:58:58 EST
- Received: by june.cs.washington.edu (5.52.1/6.10)
- id AA27669; Thu, 12 Nov 87 16:00:03 PST
- Return-Path: <rutgers!bellcore!faline!karn@eddie.mit.edu>
- Message-Id: <8711130000.AA27669@june.cs.washington.edu>
- Date: 12 Nov 87 20:27:47 GMT
- From: rutgers!bellcore!faline!karn@EDDIE.MIT.edu (Phil R. Karn)
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: Re: Level 3 protocols, some comments
- Summary: datagrams vs virtual circuits - regulatory considerations
- References: <8711112140.AA05493@decwrl.dec.com>
-
- > It [TCP/IP] is connectionless, which is better or worse than
- > connection-oriented based on your religion...
-
- There are many strong and objective technical arguments for
- connectionless network protocols, especially over broadcast media like
- radio. I therefore disagree with your use of the word "religion".
- However, I'll not repeat them here since I've already given them many
- times. But a recent development on the regulatory front gives added
- impetus to use datagrams.
-
- Australia's amateur packet radio regulations allow the use of any
- protocol as long as *every packet* identifies the originator, the
- station actually transmitting the packet (if different from the
- originator), and the destination. This is tantamount to REQUIRING a
- datagram protocol at both the link and network layers. Furthermore, this
- means that "protocol translation gateways" or any other mechanisms that
- violate the end-to-end significance of the network layer addresses are
- prohibited.
-
- Australian hams were recently ordered to shut down their NET/ROM network
- for this reason. Even though NET/ROM uses a perfectly acceptable
- datagram network protocol internally, it is not spoken by the end users.
- Rather, "uplink" and "downlink" sites act as protocol translation
- gateways between "vanilla" AX.25 and the internal NET/ROM protocols.
-
- It is not known whether the "identification" must be actual callsigns or
- whether it can be a unique bit pattern that can be translated back into
- a callsign if desired. If the latter is true, IP addresses are
- sufficient since their assignments are public and can be uniquely mapped
- back into callsigns. In the former case, it would be fairly
- straightforward to define an IP header option containing callsigns.
-
- While Australia may be a relatively small (in population) country, it is
- my understanding that many European countries are using their rules as a
- model as they draft their own packet radio regulations. There is
- considerable irony here, since we've heard a lot from the COSI camp
- about the need for "international acceptance" of amateur packet radio
- protocols. Yet the protocols they are pushing do not satisfy the
- Australian identification requirements.
-
- Phil
-
-
- 12-Nov-87 22:09:50-EST,1394;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 12 Nov 87 22:09-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA06904@EDDIE.MIT.EDU>; Thu, 12 Nov 87 20:02:15 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA06897@EDDIE.MIT.EDU>; Thu, 12 Nov 87 20:02:02 EST
- Received: from adam.dg.com by RELAY.CS.NET id aa29006; 12 Nov 87 19:34 EST
- Received: by adam.DG.COM with sendmail; Thu, 12 Nov 87 19:34:00 est
- Reply-To: Podsiadlo_Jim%WDS.CEO.DG.COM@adam.dg.com
- Mmdf-Warning: Parse error in original version of preceding line at RELAY.CS.NET
- Received: from WDS by adam.DG.COM with CEO; Thu, 12 Nov 87 16:18:54 EST
- Date: Thu, 12 Nov 87 15:03:42 EST
- From: Podsiadlo_Jim%wds.ceo.dg.com@RELAY.CS.NET
- Message-Id: <29.002685@adam.DG.COM>
- To: packet-radio%eddie.mit.edu@RELAY.CS.NET
- Subject: AMPRNET and TCP/IP
-
- I have been using TCP/IP in the Worcester area for some time now.
- Although I have been very successful with the few stations out here,
- I would like to help get something more significant going. Is there
- any interest in really getting the "net" established. Maybe some more
- full time nodes, other than W1MX. Or maybe some higher data rates on
- 440. How about trying to link the Boston area into the pocket of
- activity that exists in the DC area, or even RI. Is anyone else
- trying to advance the state of AMPRNET in the area???????
- 12-Nov-87 22:44:46-EST,1394;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 12 Nov 87 22:44-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA06904@EDDIE.MIT.EDU>; Thu, 12 Nov 87 20:02:15 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA06897@EDDIE.MIT.EDU>; Thu, 12 Nov 87 20:02:02 EST
- Received: from adam.dg.com by RELAY.CS.NET id aa29006; 12 Nov 87 19:34 EST
- Received: by adam.DG.COM with sendmail; Thu, 12 Nov 87 19:34:00 est
- Reply-To: Podsiadlo_Jim%WDS.CEO.DG.COM@adam.dg.com
- Mmdf-Warning: Parse error in original version of preceding line at RELAY.CS.NET
- Received: from WDS by adam.DG.COM with CEO; Thu, 12 Nov 87 16:18:54 EST
- Date: Thu, 12 Nov 87 15:03:42 EST
- From: Podsiadlo_Jim%wds.ceo.dg.com@RELAY.CS.NET
- Message-Id: <29.002685@adam.DG.COM>
- To: packet-radio%eddie.mit.edu@RELAY.CS.NET
- Subject: AMPRNET and TCP/IP
-
- I have been using TCP/IP in the Worcester area for some time now.
- Although I have been very successful with the few stations out here,
- I would like to help get something more significant going. Is there
- any interest in really getting the "net" established. Maybe some more
- full time nodes, other than W1MX. Or maybe some higher data rates on
- 440. How about trying to link the Boston area into the pocket of
- activity that exists in the DC area, or even RI. Is anyone else
- trying to advance the state of AMPRNET in the area???????
- 13-Nov-87 16:44:24-EST,1197;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 13 Nov 87 16:44-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA24327@EDDIE.MIT.EDU>; Fri, 13 Nov 87 12:24:14 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA24320@EDDIE.MIT.EDU>; Fri, 13 Nov 87 12:23:54 EST
- Message-Id: <8711131723.AA24320@EDDIE.MIT.EDU>
- Received: from DHHDESY3.BITNET by wiscvm.wisc.edu ; Fri, 13 Nov 87 11:23:53 CDT
- Date: 13 NOV 87 16:20:11 MEZ
- From: F33PAP%DHHDESY3.BITNET@wiscvm.wisc.edu
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: Re: Repeater advice anyone?
-
- Hello!
- We are preparing for a duplex digipeater on 70cm. Just now it is only
- AF in, AF out, without regeneration. The next step is a regenerator
- with a K9NG state machine PROM. Yes, it should work! Clock a K9NG PROM
- state machine with 16 times the data rate and use output bit 4 (I think)
- as the regenerated data output.
- We also want to have a TNC-2 with NET/ROM in parallel. The TNC-2 will
- not transmit while the receiver is on... Just switch the data with the
- key line of the TNC from the PROM to the TNC.
- If you need more details, send me a note.
- vy 73, Karl-Heinz, DK8HI, F33PAP@DHHDESY3.BITNET
-
- 13-Nov-87 22:01:02-EST,1115;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 13 Nov 87 22:01-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA04751@EDDIE.MIT.EDU>; Fri, 13 Nov 87 20:07:05 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA04695@EDDIE.MIT.EDU>; Fri, 13 Nov 87 20:04:02 EST
- Received: by LANL.GOV (5.54/1.14)
- id AA01037; Thu, 12 Nov 87 11:31:51 MST
- Received: by MILNET-GW.GOV (5.54/5.17)
- id AA06520; Thu, 12 Nov 87 08:23:12 MST
- Received: by a (5.51/5.17)
- id AA24458; Thu, 12 Nov 87 08:22:35 MST
- Date: Thu, 12 Nov 87 08:22:35 MST
- From: djw%a@LANL.GOV (Dave Wade)
- Message-Id: <8711121522.AA24458@a>
- To: PACKET-RADIO@eddie.mit.edu, ihnp4!homxb!hotps!ka2qhd!kc2wz@eddie.mit.edu
- Subject: Re: Level 3 protocols - some info?
-
- Might I suggest that a good source of information on the questions you are
- asking is:
- "The Elements of Networking Style"
- by M.A. Padlipsky
- Published by Prentice-Hall., of Englewood Cliffs, N.J. 07632
- ISBN 0-13-268129-3
-
- You'll enjoy it more than you enjoy the usual fare here... And,
- you'll undoubtedly learn what you asked.
-
- Dave
- WB5PFS
- 14-Nov-87 08:36:54-EST,1441;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 14 Nov 87 08:36-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA14988@EDDIE.MIT.EDU>; Sat, 14 Nov 87 07:18:33 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA14977@EDDIE.MIT.EDU>; Sat, 14 Nov 87 07:18:16 EST
- Resent-Message-Id: <8711141218.AA14977@EDDIE.MIT.EDU>
- Date: Friday, 13 November 1987 17:58-MST
- Message-Id: <KPETERSEN.12350524696.BABYL@SIMTEL20.ARPA>
- Sender: ZMTKeidel%UCIVMSA.BITNET@WISCVM.WISC.EDU
- From: ZMTKeidel%UCIVMSA.BITNET@WISCVM.WISC.EDU
- To: INFO-HAMS-REQUEST@SIMTEL20.ARPA
- Subject: AEA PK-232 control (terminal) program
- Resent-From: KPETERSEN@SIMTEL20.ARPA
- Resent-To: packet-radio@eddie.mit.edu
- Resent-Date: Sat 14 Nov 1987 05:18-MST
-
- I am interested in developing a control program which will interface
- an Apple MacIntosh with the AEA PK-232. However, I have heard that
- there are not many HAMS with MacIntosh computers! I have one, my
- roomate has three, and even the engineers at AEA have Mac's.
-
- Before I invest the time in writing a full-function program, which may
- also include a control section for the most popular computer
- controlled rigs such as the TS-440s, I would like to know if there is
- interest and whether or not Hams do own MacIntosh computers!
-
- Please respond directly to my bitnet address MKEIDE@UCI.BITNET or if
- your software will allow it, MKEIDEL@UCI.BITNET.
-
- KA6JAF-- 73'S...
- 14-Nov-87 11:45:24-EST,1451;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 14 Nov 87 11:45-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA17862@EDDIE.MIT.EDU>; Sat, 14 Nov 87 10:34:08 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA17855@EDDIE.MIT.EDU>; Sat, 14 Nov 87 10:33:57 EST
- Resent-Message-Id: <8711141533.AA17855@EDDIE.MIT.EDU>
- Date: Friday, 13 November 1987 09:31-MST
- Message-Id: <KPETERSEN.12350560383.BABYL@SIMTEL20.ARPA>
- Sender: <RME8650%TAMSIGMA.BITNET@WISCVM.WISC.EDU> (Robert Eden - N5GWY)
- From: <RME8650%TAMSIGMA.BITNET@WISCVM.WISC.EDU> (Robert Eden - N5GWY)
- To: info-hams@SIMTEL20.ARPA
- Subject: packet radio - VAX gateway
- Resent-From: KPETERSEN@SIMTEL20.ARPA
- Resent-To: packet-radio@eddie.mit.edu
- Resent-Date: Sat 14 Nov 1987 08:34-MST
-
- I'm in the finishing stages of a gateway between the W0RLI mailbox network
- and VAX mail.
-
- I've written a detached process that talks to a TNC over a standard VAX
- serial port (LAT is ok) and uses that port to receive and transmit messages
- to W0RLI mailboxes. Mail is sent to VAX uses through the MAIL utility.
-
- Anyone who is interested in running this system should contact me and
- I'll send the source and documentation.
-
- Robert Eden - N5GWY
-
- RME8650@TAMSIGMA.BITNET <<< first choice
- RME8650@TAMVENUS.BITNET <<< second choice
-
- reden@sys1.uucp <<< after Dec. 15th.
- rgd059@mipl3.jpl.nasa.gov<<< if all else fails.
- 14-Nov-87 13:58:28-EST,1451;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 14 Nov 87 13:58-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA17862@EDDIE.MIT.EDU>; Sat, 14 Nov 87 10:34:08 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA17855@EDDIE.MIT.EDU>; Sat, 14 Nov 87 10:33:57 EST
- Resent-Message-Id: <8711141533.AA17855@EDDIE.MIT.EDU>
- Date: Friday, 13 November 1987 09:31-MST
- Message-Id: <KPETERSEN.12350560383.BABYL@SIMTEL20.ARPA>
- Sender: <RME8650%TAMSIGMA.BITNET@WISCVM.WISC.EDU> (Robert Eden - N5GWY)
- From: <RME8650%TAMSIGMA.BITNET@WISCVM.WISC.EDU> (Robert Eden - N5GWY)
- To: info-hams@SIMTEL20.ARPA
- Subject: packet radio - VAX gateway
- Resent-From: KPETERSEN@SIMTEL20.ARPA
- Resent-To: packet-radio@eddie.mit.edu
- Resent-Date: Sat 14 Nov 1987 08:34-MST
-
- I'm in the finishing stages of a gateway between the W0RLI mailbox network
- and VAX mail.
-
- I've written a detached process that talks to a TNC over a standard VAX
- serial port (LAT is ok) and uses that port to receive and transmit messages
- to W0RLI mailboxes. Mail is sent to VAX uses through the MAIL utility.
-
- Anyone who is interested in running this system should contact me and
- I'll send the source and documentation.
-
- Robert Eden - N5GWY
-
- RME8650@TAMSIGMA.BITNET <<< first choice
- RME8650@TAMVENUS.BITNET <<< second choice
-
- reden@sys1.uucp <<< after Dec. 15th.
- rgd059@mipl3.jpl.nasa.gov<<< if all else fails.
- 16-Nov-87 14:33:47-EST,2169;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 16 Nov 87 14:33-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA28995@EDDIE.MIT.EDU>; Mon, 16 Nov 87 11:30:06 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA28979@EDDIE.MIT.EDU>; Mon, 16 Nov 87 11:29:49 EST
- Received: from ROOK.SCRC.Symbolics.COM by VALLECITO.SCRC.Symbolics.COM via CHAOS with CHAOS-MAIL id 178180; Mon 16-Nov-87 11:04:19 EST
- Date: Mon, 16 Nov 87 11:03 EST
- From: Henry Minsky <hqm@VALLECITO.SCRC.Symbolics.COM>
- Subject: Repeater advice anyone?
- To: David_G._Michelson%UBC.MAILNET@um.cc.umich.edu, packet-radio@EDDIE.MIT.EDU
- In-Reply-To: <8711110030.AA22371@june.cs.washington.edu>
- Message-Id: <871116110347.5.HQM@ROOK.SCRC.Symbolics.COM>
-
- Date: Tue, 10 Nov 87 11:32:08 PST
- From: David_G._Michelson%UBC.MAILNET@um.cc.umich.edu
- To: packet-radio@EDDIE.MIT.EDU
- Subject: Repeater advice anyone?
-
- Re: 440 MHz Repeater for Packet Radio
-
- One of the most interesting things about packet radio (in my opinion)
- is the different ways one can approach the problem of repeating.
-
- The simplex digipeater is the simplest but, as subscribers to this
- mailing list are aware, it has severe disadvantages due to contention
- between the orginal packet, all its copies, and the many acknowledge-
- ments that fly back and forth up and down the link.
-
- NET/ROM is a significant improvement to the simple digipeater concept
- which, as I understand, uses standard TNC-2 hardware but enhanced
- firmware.
-
- Cross-band and cross-channel packet radio repeaters offer significant
- advantages because they eliminate the possibility of collision between
- originator-repeater traffic and repeater-addressee traffic.
-
- And then there are the many people who use conventional duplex repeaters
- for packet traffic (i.e., no regeneration).
-
- Are these the issues you were concerned with or are you after the
- technical details of a specific implementation?
-
- I have been hearing a lot about "regeneration" in a digital full duplex
- repeater. What is that and how does it work?
-
-
- 16-Nov-87 17:26:52-EST,2191;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 16 Nov 87 17:26-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA02758@EDDIE.MIT.EDU>; Mon, 16 Nov 87 14:38:31 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA02752@EDDIE.MIT.EDU>; Mon, 16 Nov 87 14:38:15 EST
- Received: by june.cs.washington.edu (5.52.1/6.10)
- id AA26954; Mon, 16 Nov 87 11:39:40 PST
- Return-Path: <bellcore!faline!karn@eddie.mit.edu>
- Message-Id: <8711161939.AA26954@june.cs.washington.edu>
- Date: 16 Nov 87 04:33:01 GMT
- From: bellcore!faline!karn@EDDIE.MIT.edu (Phil R. Karn)
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: Re: Packet freqs
- Summary: weak-signal people have 500 khz
- References: <698@uop.UUCP> <1532@faline.bellcore.com> <250@splut.UUCP>
-
- > Where did those recommendations come from??
-
- The ARRL Digital Committee, of which I am a member.
-
- > Has anyone floated them past the 220 weak-signal types??
-
- The weak-signal types on 220 have 500 Khz of spectrum reserved to them,
- 220.0 - 220.5. Given the very low levels of weak signal activity on that
- band, I should think this would be more than plenty. Note that unlike
- 2 meters, modes other than CW are allowed in this segment. Wideband
- packet could in theory operate here.
-
- The segment just above 220.5 that was recommended for wideband packet
- is already listed as "experimental and control links". The major problem
- in getting approval for these frequencies in most areas seems to be
- that the local coordination committees seldom have complete records for
- repeater control links, which by their nature are not widely advertised.
-
- Which gives me an idea: many repeaters have separate, dedicated control
- links mainly to satisfy FCC requirements. They can and do have secondary
- command decoders on their regular inputs, and most routine commanding is
- done this way. The backup control requirement for the voice repeater
- could instead be provided as a secondary function of a co-located packet
- switch. In exchange for providing secure and more sophisticated control
- functions, the packet guys could get an existing antenna and frequency
- assignment at a good site, and everybody wins.
-
- Phil
-
-
- 16-Nov-87 18:52:24-EST,1819;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 16 Nov 87 18:52-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA02740@EDDIE.MIT.EDU>; Mon, 16 Nov 87 14:37:44 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA02729@EDDIE.MIT.EDU>; Mon, 16 Nov 87 14:37:20 EST
- Received: by june.cs.washington.edu (5.52.1/6.10)
- id AA26900; Mon, 16 Nov 87 11:38:41 PST
- Return-Path: <uunet!nuchat!splut!jay@eddie.mit.edu>
- Message-Id: <8711161938.AA26900@june.cs.washington.edu>
- Date: 15 Nov 87 14:20:20 GMT
- From: uunet!nuchat!splut!jay@EDDIE.MIT.edu (Jay Maynard)
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: Re: Packet freqs
- Summary: Uh, Phil...
- References: <698@uop.UUCP> <1532@faline.bellcore.com>
-
- In article <1532@faline.bellcore.com>, karn@faline.bellcore.com (Phil R. Karn) writes:
- > The band plan recommendation is to put conventional (i.e., AFSK/FM) packet
- > on the frequencies 221.01, .03, ... with 20 khz spacing. High speed
- > (100 Khz) channels are 220.55, .65, .75, .85 and .95.
-
- Where did those recommendations come from??
- Has anyone floated them past the 220 weak-signal types??
-
- > These recommendations are only that; you must still clear them with your
- > local frequency coordination committee. And, of course, all this assumes
- > we keep that portion of the band.
-
- A hope we all share...
- Beware, though: some coordination committees only espouse jurisdiction in
- the repeater bands, and have little-to-no contact with the folks who this
- would really affect.
-
- --
- Jay Maynard, K5ZC (@WB5BBW)...>splut!< | uucp: uunet!nuchat!splut!jay
- Never ascribe to malice that which can | or: academ!uhnix1!--^
- adequately be explained by stupidity. | GEnie: JAYMAYNARD CI$: 71036,1603
- The opinions herein are shared by none of my cats, much less anyone else.
-
-
- 16-Nov-87 18:52:41-EST,2191;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 16 Nov 87 18:52-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA02758@EDDIE.MIT.EDU>; Mon, 16 Nov 87 14:38:31 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA02752@EDDIE.MIT.EDU>; Mon, 16 Nov 87 14:38:15 EST
- Received: by june.cs.washington.edu (5.52.1/6.10)
- id AA26954; Mon, 16 Nov 87 11:39:40 PST
- Return-Path: <bellcore!faline!karn@eddie.mit.edu>
- Message-Id: <8711161939.AA26954@june.cs.washington.edu>
- Date: 16 Nov 87 04:33:01 GMT
- From: bellcore!faline!karn@EDDIE.MIT.edu (Phil R. Karn)
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: Re: Packet freqs
- Summary: weak-signal people have 500 khz
- References: <698@uop.UUCP> <1532@faline.bellcore.com> <250@splut.UUCP>
-
- > Where did those recommendations come from??
-
- The ARRL Digital Committee, of which I am a member.
-
- > Has anyone floated them past the 220 weak-signal types??
-
- The weak-signal types on 220 have 500 Khz of spectrum reserved to them,
- 220.0 - 220.5. Given the very low levels of weak signal activity on that
- band, I should think this would be more than plenty. Note that unlike
- 2 meters, modes other than CW are allowed in this segment. Wideband
- packet could in theory operate here.
-
- The segment just above 220.5 that was recommended for wideband packet
- is already listed as "experimental and control links". The major problem
- in getting approval for these frequencies in most areas seems to be
- that the local coordination committees seldom have complete records for
- repeater control links, which by their nature are not widely advertised.
-
- Which gives me an idea: many repeaters have separate, dedicated control
- links mainly to satisfy FCC requirements. They can and do have secondary
- command decoders on their regular inputs, and most routine commanding is
- done this way. The backup control requirement for the voice repeater
- could instead be provided as a secondary function of a co-located packet
- switch. In exchange for providing secure and more sophisticated control
- functions, the packet guys could get an existing antenna and frequency
- assignment at a good site, and everybody wins.
-
- Phil
-
-
- 17-Nov-87 19:47:31-EST,2526;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 17 Nov 87 19:47-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA00627@EDDIE.MIT.EDU>; Tue, 17 Nov 87 17:50:40 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA00584@EDDIE.MIT.EDU>; Tue, 17 Nov 87 17:47:38 EST
- Received: by cod.nosc.mil (5.58/1.27)
- id AA02165; Tue, 17 Nov 87 12:25:59 PST
- Date: Tue, 17 Nov 87 12:25:59 PST
- From: coleman@cod.nosc.mil (Jeffrey L. Coleman)
- Message-Id: <8711172025.AA02165@cod.nosc.mil>
- To: packet-radio@eddie.mit.edu
- Subject: DIGICOM>64
-
-
- I obtained DIGICOM>64 from an American ham who distributes it for $1.00
- if you send him a disk and stamped mailer. Sorry, I don't have his
- address with me but I got it from QST, October 1987. He includes
- 2 schematics for modems and hints on selecting and building one of them.
-
- The software seems to be very well written and documentation is included
- on the disk. I built the modem using the XR2206 and XR2211 for about
- $20 in parts. The chip set cost $7.50, the rest was mostly the box,
- board, connectors, trimpots, etc.
-
- I have 2 questions about the system, one trivial and one critical:
-
- Trivial: Why is it necessary to set HBAUD to 1156 instead of 1200?
- Are European Commodores different than U.S. Commodores? The
- explanation given in the documentation that this corrects the
- clock for the American 60Hz vs. the European 50Hz doesn't make
- sense to me.
-
- Critical: Why can't I connect to anything? I have never operated
- packet before so maybe there is something simple I'm overlooking.
- I can copy packets off the air (and have a pretty good idea what
- most of the codes and symbols mean) but when I command "c wa6mu"
- my computer always gives up after 6 retries. I can hear my
- packets being transmitted by listening to another radio so I
- know the connection to the transmitter is good but I never
- copy any response from the addressee station.
-
- I used a frequency counter to adjust my output frequencies to
- 1200 and 2200 Hz. I have tried various settings on the output
- level to prevent overmodulation but it seems to make no difference.
- I have tried to connect to several digipeaters and BBS's including
- some that I saw others using.
-
- Any ideas or suggestions? I'm running out of things to try.
-
- Thanks,
- Jeff Coleman, KI6NL
- coleman@cod.nosc.mil
-
- Disclaimer: The above views are those of a normally sane, rational person
- who has been staring at a homebrew project too long.
-
- -------
-
- 17-Nov-87 22:46:36-EST,2801;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 17 Nov 87 22:46-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA01716@EDDIE.MIT.EDU>; Tue, 17 Nov 87 18:52:10 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA01704@EDDIE.MIT.EDU>; Tue, 17 Nov 87 18:51:59 EST
- Received: from ROOK.SCRC.Symbolics.COM by VALLECITO.SCRC.Symbolics.COM via CHAOS with CHAOS-MAIL id 178890; Tue 17-Nov-87 18:52:41 EST
- Date: Tue, 17 Nov 87 18:52 EST
- From: Henry Minsky <hqm@VALLECITO.SCRC.Symbolics.COM>
- Subject: DIGICOM>64
- To: coleman@cod.nosc.mil, packet-radio@eddie.mit.edu
- In-Reply-To: <8711172025.AA02165@cod.nosc.mil>
- Message-Id: <871117185228.7.HQM@ROOK.SCRC.Symbolics.COM>
-
- From: coleman@cod.nosc.mil (Jeffrey L. Coleman)
- Message-Id: <8711172025.AA02165@cod.nosc.mil>
- To: packet-radio@eddie.mit.edu
- Subject: DIGICOM>64
- ...
- ...
- I have 2 questions about the system, one trivial and one critical:
-
- Trivial: Why is it necessary to set HBAUD to 1156 instead of 1200?
- Are European Commodores different than U.S. Commodores? The
- explanation given in the documentation that this corrects the
- clock for the American 60Hz vs. the European 50Hz doesn't make
- sense to me.
-
- Critical: Why can't I connect to anything? I have never operated
- packet before so maybe there is something simple I'm overlooking.
- I can copy packets off the air (and have a pretty good idea what
- most of the codes and symbols mean) but when I command "c wa6mu"
- my computer always gives up after 6 retries. I can hear my
- packets being transmitted by listening to another radio so I
- know the connection to the transmitter is good but I never
- copy any response from the addressee station.
-
- I used a frequency counter to adjust my output frequencies to
- 1200 and 2200 Hz. I have tried various settings on the output
- level to prevent overmodulation but it seems to make no difference.
- I have tried to connect to several digipeaters and BBS's including
- some that I saw others using.
-
- Any ideas or suggestions? I'm running out of things to try.
-
- I suggest that you try setting any parameters which control the
- "TXDELAY" and/or "TXTAIL" (or the equivalent in your system).
-
- These are parameters which control how long a delay the computer adds
- after it keys the transmitter, but before sending the packet data, and
- likewise, how long it holds the transmitter keyed after the last packet
- bits have been sent.
-
- I was unable to sucessfully transmit packets using my Yaesu handheld,
- until I set something like a 50 millisecond txdelay, and about 20
- millisecond tail.
-
- Some of these transcievers can take even longer to key up, I've heard.
-
-
- 18-Nov-87 16:35:05-EST,1616;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 18 Nov 87 16:35-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA18332@EDDIE.MIT.EDU>; Wed, 18 Nov 87 12:26:14 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA18173@EDDIE.MIT.EDU>; Wed, 18 Nov 87 12:18:05 EST
- Received: by cod.nosc.mil (5.58/1.27)
- id AA25856; Wed, 18 Nov 87 08:50:33 PST
- Date: Wed, 18 Nov 87 08:50:33 PST
- From: coleman@cod.nosc.mil (Jeffrey L. Coleman)
- Message-Id: <8711181650.AA25856@cod.nosc.mil>
- To: packet-radio@eddie.mit.edu
- Subject: DIGICOM>64 (continued)
-
-
- I made an error in yesterday's posting. The article on DIGICOM64 was in
- November issue of QST (not October), on page 59. To quote the article,
- "The software, documentation and modem schematic may be obtained by sending
- a blank disk, a self-addressed disk mailer with enough postage for at least
- 3 ounces and $1 for the cost of the printed documentation to Barry Kutner,
- W2UP, 286 Leedom Way, Newtown, PA 18940."
-
- Now the good news: SUCCESS!
- Last night I finally found a station which I could connect to. Then I also
- connected to myself using him as a digipeater. The number of retries
- seemed high and I lost the connection a couple of times when sending long
- packets so I'll still be playing with the system trying to improve it (re-tune
- the transmitter, shorten the cables) but now it has gotten more interesting.
-
- If others are interested, I'll report later on my impressions of it and
- problems I've had.
-
- 73,
- Jeff Coleman, KI6NL
- coleman@cod.nosc.mil
-
- Disclaimer: I hereby disclaim.
- ------
- -------
-
- 18-Nov-87 19:15:46-EST,3581;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 18 Nov 87 19:15-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA25585@EDDIE.MIT.EDU>; Wed, 18 Nov 87 16:54:35 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA25567@EDDIE.MIT.EDU>; Wed, 18 Nov 87 16:54:15 EST
- Message-Id: <8711182154.AA25567@EDDIE.MIT.EDU>
- Received: from MITVMA.MIT.EDU by MITVMA.MIT.EDU ; Wed, 18 Nov 87 16:53:30 EST
- Received: from TRIUMFCL.BITNET (ROSK) by MITVMA.MIT.EDU (Mailer X1.25) with
- BSMTP id 8079; Wed, 18 Nov 87 16:19:33 EST
- Date: Wed, 18 Nov 87 13:12 PST
- From: <ROSK%TRIUMFCL.BITNET@MITVMA.MIT.EDU>
- Subject: Re: DIGICOM>64 Problems.
- To: PACKET-RADIO@EDDIE.MIT.EDU
- X-Original-To: pack, ROSK
-
- In answer to a couple of questions from Jeffrey L. Coleman
- <coleman@cod.nosc.mil> about the DIGICOM>64 software:
-
- > Why is it necessary to set HBAUD to 1156 instead of 1200?
- > Are European Commodores different than U.S. Commodores? The
- > explanation given in the documentation that this corrects the
- > clock for the American 60Hz vs. the European 50Hz doesn't make
- > sense to me.
-
- The baud rate timing is derived from the CPU clock via one of the
- programmable counters. The problem is that the European CPU clock is
- at a slightly different frequency than the USA clock. This is because
- the Video output is also derived from the CPU clock. The European
- version produces a PAL picture (625 line - 50 frame/sec) and the USA
- version an NTSC picture (525 line - 60 frame/sec.) The timings for the
- two standards are quite different, but most of the difference is taken
- care of by using a different version of the VIC chip. (Thats what generates
- the video.) BUT, it cant quite do it all - the primary crystal frequency
- has to be slightly different ~17MHz for PAL, ~14MHz for NTSC. After some
- dividing in the (different) VICs, this results in CPU clock rates of
- 0.98 MHz for PAL and 1.02MHz for NTSC. So the NTSC CPU clock runs a bit
- faster, and to get 1200 baud, you have to tell the program to run 1156bd
- instead. The program could be "fixed" by adding a USA/EU switch, but I
- for one can put up with this quirk.
-
- > Why can't I connect to anything? I have never operated
- > packet before so maybe there is something simple I'm overlooking.
- > I can copy packets off the air (and have a pretty good idea what
- > most of the codes and symbols mean) but when I command "c wa6mu"
- > my computer always gives up after 6 retries. I can hear my
- > packets being transmitted by listening to another radio so I
- > know the connection to the transmitter is good but I never
- > copy any response from the addressee station.
-
- Get another packet station to MONITOR your packets. Talk to them on
- the phone, or 2m, have them tell you what they see. Try setting TXDELAY
- larger - try 20 for starters. This gives the TX longer to "get on the air"
- before the real data is transmitted. This was my problem when getting
- started.
- You can test your software and modem without the radio - connect the
- modem output to the modem input. Set the following:
- MYCALL KI6NL < thats your call sign.
- CONOK ON < that allows someone to connect to you
- FULL ON < that allows full duplex operation
- Now try CONNECT KI6NL
- You should connect to yourself! Note - it doesn't always work perfectly.
- The C64 is working hard to generate output and copy the input at the same
- time, all on timer interrupts, when running Full Duplex. You may not get
- perfect packet transmission every time.
-
- 73 VE7AII.
-
- 18-Nov-87 19:40:37-EST,4837;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 18 Nov 87 19:40-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA22527@EDDIE.MIT.EDU>; Wed, 18 Nov 87 15:03:36 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA22499@EDDIE.MIT.EDU>; Wed, 18 Nov 87 15:02:56 EST
- Received: by june.cs.washington.edu (5.52.1/6.10)
- id AA09409; Wed, 18 Nov 87 12:04:07 PST
- From: ihnp4!homxb!mtuxo!mtune!petsd!pedsgo!pedsga!tsdiag!tom@eddie.mit.edu
- Return-Path: <ihnp4!homxb!mtuxo!mtune!petsd!pedsgo!pedsga!tsdiag!tom@eddie.mit.edu>
- Message-Id: <8711182004.AA09409@june.cs.washington.edu>
- Date: 13 Nov 87 15:40:35 GMT
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: Re: Level 3 protocols - some info?
- Keywords: TCPIP,COSI,NETROM
- References: <242@ka2qhd.UUCP>
-
- Bob, You are right it's time these questions were asked by someone who
- has no stake in the protocol wars!!!
-
- Your questions in order...
-
- 1) Is TCP/IP best?
- There are people who do believe that this is the only networking
- protocol that can solve everyone's needs. It does have advantages
- some environments, I personally do not agree. The basic disadvantage
- is that it is an all or nothing and you must have the whole protocol
- running in your computer to access the network.
-
- 2) COSI Switch - What is it?
- I am the programmer writing the switch, it is based on CCITT X.25,
- (aka ISO 8208) which is the International standard used to connect
- data networks throughout the world. [note that tcp/ip is also used
- globally, but isn't accepted by all PTT's, in fact many tcp/ip
- connections are running on top of X.25]
-
- Why X.25? I believe that to run a network on packet radio it is very
- important to provide error recovery at all levels. [for ip it's end to
- end, very much like good 'ole digipeating is now] Which means that if
- any data gets lost you get notified and can re-send the information.
-
- An X.25 network can not be an ad hoc network, it must be expanded in
- a coordinated fashion. [remember kaos on 145.01?]
-
- Politics, many PTT's won't accept a new protocol, the COSI suite is
- going to be following the same standards that they are seeing in the
- commerical world, how can they say no!
-
- The ISO protocols are the next generation of networking which the
- commerical world is going to. TCP/IP was developed on the ARPAnet
- [a US DoD funded project] and while many companies use this protocol
- many of the "small" companies are working towards these protocols.
- [companies such as IBM, DEC, Xerox to name the "small ones"]
-
- COSI Switch will provide a Level 2 User interface so users don't have
- to run on a PC/Clone, they just need a terminal! This is not the best
- way to use the network, but it is provided to allow *any* user to take
- advantage of the network, without having to buy a new computer!
-
- Disadvantages - it's coming out in Jan '88 and inspite of my extensive
- testing I know it will contain a bug of two. Remember I started the
- project in June '87.
-
- 3) NET/ROM
- NET/ROM is a $65 rom with a non-standard datagrame [TCP/IP-ish]
- protocol. I don't know much about it execpt user comments about
- problems or other things they don't like. [no verified information]
-
-
- 4) Other Protocols - There isn't much else around in the way of commonly
- accepted netwotking protocols.
-
- 5) IBM/Clones
- Yes many of us have clones and use them for development, the TCP/IP
- runs there (from KA9Q), as well as a few other machines, I hear that
- new evironments are being worked on.
- The COSI suite will be ported to many machines also, the switch will
- be running on a TNC 2/Clone first and then the DR-200 and PS-186 as well
- as the PC. We will also have a Level 3 User PAD (aka TNC code) for
- various TNC's (I'll do it for the TNC 1 if there is a C compiler for it)
- Give me a 68000 machine for 2-3 Weekends and it'll be up there too!
- Any other machines, if there is a C compiler it should be easy.
-
- 6) What does a User need?
- For TCP/IP you should have one of the machines that the KA9Q code has
- been ported to.
-
- For COSI you need WHAT YOU ARE USING RIGHT NOW! (a terminal)
- In the long run it would be good for the TNC to run level 3 (it's
- called a PAD) or to have it in the host[user] computer.
-
- I have tried to be as fair as possible and have presented the information as
- I know it. [Phil - I hope I was FAIR-ly accurate on your stuff]
-
- Bob, please question anything you want, this topic needs clearing up to OUR
- [packet radio's] users in the worst way!!
-
- 73
- tom
- --
- Thomas A. Moulton, W2VY Life is too short to be mad about things.
- Home: (201) 779-W2VY Packet: w2vy@kd6th Voice: 145.190 (r)
- Work: (201) 492-4880 x3226 FAX: (201) 493-9167
- Concurrent Computer Corp. uucp: ...!ihnp4!hotps!ka2qhd!w2vy
-
-
- 19-Nov-87 15:45:55-EST,6188;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 19 Nov 87 15:45-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA21876@EDDIE.MIT.EDU>; Thu, 19 Nov 87 13:35:18 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA21514@EDDIE.MIT.EDU>; Thu, 19 Nov 87 13:24:58 EST
- Received: by june.cs.washington.edu (5.52.1/6.11)
- id AA13072; Thu, 19 Nov 87 10:06:09 PST
- Return-Path: <bellcore!faline!karn@EDDIE.MIT.EDU>
- Message-Id: <8711191806.AA13072@june.cs.washington.edu>
- Date: 18 Nov 87 21:34:52 GMT
- From: bellcore!faline!karn@EDDIE.MIT.edu (Phil R. Karn)
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: Re: Level 3 protocols - some info?
- Summary: response to W2VY
- Keywords: TCPIP,COSI,NETROM
- References: <242@ka2qhd.UUCP> <129@tsdiag.UUCP>
-
- > 1) Is TCP/IP best?
- > There are people who do believe that this is the only networking
- > protocol that can solve everyone's needs. It does have advantages
- > some environments, I personally do not agree. The basic disadvantage
- > is that it is an all or nothing and you must have the whole protocol
- > running in your computer to access the network.
-
- TCP/IP can't solve everyone's needs, but of all things existing (or
- promised) it comes closest. TCP and IP are just two members (albeit the
- most important ones) of large and highly modular family of protocols
- more properly called the ARPA Internet Protocol Suite. Some (e.g., TCP
- and IP) are more important than others. The only "essential" protocol is
- IP. Because everyone on the Internet has to implement it, it was kept as
- simple as possible. Many of the applications are optional. The ARPA suite
- is by no means "all or nothing".
-
- > ... [note that tcp/ip is also used
- > globally, but isn't accepted by all PTT's, in fact many tcp/ip
- > connections are running on top of X.25]
- >[...]
- > Politics, many PTT's won't accept a new protocol, the COSI suite is
- > going to be following the same standards that they are seeing in the
- > commerical world, how can they say no!
- >
-
- This reflects a fundamental misunderstanding (or misrepresentation) of
- the role of TCP/IP in computer networking. TCP and IP are transport
- (layer 4) and internetwork (layer 3B) protocols. They run ON TOP of
- existing subnetworks, including, as Tom points out, X.25 networks. As
- such, TCP/IP packets are carried AS USER DATA by these subnetworks. The
- decision to use TCP/IP is made by the subscribers to the PTT data
- networks. Whether the PTTs "accept" TCP/IP is completely irrelevant.
- They have no more say in the matter than they do over what language you and
- your friends speak in your voice telephone calls.
-
- > Why X.25? I believe that to run a network on packet radio it is very
- > important to provide error recovery at all levels. [for ip it's end to
- > end, very much like good 'ole digipeating is now] Which means that if
- > any data gets lost you get notified and can re-send the information.
-
- It is almost universally believed within the computer networking
- industry that full end-to-end error control is both necessary AND
- sufficient for reliable transfer of data between computers. TCP is one
- way of doing this. Once you have such a mechanism in place, the question
- of whether additional error recovery mechanisms are also needed at the
- lower levels becomes a question of performance, not reliability. More
- often than not, such mechanisms merely result in needless complexity and
- overhead, as they can never eliminate the need for the end-to-end
- mechanism.
-
- Amateur packet radio as it now stands is one of the few areas where
- these low-level mechanisms may help. Nevertheless, it would be far
- better to prevent collisions in the first place through the the use of
- full duplex and controlled frequency reuse within the backbone network.
- Once this is done, low level mechanisms again become superfluous.
-
- Nevertheless, to quiet the critics who imply that TCP/IP somehow
- *precludes* lower level recovery mechanisms, I have added full-blown
- AX.25 Level 2 to the KA9Q Internet Package. IP datagrams may now be
- optionally acknowledged on a hop-by-hop basis. However, this technique
- suffers from exactly the same collision-aggravation problem I described
- in my recent note "The Trouble with NET/ROM". I therefore strongly
- recommend that you use this technique only as a last resort, after you
- have exhausted all other techniques to improve the raw packet loss rate.
-
- > COSI Switch will provide a Level 2 User interface so users don't have
- > to run on a PC/Clone, they just need a terminal! This is not the best
- > way to use the network, but it is provided to allow *any* user to take
- > advantage of the network, without having to buy a new computer!
- > [...]
- > For COSI you need WHAT YOU ARE USING RIGHT NOW! (a terminal)
- > In the long run it would be good for the TNC to run level 3 (it's
- > called a PAD) or to have it in the host[user] computer.
-
- This is also misleading. The reason the KA9Q Internet Protocol Package
- requires a computer to run on is that many of its high-level services
- are only meaningful on computers with file systems. How do you provide
- services like "file transfer" or "mail transfer" when all you have is a
- dumb terminal? Yes, a stripped-down version of TCP/IP and Telnet
- *could* be put on a TNC-2 for use with a dumb terminal. This would gain
- the ability to "internetwork" between an amateur packet radio channel
- and computers on Ethernets, X.25 nets, etc, also running TCP/IP, and
- this may make such a project worthwhile. However, you would gain no
- additional FUNCTIONALITY over what you can already do with a dumb
- terminal and "vanilla" AX.25. All you get is "remote terminal access",
- not true computer networking.
-
- Comparing the ARPA Suite and X.25 is comparing apples and oranges. Or,
- to use Michael Padlipsky's favorite analogy, it's like comparing a
- complete, working automobile with a bare chassis and tires. Arguing that
- X.25 is somehow "better" than the ARPA suite because you don't need a
- real computer to run it is like arguing that a bare chassis is better
- than a complete car because you can save money on body paint.
-
- Phil
-
-
- 19-Nov-87 16:01:31-EST,11202;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 19 Nov 87 16:01-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA21950@EDDIE.MIT.EDU>; Thu, 19 Nov 87 13:38:16 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA21545@EDDIE.MIT.EDU>; Thu, 19 Nov 87 13:26:07 EST
- Received: by june.cs.washington.edu (5.52.1/6.11)
- id AA13166; Thu, 19 Nov 87 10:09:55 PST
- From: ucbvax!jade!aurora!labrea!umunhum!paulf@EDDIE.MIT.edu
- Return-Path: <ucbvax!jade!aurora!labrea!umunhum!paulf@EDDIE.MIT.EDU>
- Message-Id: <8711191809.AA13166@june.cs.washington.edu>
- Date: 19 Nov 87 01:45:34 GMT
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: Re: Level 3 protocols - some info?
- Keywords: TCPIP,COSI,NETROM
- References: <242@ka2qhd.UUCP> <129@tsdiag.UUCP>
- Reply-To: paulf@umunhum.stanford.edu (Paul Flaherty)
-
- In article <129@tsdiag.UUCP> tom@tsdiag.UUCP (Tom Moulton) writes:
- >Bob, You are right it's time these questions were asked by someone who
- >has no stake in the protocol wars!!!
-
- Perhaps it would be better if they were ANSWERED by someone who has no
- stake in the protocol wars. Personally, I havn't written any code for
- either system, although I have used both sets of protocols at various times.
-
- >1) Is TCP/IP best?
- > There are people who do believe that this is the only networking
- > protocol that can solve everyone's needs. It does have advantages
- > some environments, I personally do not agree. The basic disadvantage
- > is that it is an all or nothing and you must have the whole protocol
- > running in your computer to access the network.
-
- People generally don't care who propagates a standard nearly as
- much as the care whether or not it solves a problem. This is why de facto
- standards get started; and, in a sense, the DDN protocols have become
- the de facto international protocol.
-
- I disagree with your contention that the DDN protocols are all or
- nothing. In fact, the only required protocol is IP; the user then implents
- only those layers which are useful to the solution of the problem. Take,
- for example, the gateway box that I'm using as a terminal server (at this
- very moment); it only implements IP, UDP, and BOOTP (tcp and telnet are
- added features that aren't really necessary). I'd much rather implement
- a switch based on IP and UDP (relatively simple protocols) than x.75 and
- tp4 (much more complex protocols).
-
- Also, look at the Imagen TCP implementation of tcp/ip, which is
- ROMable. If we want people to use tnc's, all we need to do is
- implement a version of TELNET in ROM for the 6809, which is both doable
- and desirable...
-
- >2) COSI Switch - What is it?
- > I am the programmer writing the switch, it is based on CCITT X.25,
- > (aka ISO 8208) which is the International standard used to connect
- > data networks throughout the world. [note that tcp/ip is also used
- > globally, but isn't accepted by all PTT's, in fact many tcp/ip
- > connections are running on top of X.25]
-
- Well, now we know your biases, anyway.
-
- You are confusing medium access protocols with internetworking and
- transport protocols. In the first place, in order to accomplish roughly
- the same goals as TCP/IP, one need x.75 AND TP4, in addition to x.25.
- Of course, one could just as easily implement TCP/IP on top of x.25, and
- this has been done in practice. Moreover, the number of TCP/IP installations
- in Europe is growing exponentially, so if the PTTs want to make money,
- they'll allow it. Finally, it's worth noting that IP gives one a much
- higher degree of internetworking flexibility than x.75, which is designed
- for x.25; IP runs just as easily with Ethernet, x.25, AppleTalk, or any of
- a myriad of data link protocols. Try doing THAT with x.75.
-
- > Why X.25? I believe that to run a network on packet radio it is very
- > important to provide error recovery at all levels. [for ip it's end to
- > end, very much like good 'ole digipeating is now] Which means that if
- > any data gets lost you get notified and can re-send the information.
-
- Your beliefs do not corroborate with the results of research done
- in this area. In particular, Karn's excellent paper on X.25 in the
- multiaccess environment shows a number of pitfalls. In particular, the
- NAK packets alone can send the network into the thrashing region. Boesch's
- thesis work here at Stanford indicates that the only real way to do multihop
- packet radio is to violate the OSI model, and establish communications
- between nonadjacent layers. Needless to say, IP does this much more
- gracefully than x.75.
-
- The real problem here is that, in a lossy multihop subnet, multi
- layer protocols don't perform terribly well; the solution is to implement
- out - of - band acknowledgements, which IP is more than capable of handling.
-
- > An X.25 network can not be an ad hoc network, it must be expanded in
- > a coordinated fashion. [remember kaos on 145.01?]
-
- Big disadvantage. As I stated in an earlier article, the amprnet
- is closer in organization to DDN than any PTT. And the DDN protocols were
- meant to connect nets run by different administrations, which will definitely
- be the case for amprnet (we don't want them funny folks from Back East telling
- us how to run OUR network...). And, as I stated before, do YOU want the job
- of coordinating all of this?
-
- > Politics, many PTT's won't accept a new protocol, the COSI suite is
- > going to be following the same standards that they are seeing in the
- > commerical world, how can they say no!
-
- They can say no if nobody is using the product. So far, the growth
- rate of TCP/IP has outstripped TP4. PTTs may be political, but they also
- like to make money (well, ok, MOST of them, anyway). Also, TCP/IP is hardly
- a new protocol; if anything, as a PTT, I'd rather have TCP/IP running, since
- its failure modes are fairly well known and documented.
-
- >
- > The ISO protocols are the next generation of networking which the
- > commerical world is going to. TCP/IP was developed on the ARPAnet
- > [a US DoD funded project] and while many companies use this protocol
- > many of the "small" companies are working towards these protocols.
- > [companies such as IBM, DEC, Xerox to name the "small ones"]
-
- The "next generation" argument smacks of innuendo. TCP is in fact
- a "next generation" protocol, having replace NTP; assuming that TP4 actually
- improves on TCP (evidence which is lacking at this point), we could implement
- changes in TCP to accomplish the same thing, and still use the existing IP
- layer and everything below it.
-
- TCP/IP was developed on the ARPA _InterNet_, which includes the
- ARPANET as a member. Not all research on the DDN protocols has been done
- with DOD funding; in particular, NFS and BOOTP were developed at Stanford
- to support our internal networking facilities.
-
- IBM, DEC and Xerox all support implementations of TCP/IP, as well
- as their own internal protocols (SNA, DECNET, and XNS). If anything, by
- promulgating their own internal architectures, they have shown a lack of
- faith in the OSI committees to produce anything practical.
-
- >
- > COSI Switch will provide a Level 2 User interface so users don't have
- > to run on a PC/Clone, they just need a terminal! This is not the best
- > way to use the network, but it is provided to allow *any* user to take
- > advantage of the network, without having to buy a new computer!
-
- Again, this is not unique. It is just as easy to write TNC code to
- do a simple TELNET implementation. Conversely, you get what you pay for.
-
- >
- > Disadvantages - it's coming out in Jan '88 and inspite of my extensive
- > testing I know it will contain a bug of two. Remember I started the
- > project in June '87.
-
- Of course, this is the disadvantage of all COSI software. Not only
- is it untried, and full of bugs, but we really don't know its failure
- modes. And, even assuming a marginal benefit from converting to the OSI
- protocols (and that's really stretching things), lack of known failure modes
- introduces unknown and unquantifiable risks into the network, which easily
- outweighs any gain.
-
- >3) NET/ROM
- > NET/ROM is a $65 rom with a non-standard datagrame [TCP/IP-ish]
- > protocol. I don't know much about it execpt user comments about
- > problems or other things they don't like. [no verified information]
-
- Odd, I thought NET/ROM implemented virtual circuits. Or at least it
- did the last time I used it. (Or, should I say, almost did...)
-
- >4) Other Protocols - There isn't much else around in the way of commonly
- > accepted netwotking protocols.
-
- Unless, of course, one counts TCP/IP, which is the de facto
- international standard.
-
- >5) IBM/Clones
- > Yes many of us have clones and use them for development, the TCP/IP
- > runs there (from KA9Q), as well as a few other machines, I hear that
- > new evironments are being worked on.
-
- TCP/IP is available for almost all pc's and mainframes; besides
- the KA9Q implementation, there are the MIT and SUMEX implementations for
- the IBM PC domain. There are several implementations for the Mac, including
- the SUMEX code, which is available on a basis similar to PCIP (CMU also has
- a MAC TCP/IP suite). The KA9Q code has been ported, or is in the process
- of being ported, to UNIX System V (Bdale Garbee et al), the Commmodore
- Amiga, laptop clones, and many others.
-
- >6) What does a User need?
- > For TCP/IP you should have one of the machines that the KA9Q code has
- > been ported to.
-
- Which covers most of the market, except for the games machines...
-
- > For COSI you need WHAT YOU ARE USING RIGHT NOW! (a terminal)
- > In the long run it would be good for the TNC to run level 3 (it's
- > called a PAD) or to have it in the host[user] computer.
-
- Like I said, this is not unique. "Given a few weeks of free time",
- I could have telnet up and running on a TNC. Unfortunately, my free time
- comes in minutes, not weeks...
-
- >I have tried to be as fair as possible and have presented the information as
- >I know it. [Phil - I hope I was FAIR-ly accurate on your stuff]
-
- Well, I don't think that you were unfair, but I think that it is clear
- that you were inaccurate, but that's probably not your fault anyway.
-
- To reiterate my arguments from my last posting, the DDN protocols
- have several clear and evident advantages over the OSI suite. Some of these
- are related to the fact that TCP/IP has been around awhile (good routing,
- lots of user layer protocols, and availability), but most are intrinsic
- (strong internetworking, layering flexibility, subnetting and distributed
- management, tremendous reliability and throughput).
-
- Given all of this, why make such a major effort to convert to
- a protocol suite which offers no additional user funtionality?
-
- Still not convinced? Well, just remember, when Phase IV flies,
- the only cost free long distance network in the country will be running
- TCP/IP...
-
-
-
-
-
-
-
- -=Paul Flaherty, N9FZX | "One Internet to rule them all,
- Computer Systems Laboratory | One Internet to find them,
- Stanford University | One Internet to bring them all
- Domain: paulf@shasta.Stanford.EDU| and in the ether bind them." -ToIH
-
-
- 19-Nov-87 16:18:53-EST,1162;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 19 Nov 87 16:18-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA24560@EDDIE.MIT.EDU>; Thu, 19 Nov 87 15:05:46 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA24547@EDDIE.MIT.EDU>; Thu, 19 Nov 87 15:05:25 EST
- Received: from huey.udel.edu by Louie.UDEL.EDU id ab01625; 19 Nov 87 13:48 EST
- Date: Thu, 19 Nov 87 13:44:49 EST
- From: Mills@UDEL.EDU
- To: "Phil R. Karn" <bellcore!faline!karn@eddie.mit.edu>
- Cc: PACKET-RADIO@eddie.mit.edu
- Subject: Re: Repeater advice anyone?
- Message-Id: <8711191344.aa02654@Huey.UDEL.EDU>
-
- Phil,
-
- DARPA packet radios use a technique called pacing, in which the onward
- transmission of a packet from the digipeater to the next hop is heard
- by the original transmitter as the ACK for its transmission. Since the
- digipeater will delay until the channel is quiet, this forms an automatic
- sensing mechanism for hidden transmitters and without the channel overhead
- of additional ACKs. The sender, of course, should delay once his relayed
- transmission is heard to avoid collision on the next hop, just as you
- describe.
-
- Dave
- 22-Nov-87 22:25:09-EST,2495;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 22 Nov 87 22:25-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA18425@EDDIE.MIT.EDU>; Sun, 22 Nov 87 21:08:06 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA18419@EDDIE.MIT.EDU>; Sun, 22 Nov 87 21:07:46 EST
- Received: by june.cs.washington.edu (5.52.1/6.11)
- id AA02759; Sun, 22 Nov 87 18:08:37 PST
- Return-Path: <princeton!idacrd!mac@eddie.mit.edu>
- Message-Id: <8711230208.AA02759@june.cs.washington.edu>
- Date: 20 Nov 87 15:25:30 GMT
- From: princeton!idacrd!mac@eddie.mit.edu (Bob McGwier)
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: protowars
-
- I would like to provide a quote for the nets perusal. I am quoting
- AA4RE, Roy Engehausen's, column from PSR, "In the Mailbox".
-
- Food for Thought - One Man's Opinion
-
- Competition has been with packet radio a long time. I think it all started
- with the VADCG versus AX25 controversy. We now have all sorts of things
- bouncing around in the packet world. Zip code versus area code; NETROM versus
- COSI versus TCP/IP versus TEXNET; W0RLI CBBS versus WA7MBL versus KA2BQE
- mailboxes; etc. etc.
-
- (Editorial comment: Roy either doesn't understand internetworking or is
- writing for the "unwashed masses" as IP will run on top of ALL these
- offerings with varying degrees of difficulty depending on whether the
- internal structure is datagrammes or not)
-
- Competition is good. Unfortunately, the real world does not have the nice
- black and white decisions we would like. Each competiting system has
- advantages and disadvantages that have to be carefully examined, and, in
- most cases tried. Opinions have to be heard and the facts weighed.
-
- The result is usually a good (if not right) decision. Most packeteers can
- agree today that AX25 was the better way to go initially so the system does
- work (sic).
-
- The trick to competition is timing. At some point, a majority of packeteers
- will be going in the same direction. The proponents of the other alternatives
- have a choice: they can either abandon their approach and help improve packet
- for all of us or they can keep sapping the strength of amateur radio by
- spending resources and energy on a "dead end".
-
- If you find your self North while everyone else is going South, it may be
- time to say "Ah, What the Heck" and join the crowd.
-
- 73, Roy AA4RE
-
-
-
- I thought this might be appropriate at this juncture in the protocol
- wars to give it a wider audience.
-
- Bob N4HY
-
-
- 22-Nov-87 22:31:06-EST,9713;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 22 Nov 87 22:31-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA18488@EDDIE.MIT.EDU>; Sun, 22 Nov 87 21:13:31 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA18483@EDDIE.MIT.EDU>; Sun, 22 Nov 87 21:13:05 EST
- Received: by june.cs.washington.edu (5.52.1/6.11)
- id AA03078; Sun, 22 Nov 87 18:13:56 PST
- Return-Path: <delni.dec.com!goldstein@decwrl.dec.com>
- Message-Id: <8711230213.AA03078@june.cs.washington.edu>
- Date: 20 Nov 87 18:50:00 GMT
- From: delni.dec.com!goldstein@DECWRL.DEC.COM (Fred R. Goldstein dtn226-7388)
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: Let's calm down the CONS/CLNS flames just a bit?
-
- The recent go-round by Tom Moulton W2VY and Paul Flaherty N9FZX was fun
- to read but seems a bit too flamish... let me just correct some
- misunderstandings based upon my (agnostic) rather finite knowledge.
-
- VY> 1) Is TCP/IP best?
- VY> The basic disadvantage
- VY> is that it is an all or nothing and you must have the whole protocol
- VY> running in your computer to access the network.
-
- FZX> I disagree with your contention that the DDN protocols are all or
- FZX> nothing. In fact, the only required protocol is IP; the user then
- FZX> implents only those layers which are useful to the solution of the
- FZX> problem.
-
- Paul is closer to the mark than Tom, but the implementations,
- not protocols, are what make it so. Both stacks (COSI and TCP/IP)
- require L3 to be implemented in the ISs (routers/gateways); the KA9Q
- implementation of IP puts the PAD function (TELNET) in the host only,
- while Tom's COSI will provide an AX.25L2 to COSI gateway PAD. The
- TCP and other higher-layers don't have to go in routers anyway.
-
- VY> 2) COSI Switch - What is it?
- VY> I am the programmer writing the switch, it is based on CCITT X.25,
- VY> (aka ISO 8208) which is the International standard used to connect
- VY> data networks throughout the world. [note that tcp/ip is also used
- VY> globally, but isn't accepted by all PTT's, in fact many tcp/ip
- VY> connections are running on top of X.25]
-
- VY> Politics, many PTT's won't accept a new protocol, the COSI suite is
- VY> going to be following the same standards that they are seeing in the
- VY> commerical world, how can they say no!
- VY>
- VY> The ISO protocols are the next generation of networking which the
- VY> commerical world is going to. TCP/IP was developed on the ARPAnet
- VY> [a US DoD funded project] and while many companies use this protocol
- VY> many of the "small" companies are working towards these protocols.
- VY> [companies such as IBM, DEC, Xerox to name the "small ones"]
-
- This isn't exactly fair about OSI. The PTT's adopted X.25
- in 1976 based on their "connection" model of telephony applied to data.
- They know how to bill for connections! OSI has adopted both connection-
- oriented (ISO 8208) and connectionless (8473) network layers. The
- former is really X.25-1984; the latter is new to OSI.
-
- And while many companies are adopting OSI, it doesn't mean we're
- throwing out our other protocols. OSI is open systems
- _interconnection_, meaning a common means to link multi-vendor nets
- together. DECnet isn't going away even after it adopts OSI (8473) as
- the basis of its network layer. This doesn't discredit OSI per se;
- it just means that we don't speak "Church Latin" (or Esperanto, if you
- prefer that analogy) to members of our own community, even if it's the
- common language we share with other communities.
-
- FZX> People generally don't care who propagates a standard nearly as
- FZX> much as the care whether or not it solves a problem. This is why de facto
- FZX> standards get started; and, in a sense, the DDN protocols have become
- FZX> the de facto international protocol.
-
- True enough at the moment; DDN Suite provides the most popular network
- functions on a public domain multi-vendor basis. It does provide an
- open systems interconnection (lower case) function! ISO OSI is a more
- ambitious project of which X.25-1984 is a small piece.
-
- FZX> You are confusing medium access protocols with internetworking and
- FZX> transport protocols. In the first place, in order to accomplish roughly
- FZX> the same goals as TCP/IP, one need x.75 AND TP4, in addition to x.25.
- FZX> Of course, one could just as easily implement TCP/IP on top of x.25, and
- FZX> this has been done in practice. Moreover, the number of TCP/IP
- FZX> installations
- FZX> in Europe is growing exponentially, so if the PTTs want to make money,
- FZX> they'll allow it. Finally, it's worth noting that IP gives one a much
- FZX> higher degree of internetworking flexibility than x.75, which is designed
- FZX> for x.25; IP runs just as easily with Ethernet, x.25, AppleTalk, or any of
- FZX> a myriad of data link protocols. Try doing THAT with x.75.
-
- Let's not get TP4 confused with TP0-3. X.75 is the IS-IS
- protocol that goes with X.25. ISO IP (8473) does pretty much what DDN
- IP does. When you run either datagramme network layer, you need a good
- transport to fix it up; TP4 and TCP are quite similar in function though
- the headers parse quite differently. If you like TCP you shouldn't
- dislike TP4, though you might not like TP0 (the "do nothing" Transport)!
-
- The hard-core CONS fans in the PTT world prefer lower-class TPs,
- which require a "perfect" virtual circuit in order to operate. TP4 is
- the favorite of datagramme fans _and_ those who don't trust in the perfect
- reliability of packet switch implementations.
-
- VY> Politics, many PTT's won't accept a new protocol, the COSI suite is
- VY> going to be following the same standards that they are seeing in the
- VY> commerical world, how can they say no!
-
- FZX> They can say no if nobody is using the product. So far, the growth
- FZX> rate of TCP/IP has outstripped TP4. PTTs may be political, but they also
- FZX> like to make money (well, ok, MOST of them, anyway). Also, TCP/IP is hardly
- FZX> a new protocol; if anything, as a PTT, I'd rather have TCP/IP running, since
- FZX> its failure modes are fairly well known and documented.
-
- VY> The ISO protocols are the next generation of networking which the
- VY> commerical world is going to. TCP/IP was developed on the ARPAnet
- VY> [a US DoD funded project] and while many companies use this protocol
- VY> many of the "small" companies are working towards these protocols.
- VY> [companies such as IBM, DEC, Xerox to name the "small ones"]
-
- FZX> The "next generation" argument smacks of innuendo. TCP is in fact
- FZX>a "next generation" protocol, having replace NTP; assuming that TP4 actually
- FZX>improves on TCP (evidence which is lacking at this point), we could implement
- FZX> changes in TCP to accomplish the same thing, and still use the existing IP
- FZX> layer and everything below it.
-
- Better, you could use 8473 (ISO IP) and maybe run TCP et al
- above it, since IP's addressing is ill-suited to AMPRnet. Again TP4 is
- not the issue since _it_ runs on top of most anything; TP classes 0 to 3 don't.
-
- FZX> TCP/IP was developed on the ARPA _InterNet_, which includes the
- FZX> ARPANET as a member. Not all research on the DDN protocols has been done
- FZX> with DOD funding; in particular, NFS and BOOTP were developed at Stanford
- FZX> to support our internal networking facilities.
-
- FZX> IBM, DEC and Xerox all support implementations of TCP/IP, as well
- FZX> as their own internal protocols (SNA, DECNET, and XNS). If anything, by
- FZX> promulgating their own internal architectures, they have shown a lack of
- FZX> faith in the OSI committees to produce anything practical.
-
- Horse hockey. OSI is eminently practical for inter-vendor
- communication. SNA wouldn't be too efficient on VAXen nor would DECnet
- on 370s. (Though gateways exist in both directions.) OSI will allow
- both to communicate on neutral terms. Maybe TCP/IP would too, but that's
- not for me to say.
-
- FZX> And, even assuming a marginal benefit from converting to the OSI
- FZX> protocols (and that's really stretching things), lack of known failure modes
- FZX> introduces unknown and unquantifiable risks into the network, which easily
- FZX> outweighs any gain.
-
- Again confusing implementations with protocol. There are bugs
- in Phil's TCP/IP code too, and he continues to work on them. Tom's code
- is new and of course buggy too -- what code isn't? Star wars? (HI)
- Implementations aren't the key to protocol beauty contests like this
- one.
-
- FZX> Odd, I thought NET/ROM implemented virtual circuits. Or at least it
- FZX> did the last time I used it. (Or, should I say, almost did...)
-
- NET/ROM uses datagrammes within its private subnet and VCs to
- the outside world. So there. (Telenet did this once, before moving its
- innards to X.75 since that matched up better with its X.25 outards.)
- Best or worst of both worlds if you're so inclined, but expedient.
-
- FZX> To reiterate my arguments from my last posting, the DDN protocols
- FZX> have several clear and evident advantages over the OSI suite. Some of these
- FZX> are related to the fact that TCP/IP has been around awhile (good routing,
- FZX> lots of user layer protocols, and availability), but most are intrinsic
- FZX> (strong internetworking, layering flexibility, subnetting and distributed
- FZX> management, tremendous reliability and throughput).
-
- The OSI suite is newer but not intrinsically that different,
- since it includes connectionless as well as X.25-style VC networking.
- Remember that in standardsland, it's easier to write more standards to
- include everybody than to pick a single solution.
- fred k1io
-
-
- 22-Nov-87 22:42:15-EST,2398;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 22 Nov 87 22:42-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA18814@EDDIE.MIT.EDU>; Sun, 22 Nov 87 21:38:28 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA18810@EDDIE.MIT.EDU>; Sun, 22 Nov 87 21:38:15 EST
- Received: by june.cs.washington.edu (5.52.1/6.11)
- id AA03741; Sun, 22 Nov 87 18:39:12 PST
- Return-Path: <pyramid!voder!apple!winter@eddie.mit.edu>
- Message-Id: <8711230239.AA03741@june.cs.washington.edu>
- Date: 21 Nov 87 00:00:01 GMT
- From: pyramid!voder!apple!winter@eddie.mit.edu (Patty Winter)
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: Re: Packet freqs
- Summary: Would packet control of a repeater really be secure?
- References: <698@uop.UUCP> <1532@faline.bellcore.com> <250@splut.UUCP> <1548@faline.bellcore.com>
-
- In article <1548@faline.bellcore.com> karn@faline.bellcore.com (Phil R. Karn)
- writes:
- > [discussion of amateur spectrum above 220.5 MHz]
- >Which gives me an idea: many repeaters have separate, dedicated control
- >links mainly to satisfy FCC requirements. They can and do have secondary
- >command decoders on their regular inputs, and most routine commanding is
- >done this way. The backup control requirement for the voice repeater
- >could instead be provided as a secondary function of a co-located packet
- >switch. In exchange for providing secure and more sophisticated control
- >functions, the packet guys could get an existing antenna and frequency
- >assignment at a good site, and everybody wins.
-
- Hmmm, intriguing idea. But I don't see how such control would be
- more secure. Control codes flying by on a packet channel would
- require less (read: none) effort to view than would the DTMF tones
- normally used. I don't know how many packeteers leave MON ON
- nowadays with all the activity going on, but it wouldn't be
- difficult to set your TNC to grab only those messages going from
- the repeater's control operators.
-
- And even if the control codes were encrypted and then translated
- at the repeater controller (would that be legal?), I don't expect
- it would take a diehard hacker long to crack them. Right, Phil? :-)
-
-
- Patty
-
- --
- Patty Winter N6BIS (408) 973-2814
- M/S 2C, Apple Computer, Inc., 20525 Mariani Ave., Cupertino, CA 95014
- {decwrl,nsc,sun,dual}!apple!winter
-
-
- 22-Nov-87 22:54:58-EST,1908;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 22 Nov 87 22:54-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA18953@EDDIE.MIT.EDU>; Sun, 22 Nov 87 21:50:36 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA18948@EDDIE.MIT.EDU>; Sun, 22 Nov 87 21:50:18 EST
- Received: by june.cs.washington.edu (5.52.1/6.11)
- id AA04094; Sun, 22 Nov 87 18:51:12 PST
- Return-Path: <ulysses!faline!karn@eddie.mit.edu>
- Message-Id: <8711230251.AA04094@june.cs.washington.edu>
- Date: 22 Nov 87 01:49:01 GMT
- From: ulysses!faline!karn@EDDIE.MIT.edu (Phil R. Karn)
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: Re: Packet freqs
- Summary: packet security is quite possible
- Keywords: repeater control, security
- References: <698@uop.UUCP> <1532@faline.bellcore.com> <250@splut.UUCP> <6791@apple.UUCP>
-
- > Hmmm, intriguing idea. But I don't see how such control would be
- > more secure.
-
- No problem. See my paper "Another Look at Authentication" in the 6th
- ARRL Computer Networking conference proceedings.
-
- Basically, the idea is to use DES encryption on the packet. The original
- packet is sent in the clear, but the DES is used to compute a "cipher
- checksum" that is sent along with the packet. By running the same
- algorithm and comparing the results, the receiver can detect bogus or
- corrupted packets. This technique is in widespread use in the banking
- industry; the detection of forged EFT orders is much more important than
- keeping legitimate orders secret.
-
- For those of you familiar with DES, the exact technique is to encrypt
- the packet with DES in the Cipher Block Chaining mode, and then transmit
- the original plaintext packet along with the last cipherword.
-
- This does not violate the FCC restriction on codes and ciphers because no
- information is being hidden; the "ciphertext" (such as it is) is sent along
- with the plaintext that produced it.
-
- Phil
-
-
- 22-Nov-87 22:55:10-EST,2400;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 22 Nov 87 22:55-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA18930@EDDIE.MIT.EDU>; Sun, 22 Nov 87 21:48:00 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA18925@EDDIE.MIT.EDU>; Sun, 22 Nov 87 21:47:47 EST
- Received: by june.cs.washington.edu (5.52.1/6.11)
- id AA04011; Sun, 22 Nov 87 18:48:40 PST
- From: ihnp4!homxb!whuts!mtune!petsd!pedsgo!pedsga!tsdiag!tom@EDDIE.MIT.edu
- Return-Path: <ihnp4!homxb!whuts!mtune!petsd!pedsgo!pedsga!tsdiag!tom@eddie.mit.edu>
- Message-Id: <8711230248.AA04011@june.cs.washington.edu>
- Date: 21 Nov 87 17:27:12 GMT
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: Re: Level 3 protocols - some info?
- Summary: Sorry Bob some people don't read directions.
- References: <242@ka2qhd.UUCP> <129@tsdiag.UUCP> <1556@faline.bellcore.com> <103@umunhum.STANFORD.EDU>
-
- Bob, KC2WZ,
- I hope you noted how your request did not get any reaction until there
- was someone to jump all over. I personally am trying to have fun with packet
- after all it's a hobby, and I really don't have time for this crap.
-
- I am tired of this bickering back and forth about the exact words i use in my
- messages, it seems some people don't understand that by attacking my words and
- ideas they are not going to get anywhere, I'm going to do what I believe is
- best for the network.
-
- I feel that the NETWORK I am working on building will be best to serve our needs.
-
- I am also sorry that this group did not see fit to honor your request to avoid
- inflammatory remarks, in my comment I may not have succeded 100% but I tried
- and do think I was fair in my remarks.
-
- We all have two things in common, 1) We are building Alternatives for our USERS
- and 2) We are following the path we feel is best.
-
- 73 and remember IT'S ONLY A HOBBY
- Tom, W2VY
-
- Thomas A. Moulton, W2VY Life is too short to be mad about things.
- Home: (201) 779-W2VY Packet: w2vy@kd6th Voice: 145.190 (r)
- Work: (201) 492-4880 x3226 FAX: (201) 493-9167
- Concurrent Computer Corp. uucp: ...!ihnp4!hotps!ka2qhd!w2vy
- --
- Thomas A. Moulton, W2VY Life is too short to be mad about things.
- Home: (201) 779-W2VY Packet: w2vy@kd6th Voice: 145.190 (r)
- Work: (201) 492-4880 x3226 FAX: (201) 493-9167
- Concurrent Computer Corp. uucp: ...!ihnp4!hotps!ka2qhd!w2vy
-
-
- 22-Nov-87 22:55:57-EST,6809;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 22 Nov 87 22:55-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA18836@EDDIE.MIT.EDU>; Sun, 22 Nov 87 21:41:23 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA18830@EDDIE.MIT.EDU>; Sun, 22 Nov 87 21:41:05 EST
- Received: by june.cs.washington.edu (5.52.1/6.11)
- id AA03782; Sun, 22 Nov 87 18:41:56 PST
- Return-Path: <umunhum!paulf@umunhum.stanford.edu>
- Message-Id: <8711230241.AA03782@june.cs.washington.edu>
- Date: 21 Nov 87 04:29:57 GMT
- From: umunhum!paulf@umunhum.stanford.edu (paulf)
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: %&^#*$BOOM&^%&$
- Reply-To: paulf@umunhum.STANFORD.edu (Paul Flaherty)
- Distribution: world
-
-
- >True enough at the moment; DDN Suite provides the most popular network
- >functions on a public domain multi-vendor basis. It does provide an
- >open systems interconnection (lower case) function! ISO OSI is a more
- >ambitious project of which X.25-1984 is a small piece.
-
- The key operative here is "public domain". Most of the TCP/IP
- implementations running today were hacked out on the govenment dole, and
- are henceforth free. Some individuals have even been known to use all of
- their "recreation time" to do this.
-
- This, of course, is entirely unacceptable to certain vendors, who
- woudl stand to make tons of money selling networking software, if the
- public domain stuff wasn't there. So, don't be suprised to see them
- hawking their wares here.
-
- I strongly disagree with your "more ambitious" labelling of
- the OSI protocols. The degree of difference in offerred services is
- slight, yet large enough to make the two protocol sets incompatible.
-
- Frankly, all of this reminds me of the US T1 vs. CCITT T1 battles;
- the US poured tons of money to develop the concept, and the CCITT made
- subtle changes, so that the US couldn't sell its technology overseas.
-
- > Let's not get TP4 confused with TP0-3. X.75 is the IS-IS
- >protocol that goes with X.25. ISO IP (8473) does pretty much what DDN
- >IP does. When you run either datagramme network layer, you need a good
- >transport to fix it up; TP4 and TCP are quite similar in function though
- >the headers parse quite differently. If you like TCP you shouldn't
- >dislike TP4, though you might not like TP0 (the "do nothing" Transport)!
-
- Okay, so if they're so similar, why waste money and effort?
- I know, I know, because it's an INTERNATIONAL STANDARD. Ptooey. I like
- TCP because it has fewer internal states than TP4, therefore it will run
- faster almost universally. And, speaking of fast, what about UDP?
- Incidentally, I'm assuming x.75, since that is the hack that belongs with
- x.25. I'd rather see otherwise...
-
- > Better, you could use 8473 (ISO IP) and maybe run TCP et al
- >above it, since IP's addressing is ill-suited to AMPRnet. Again TP4 is
- >not the issue since _it_ runs on top of most anything; TP classes 0 to 3 don't.
-
- Okay, I'll bite. Why is IP addressing unsuitable to the ampr?
- Most of the problems which have been brought up (like location routing) could
- be solved via communicating gateways. The problems that have been mentioned
- are not unknown to research, and have been discussed before in many different
- contexts.
-
- >FZX> IBM, DEC and Xerox all support implementations of TCP/IP, as well
- >FZX> as their own internal protocols (SNA, DECNET, and XNS). If anything, by
- >FZX> promulgating their own internal architectures, they have shown a lack of
- >FZX> faith in the OSI committees to produce anything practical.
- > Horse hockey. OSI is eminently practical for inter-vendor
- >communication. SNA wouldn't be too efficient on VAXen nor would DECnet
- >on 370s. (Though gateways exist in both directions.) OSI will allow
- >both to communicate on neutral terms. Maybe TCP/IP would too, but that's
- >not for me to say.
-
- Of course TCP/IP isn't practical for inter - vendor communication,
- because they can't make any money off of it.
-
- TCP/IP is the most practical way (at the moment, and otherwise) to
- do inter - vendor machine linking. And with NFS (which doesn't exist for
- OSI), that linking becomes even more practical and useful.
-
- > Again confusing implementations with protocol. There are bugs
- >in Phil's TCP/IP code too, and he continues to work on them. Tom's code
- >is new and of course buggy too -- what code isn't? Star wars? (HI)
- >Implementations aren't the key to protocol beauty contests like this
- >one.
-
- Wrong. Certain features of protocols can make them immune to
- software faults; conversely, some features make protocols error prone.
- In particular, DDN subnetting isolates faults, and bogus packets (thank
- you, oh Martian Filters). TP4 has a greater number of internal states,
- which increases the complexity of the code (or state machine) required
- to implement it, ergo more bugs.
-
- This is NOT a beauty contest. We are trying to make a serious
- comparison of two protocol families, before we lock ourselves into one.
- That way, we can hopefully avoid expensive mistakes like AX.25.
-
- >FZX> Odd, I thought NET/ROM implemented virtual circuits. Or at least it
- >FZX> did the last time I used it. (Or, should I say, almost did...)
- > NET/ROM uses datagrammes within its private subnet and VCs to
- >the outside world. So there. (Telenet did this once, before moving its
- >innards to X.75 since that matched up better with its X.25 outards.)
- >Best or worst of both worlds if you're so inclined, but expedient.
-
- Like I said, NET/ROM implements virtual circuits (out of datagrams).
- Of course, designing a network on TOP of this is uneconomical, since you
- can't take advantage of the economy of UDP packets. (Yeah, I know about
- x.25 datagrams...nice afterthought.)
-
- > The OSI suite is newer but not intrinsically that different,
- >since it includes connectionless as well as X.25-style VC networking.
- >Remember that in standardsland, it's easier to write more standards to
- >include everybody than to pick a single solution.
- > fred k1io
-
- "Newer". "International". "Advanced". Bull.
-
- The de facto international open systems interconnect is TCP/IP.
- The burden of proof clearly lies with those in COSI - land. Show me
- a significant advantage. Show me evidence of growth. Show me freeware.
- Show me a working implentation. Show me Network Virtual Disk. Show me
- what's wrong with the DDN protocols, and the weight of research.
- And, most of all, show me that CCITT isn't screwing the pooch.
-
-
-
-
-
- -=Paul Flaherty, N9FZX | "One Internet to rule them all,
- Computer Systems Laboratory | One Internet to find them,
- Stanford University | One Internet to bring them all
- Domain: paulf@shasta.Stanford.EDU| and in the ether bind them." -ToIH
-
-
- 22-Nov-87 23:10:31-EST,1886;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 22 Nov 87 23:10-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA18844@EDDIE.MIT.EDU>; Sun, 22 Nov 87 21:43:31 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA18840@EDDIE.MIT.EDU>; Sun, 22 Nov 87 21:43:02 EST
- Received: by june.cs.washington.edu (5.52.1/6.11)
- id AA03840; Sun, 22 Nov 87 18:43:54 PST
- Return-Path: <Shasta!kaufman@shasta.stanford.edu>
- Message-Id: <8711230243.AA03840@june.cs.washington.edu>
- Date: 21 Nov 87 17:00:06 GMT
- From: Shasta!kaufman@shasta.stanford.edu (Marc Kaufman)
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: Re: Packet freqs
- Keywords: repeater control, security
- References: <698@uop.UUCP> <1532@faline.bellcore.com> <250@splut.UUCP> <1548@faline.bellcore.com> <6791@apple.UUCP>
-
- In article <6791@apple.UUCP> winter@apple.UUCP (Patty Winter) writes:
-
- >Hmmm, intriguing idea. But I don't see how such control would be
- >more secure. Control codes flying by on a packet channel would
- >require less (read: none) effort to view than would the DTMF tones
- >normally used. I don't know how many packeteers leave MON ON
- >nowadays with all the activity going on, but it wouldn't be
- >difficult to set your TNC to grab only those messages going from
- >the repeater's control operators.
-
- Well, my station has an Apple II with a tone decoder chip, that I can
- use to listen to DTMF tones... no problem. Besides, most people can
- remember tones well enough, for a few minutes, to match them with a telephone.
-
- I submit that one is not more secure intrinsicly than another, and that
- if the packet message included control characters, or those with bit 8 set,
- it would be just as secure as DTMF. For real security, you could implement
- a challenge/response algorithm with time varying keys.
-
- Marc Kaufman, WB6ECE (kaufman@Shasta.stanford.edu)
-
-
- 23-Nov-87 15:05:32-EST,540;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 23 Nov 87 15:05-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA02779@EDDIE.MIT.EDU>; Mon, 23 Nov 87 13:47:42 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA02771@EDDIE.MIT.EDU>; Mon, 23 Nov 87 13:47:27 EST
- Date: Mon 23 Nov 87 10:47:35-PST
- From: Allen Takahashi <TAKAHASHI@KL.SRI.COM>
- Subject: Removal
- To: packet-radio@EDDIE.MIT.EDU
- Message-Id: <12352954839.28.TAKAHASHI@KL.SRI.COM>
-
- Please remove me from the packet-radio list.
- -------
- 23-Nov-87 17:28:55-EST,540;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 23 Nov 87 17:28-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA02779@EDDIE.MIT.EDU>; Mon, 23 Nov 87 13:47:42 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA02771@EDDIE.MIT.EDU>; Mon, 23 Nov 87 13:47:27 EST
- Date: Mon 23 Nov 87 10:47:35-PST
- From: Allen Takahashi <TAKAHASHI@KL.SRI.COM>
- Subject: Removal
- To: packet-radio@EDDIE.MIT.EDU
- Message-Id: <12352954839.28.TAKAHASHI@KL.SRI.COM>
-
- Please remove me from the packet-radio list.
- -------
- 24-Nov-87 04:06:26-EST,774;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 24 Nov 87 04:06-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA20691@EDDIE.MIT.EDU>; Tue, 24 Nov 87 02:58:28 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA20682@EDDIE.MIT.EDU>; Tue, 24 Nov 87 02:58:14 EST
- Message-Id: <8711240758.AA20682@EDDIE.MIT.EDU>
- Received: from VMS.CIS.PITTSBURGH.EDU by VB.CC.CMU.EDU; Tue, 24 Nov 87 02:58 EST
- Date: Tue, 24 Nov 87 02:59 EDT
- From: Ray Wong <UCONRAY%Vms.Cis.Pittsburgh.Edu@VB.CC.CMU.EDU>
- Subject: RE: Delete me for the list
- To: packet-radio@EDDIE.MIT.EDU
- X-Vms-To: IN%"packet-radio@eddie.mit.EDU"
-
-
-
- Yes, please delete the user 000544@pittvms from the list since this
- person no long is at the university.
-
- Thank you,
-
- Ray
- 24-Nov-87 07:49:36-EST,716;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 24 Nov 87 07:49-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA23888@EDDIE.MIT.EDU>; Tue, 24 Nov 87 06:52:15 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA23870@EDDIE.MIT.EDU>; Tue, 24 Nov 87 06:51:38 EST
- Received: by trout.nosc.mil (5.58/1.27)
- id AA10279; Tue, 24 Nov 87 03:53:10 PST
- Date: Tue, 24 Nov 87 03:53:10 PST
- From: maziarz@trout.nosc.mil (Donald R. Maziarz)
- Message-Id: <8711241153.AA10279@trout.nosc.mil>
- To: packet-radio@eddie.mit.edu
- Cc: johnb@trout.nosc.mil, beeler@trout.nosc.mil
- Subject: remove me
-
- -------
- Please remove me from the packet-radio mail list. Its been interesting.
- -------
-
- 25-Nov-87 14:32:08-EST,5430;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 25 Nov 87 14:32-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA28061@EDDIE.MIT.EDU>; Wed, 25 Nov 87 12:54:40 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA26466@EDDIE.MIT.EDU>; Wed, 25 Nov 87 12:12:34 EST
- Message-Id: <8711251712.AA26466@EDDIE.MIT.EDU>
- To: packet-radio@EDDIE.MIT.EDU
- Cc: clements@LF-SERVER-2.BBN.COM
- Subject: PBBSes and Philosophy
- Date: Wed, 25 Nov 87 12:12:53 -0500
- From: clements@LF-SERVER-2.BBN.COM
-
- >Date: 19 Nov 87 13:34:22 GMT
- >From n2dsy-4!hotps!hou2d!n2dsy Wed Nov 18 12:17:29 1987 remote from hotps
- >
- >Bob,
- > Please forward this to W0RLI.
- >
- > J. Gordon Beattie, Jr., N2DSY
-
- Hank can be reached as W0RLI @ W0RLI, or at his postal address
- which is published with his widely distributed source code. :-)
-
-
- >REGARDING these comments>>>>>
- >>
- >> Hank did mention that he HAS become ticked off at two people, though,
- >> one JA op and one 2-lander named Brian something, who have
- >> taken Hank's code, modified it slightly and claimed it as their own,
- >>
- >> ...
-
- The above "..." deletes the main point of Hank's comment, about
- COPYRIGHTING the modified code.
-
- >>
- >> 73, Bob, K1BC clements@bbn.com
- >>
-
- >[From Brian, KA2BQE]
- > Hank and/or Bob,
- >
- > The code in question started with a pre-release of
- > W0RLI's earliest posted version (on CompuServe) and at this
- > point contains less than 10% of the original code.
-
- [ Long list of changes and additions deleted here. ]
-
- Well, Hank's code has certainly improved and changed a lot since
- then, too, including some of the items listed here. Things he didn't
- do were mostly in the area that your list includes, improving the
- user and sysop interfaces. I agree, those drastically need
- improvement in Hank's version.
-
- > In addition, during the
- > first three months, I sent at least 25 bug fixes to the RLI
- > team, most of which were NEVER acknowledged...
-
- I don't know what to say about this. My fixes and additions
- were acknowledged, and usually included in the next version.
-
- > Last but not least, I would very much appreciate your thorough
- > examination of the code I will send you. This should prove my
- > contentions.
-
- Thank you for the offer, but, as stated in my previous message, I
- have retired from the PBBS business. I was the second 820 PBBS on
- the air (after W0RLI) back in the dark ages, and I retired my
- 8088 systems this summer.
-
- > I would ask you, after such examination, to
- > consider sending a retraction/apology onto the NETWORKS as
- > your initial statement which was widely circulated did much to
- > impugn my reputation and the 14 months of intensive work that
- > I have done.
- >
- > Sincerely,
- > Brian Riley, KA2BQE @ KA2BQE
- > 609-268-9497
-
- I have already posted a partial retraction on the word
- "slightly". I posted it to the same list as the first message.
- If you didn't see it, I'm sorry. I said I had been repeating my
- impression of what Hank said, and that he may not have used the
- word "slightly". I said I did not personally know the extent of
- the changes.
-
- But here's the point:
-
- The world benefits by the free exchange of ideas. In the packet
- environment, that includes software.
-
- Hank's PBBS code, both Xerox 820 based and MS-DOS based, has been
- freely distributed, in both source and binary, since day one.
- It has benefited from the work of many others who have provided
- their efforts just as freely for the common good.
-
- KA9Q's TCP/IP code has been spreading the same way. See the book
- "Hackers" by Steve Levy (in which I am mentioned briefly) for a
- clear statement of this philosophy.
-
- Consider as a counterexample the hoarding of source code for the
- TNC-1 which held us all back for years. The TNC-2 code is also
- being kept secret. If sources had been available, we wouldn't
- have all that kludgey code dealing with "*** CONNECT REQUEST ***
- messages and XON/XOFF handling. We could have fixed the TNCs.
-
- WA7MBL implemented reverse-forwarding a long time ago. I don't
- know whether he was first, but he was the first that Hank or I
- saw. And he keeps his code secret and didn't respond to requests
- for his algorithm so Hank could put it in his code compatibly.
- So W0RLI PBBSes didn't get reverse forwarding until recently.
-
- This message from Brian is the first I have seen offering a copy
- of his sources. Maybe he has made the offer before, but I
- haven't seen it. They never made it to any site in the New
- England area THAT I KNOW OF during the development stages.
-
- Contrast that with the W0RLI and KA9Q systems which are always
- released with sources. They are posted on Compuserve, on
- SIMTEL-20 on the ARPANET, and on many dial-up BBSes. And this
- benefits us all. That difference in philosophy is what Hank was
- objecting to, especially when the undistributed code had been
- based on his freely distributed code. And I agree with him.
-
- I do not intend to get into a "My PBBS is better than yours" war.
- Or any other kind of war. I'm sorry the message I posted before
- got people bent out of shape. I'm sorry Brian missed the point
- of the message (and deleted it from his followup).
-
- Let's all try to cooperate a little more, people.
-
- 73, Bob, K1BC clements@bbn.com "Good in the callbook"
- 25-Nov-87 16:17:22-EST,1960;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 25 Nov 87 16:17-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA01178@EDDIE.MIT.EDU>; Wed, 25 Nov 87 15:09:02 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA01169@EDDIE.MIT.EDU>; Wed, 25 Nov 87 15:08:47 EST
- Received: by june.cs.washington.edu (5.52.1/6.11)
- id AA16691; Wed, 25 Nov 87 12:10:19 PST
- Return-Path: <uunet!mnetor!jim@eddie.mit.edu>
- Message-Id: <8711252010.AA16691@june.cs.washington.edu>
- Date: 23 Nov 87 18:35:11 GMT
- From: uunet!mnetor!jim@EDDIE.MIT.edu (Jim Stewart)
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: Re: Packet freqs
- Keywords: repeater control, security
- References: <698@uop.UUCP> <1532@faline.bellcore.com> <250@splut.UUCP> <6791@apple.UUCP> <1565@faline.bellcore.com>
-
- In article <1565@faline.bellcore.com> karn@faline.bellcore.com (Phil R. Karn) writes:
-
- >Basically, the idea is to use DES encryption on the packet.
- >...This technique is in widespread use in the banking industry;...
-
- >For those of you familiar with DES, the exact technique is to encrypt
- >the packet with DES in the Cipher Block Chaining mode, and then transmit
- >the original plaintext packet along with the last cipherword.
-
- Why go to all the trouble of chaining? As I recall the IBM 3624 cash
- dispenser, while the pin is sent only encrypted, does the verification
- you speak of, with a single random byte sent both inside the encrypted
- field and outside in the clear. Even though only a single byte is
- used for verification, since DES works in groups of eight bytes, it
- is difficult to tell what the resultant cipher will be without the key.
-
- In other words, you don't need to chain, just encrypt the first 8 bytes,
- and make sure at least one of those bytes is different in every packet.
-
- --
- Jim Stewart, VE3SRJ
- UUCP: {decvax|allegra|ihnp4|linus|utcsri}!utzoo!mnetor!jim
- ARPA: mnetor!jim@seismo.css.gov
- BELL: (416)475-8980 x303
-
-
- 25-Nov-87 16:39:05-EST,6276;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 25 Nov 87 16:39-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA01235@EDDIE.MIT.EDU>; Wed, 25 Nov 87 15:11:47 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA01213@EDDIE.MIT.EDU>; Wed, 25 Nov 87 15:10:54 EST
- Received: by june.cs.washington.edu (5.52.1/6.11)
- id AA16790; Wed, 25 Nov 87 12:12:26 PST
- From: umunhum!!paulf@umunhum.stanford.edu
- Return-Path: <umunhum!!paulf@umunhum.stanford.edu>
- Message-Id: <8711252012.AA16790@june.cs.washington.edu>
- Date: 23 Nov 87 19:35:02 GMT
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: Oh, no, not AGAIN...
- Reply-To: paulf@umunhum.stanford.edu (Paul Flaherty)
-
- > Real Live OSI is "public domain" too -- ISO doesn't really like
- >to bless things that cost significant money to license (since ISO
- >represents many parties, most of whom by definition would be licensees).
- >Implementations are a different story -- somebody could sell commercial
- >TCP/IP code (Hmmm, don't Wollangong, NRC, etc.?) or OSI code.
-
- I never implied that the Standard itself wouldn't be public domain.
- I'm more concerned with the Implementations, which won't be. To repeat a
- point, most TCP/IP sites right now are running public domain software. Oh
- sure, there are a few people, like Wollongong, who are selling code, but
- what they're really selling is support. And, TWG WIN-TCP isn't used
- nearly as much as the Tek or CMU ports for VMS, which are available for the
- cost of the tape.
-
- The difference here is that the OSI protocols are commercial entities,
- while the DDN protocols are the property of the govenment, and by extension,
- all of us (well, at least in the US). Commercial entities exist to make
- money (certainly nothing wrong with that), while governmental entities serve
- their constituency (hopefully). The long of the short of it is that I don't
- see the same forces in OSI which propelled large numbers of Public Domain
- DDN implementations.
-
- > As one who lives in that space, I'll disagree: AT&T DS-1 was
- >a first stab; CEPT-1 (CCITT style) is a hell of a lot better. Like
- >PAL and SECAM vs. NTSC for color TV: The US innovates, Europe debugs
- >and comes up with a better standard(s), and other countries manufacture!
-
- As one who lived in that space (I used to work for Bell Labs, Toll
- Switching), I'll press the point: the result of the split was hideously
- expensive to the US, not only in lost overseas sales, but in the expense
- required to maintain the 4ESS International Interfaces (not to mention
- development, and Bell Labs ain't cheap). CEPT-1 may be "better" in that
- more calls are carried per trunk, but "what price betterment"?
-
- > IP addressing is unsuitable because it 1) requires call to IP
- >address mapping which is difficult in practice as nets get big (I guess
- >I don't have faith in on-air nameservers); 2) doesn't allow callsigns in
- >the datagram level headers; and 3) doesn't provide geographic
- >information unless you rule out portable/mobile operation. And probably
- >other reasons. IP procedures are not the problem, but to be in favor of
- >IP-style connectionless networks, one doesn't have to clone the wireline
- >Internet hook, line and sinker.
-
- Anything can be unsuitable if you don't implent it right. To
- address your concerns one by one:
-
- 1) Empirical experience with the 4.3bsd caching software has shown
- that most Internet sites do indeed exhibit locality; that is, they
- tend to access the same few sites over and over again. A cache size
- of about 16 is fine for most sites. With a limited table size, the
- performance limitation lies with a particular processor's ability,
- or lack thereof, to handle comparisons within a table. This is why
- we've been after the RISC people to keep fast compares, or provide
- on - chip ARP caches. Ergo, if this is a problem at all, it is
- a processor architecture one. This is also an artifact of running
- TCP/IP on top of AX.25, which is dog meat anyway.
-
- 2) Again, this is an artifact of AX.25.
-
- 3) You don't WANT your addresses to be geographically significant.
- If they are, then I'll need at least ten personally, as opposed to one with
- CGP.
-
- > Priced any Wollangong software lately? BTW, DECnet and SNA
- >protocols (most of 'em) are "free", but implementations aren't.
-
- That's because DECnet and SNA were promulgated by commercial concerns,
- like OSI. Again, most implentations of TCP/IP are freeware, or close to it.
-
- > Come on. It may be easier to implement a simpler protocol
- >(how about Kermit for everything? :-) ) but we're not at the stage yet
- >where we can dismiss DDN-IP/TCP _or_ OSI-IP/TP4 as inherently buggy.
-
- Time to go shaving with Occam's Razor. Clearly kermit does not
- provide the desired user functionality of either DDN or OSI. But of the
- two, there is little doubt as to which one is simpler internally. Just going
- by paper alone, the OSI TP4 and IP specs make up twice as many pages as
- TCP and IP. (RFC's 905 and 926 vs 791, 793.)
-
- > OSI and COSI aren't the same thing! Connectionless OSI is alive
- >and well. There's lots more work to be done, and TCP/IP is a great
- >thing for today, but we can't rest on our laurels and halt any progress
- >that isn't directly upward-compatible with it.
- > fred k1io
-
- Somehow I've never seen it that way. It makes far more sense to me
- to build on DDN TCP/IP, than to scuttle it and go to the OSI protocols.
- Again, this is because I have yet to see something significant that OSI can do
- that DDN cannot. Moreover, I see the move as being extremely costly to the
- end user.
-
- Upward compatibility is not necessarily a good thing. It becomes
- significant only when the foundation architecture has demonstrated enormous
- utility, which TCP/IP has, unquestioningly so. It becomes a bad thing when
- it gets in the way of clear and significant architectural improvement. And
- that has yet to be demonstrated with OSI.
-
-
-
-
-
-
-
- -=Paul Flaherty, N9FZX | "One Internet to rule them all,
- Computer Systems Laboratory | One Internet to find them,
- Stanford University | One Internet to bring them all
- Domain: paulf@shasta.Stanford.EDU| and in the ether bind them." -ToIH
-
-
- 25-Nov-87 16:44:07-EST,1622;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 25 Nov 87 16:44-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA01292@EDDIE.MIT.EDU>; Wed, 25 Nov 87 15:14:07 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA01287@EDDIE.MIT.EDU>; Wed, 25 Nov 87 15:13:49 EST
- Received: by june.cs.washington.edu (5.52.1/6.11)
- id AA16966; Wed, 25 Nov 87 12:15:21 PST
- Return-Path: <pyramid!voder!apple!winter@eddie.mit.edu>
- Message-Id: <8711252015.AA16966@june.cs.washington.edu>
- Date: 23 Nov 87 07:15:37 GMT
- From: pyramid!voder!apple!winter@EDDIE.MIT.edu (Patty Winter)
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: Re: AEA PK-232 control (terminal) program
- References: <7410@eddie.MIT.EDU>
-
- In article <7410@eddie.MIT.EDU> ZMTKeidel%UCIVMSA.BITNET@WISCVM.WISC.EDU writes:
- >I am interested in developing a control program which will interface
- >an Apple MacIntosh with the AEA PK-232.
-
- You might want to see whether Jack Brindle has added an AEA package
- to his TAPR and Kantronics versions of MacPacket. He's WA4FIB and is
- at 3155 Resin Street, Marietta, GA 30066. Might save you some work.
-
- >However, I have heard that
- >there are not many HAMS with MacIntosh computers! I have one, my
- >roomate has three, and even the engineers at AEA have Mac's.
-
- Well, I have two. :-) :-) Unfortunately for your survey, I'm not
- using either of them for packet....
-
-
- Patty
-
- --
- Patty Winter N6BIS (408) 973-2814
- M/S 2C, Apple Computer, Inc., 20525 Mariani Ave., Cupertino, CA 95014
- {decwrl,nsc,sun,dual}!apple!winter
-
-
- 25-Nov-87 18:24:07-EST,1649;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 25 Nov 87 18:24-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA01479@EDDIE.MIT.EDU>; Wed, 25 Nov 87 15:22:59 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA01474@EDDIE.MIT.EDU>; Wed, 25 Nov 87 15:22:45 EST
- Received: by june.cs.washington.edu (5.52.1/6.11)
- id AA17930; Wed, 25 Nov 87 12:24:20 PST
- Return-Path: <bellcore!faline!karn@eddie.mit.edu>
- Message-Id: <8711252024.AA17930@june.cs.washington.edu>
- Date: 25 Nov 87 01:16:25 GMT
- From: bellcore!faline!karn@EDDIE.MIT.edu (Phil R. Karn)
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: Re: Packet freqs
- Summary: authentication
- Keywords: repeater control, security
- References: <698@uop.UUCP> <1532@faline.bellcore.com> <250@splut.UUCP> <4346@mnetor.UUCP>
-
- > Why go to all the trouble of chaining?
-
- Well, it protects against any part of your packet being modified. Of
- course, repeater control packets aren't likely to be very big in the
- first place. If they're 8 bytes or less, then "chaining" has no meaning.
- You just transmit them along with their encrypted equivalents. Remember
- to include a sequence number or time-of-day clock to prevent playback
- attacks.
-
- I'm surprised no one has pointed out the one vulnerability of my scheme
- for providing repeater control functions off a packet switch: the
- control frequency becomes known. You may not be able to enter an
- unauthorized command, but you could probably jam out an authorized
- control operator. Of course, in this day of R-7000s it probably woudn't
- take THAT long to track down a control channel if it is used at all.
-
- Phil
-
-
- 25-Nov-87 21:10:39-EST,861;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 25 Nov 87 21:10-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA06583@EDDIE.MIT.EDU>; Wed, 25 Nov 87 19:46:34 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA06573@EDDIE.MIT.EDU>; Wed, 25 Nov 87 19:46:18 EST
- Received: from huey.udel.edu by Louie.UDEL.EDU id aa07498; 25 Nov 87 19:10 EST
- Date: Wed, 25 Nov 87 19:08:53 EST
- From: Mills@UDEL.EDU
- To: "Phil R. Karn" <bellcore!faline!karn@eddie.mit.edu>
- Cc: PACKET-RADIO@eddie.mit.edu
- Subject: Re: Packet freqs
- Message-Id: <8711251908.aa08427@Huey.UDEL.EDU>
-
- Phil,
-
- The easy way to do it is to encrypt the checksums. Yaas, yaas, you can
- byte-swamp the entire message, sort the words and still come out with
- the same checksums, but that is independent of the encryption. Life
- is much too short.
-
- Dave
- 25-Nov-87 21:31:33-EST,2506;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 25 Nov 87 21:31-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA06743@EDDIE.MIT.EDU>; Wed, 25 Nov 87 19:55:05 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA06739@EDDIE.MIT.EDU>; Wed, 25 Nov 87 19:54:43 EST
- Received: by june.cs.washington.edu (5.52.1/6.11)
- id AA25005; Wed, 25 Nov 87 15:12:03 PST
- Return-Path: <athena.mit.edu!wesommer@athena.mit.edu>
- Message-Id: <8711252312.AA25005@june.cs.washington.edu>
- From: athena.mit.edu!wesommer@athena.mit.edu (William Sommerfeld)
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: Re: Packet freqs
- Keywords: repeater control, security
- Date: 24 Nov 87 10:53:53 GMT
- References: <698@uop.UUCP> <1532@faline.bellcore.com> <250@splut.UUCP> <6791@apple.UUCP> <1565@faline.bellcore.com> <4346@mnetor.UUCP>
- Reply-To: wesommer@athena.mit.edu (William Sommerfeld)
-
- In article <4346@mnetor.UUCP> jim@mnetor.UUCP (Jim Stewart) writes:
- >Why go to all the trouble of chaining? As I recall the IBM 3624 cash
- >dispenser, while the pin is sent only encrypted, does the verification
- >you speak of, with a single random byte sent both inside the encrypted
- >field and outside in the clear. Even though only a single byte is
- >used for verification, since DES works in groups of eight bytes, it
- >is difficult to tell what the resultant cipher will be without the key.
- >
- >In other words, you don't need to chain, just encrypt the first 8 bytes,
- >and make sure at least one of those bytes is different in every packet.
-
- If the "command" you send extends outside the first cypher block, then
- someone wanting to spoof the repeater could listen in and grab the
- first and last cypher blocks of the transmission, and later replay
- those blocks, bracketing a modified command.
-
- Of course, even if you do CBC encrypt the whole message, and send the
- final checksum, you still stand the risk of someone just replaying
- your transmission verbatim at a later time (not being particularly
- familiar with the workings of repeaters, I assume that typical
- repeater control commands are simple things such as "shut down the
- repeater" or "disable the autopatch"); this kind of defeats the
- purpose of encrypting.
-
- To _really_ guard against tampering, you have to include some sort of
- timestamp/sequence number in the plaintext message, which the repeater
- would know how to verify for each authorized control operator.
-
- Bill Sommerfeld, N1FHN
- wesommer@athena.mit.edu
-
-
- 25-Nov-87 21:39:10-EST,1954;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 25 Nov 87 21:39-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA06756@EDDIE.MIT.EDU>; Wed, 25 Nov 87 19:55:31 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA06745@EDDIE.MIT.EDU>; Wed, 25 Nov 87 19:55:15 EST
- Received: by june.cs.washington.edu (5.52.1/6.11)
- id AA25901; Wed, 25 Nov 87 15:23:28 PST
- Return-Path: <bellcore!faline!karn@EDDIE.MIT.EDU>
- Message-Id: <8711252323.AA25901@june.cs.washington.edu>
- Date: 25 Nov 87 03:43:15 GMT
- From: bellcore!faline!karn@eddie.mit.edu (Phil R. Karn)
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: Re: IP and roundtables/nets?
- Summary: multicasting is much easier if the subnet supports it
- Keywords: multicasting, UDP
- References: <4732@spool.wisc.edu>
-
- > problems involved, but I was wondering if anyone had done any work
- > on multiconnection or multicast protocols that would support such things.
-
- This is one of my favorite back-burner topics. Right now we are almost
- completely squandering one of the few advantages that radio has for
- sending data (other than it being "fun", of course!). That is, the
- ability for a single transmission to be heard by multiple receivers.
- Existing "conference bridges" that send N copies of the data out on N
- separate virtual circuits to N different receivers (all on the same
- channel) are utterly silly.
-
- Multicasting is FAR easier when the "subnet" (e.g., radio channel and
- associated hardware) supports it directly. Lossy channels make reliable
- multicasting almost impossible, so things should be configured such that
- each packet will get through with a very high degree of probability
- WITHOUT acknowledgements from each receiver. This means coordinated
- frequency assignments, full-duplex operation (most likely cross-band)
- and other things that I talked about in my ARRL conference paper "A High
- Performance Collision Free Packet Radio Network".
-
- Phil
-
-
- 25-Nov-87 21:45:58-EST,1679;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 25 Nov 87 21:45-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA07231@EDDIE.MIT.EDU>; Wed, 25 Nov 87 20:24:45 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA07224@EDDIE.MIT.EDU>; Wed, 25 Nov 87 20:24:24 EST
- Received: by june.cs.washington.edu (5.52.1/6.11)
- id AA25151; Wed, 25 Nov 87 15:13:15 PST
- Return-Path: <husc6!uwvax!caseus!dan@EDDIE.MIT.EDU>
- Message-Id: <8711252313.AA25151@june.cs.washington.edu>
- Date: 24 Nov 87 14:38:56 GMT
- From: husc6!uwvax!caseus!dan@eddie.mit.edu (Daniel M. Frank)
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: IP and roundtables/nets?
- Keywords: multicasting, UDP
- Reply-To: dan@caseus.wisc.edu (Daniel M. Frank)
-
- I vaguely remember this question being brought up a while back, so
- I apologize for the repetition ...
-
- The NTS people in Wisconsin are starting to talk about setting up
- an organized packet network in the state, mostly for traffic handling.
- One of the questions that came up was about support for roundtable
- traffic nets. I'm aware of some of the technical and performance
- problems involved, but I was wondering if anyone had done any work
- on multiconnection or multicast protocols that would support such things.
- I know there are some RFCs on multicasting. Is any of this work
- applicable?
-
- Also, I'd appreciate getting any "case histories" of TCP/IP use for
- intrastate traffic handling, for obvious reasons.
-
- Thanks!
-
- Dan Frank (w9nk)
- ARPA: dan@db.wisc.edu UUCP: ... uwvax!dan
- Attention Democratic Party Line Eater:
- {prosperity,freedom,work,responsibility,honesty,self-reliance}
-
-
- 25-Nov-87 21:56:04-EST,1999;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 25 Nov 87 21:56-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA07248@EDDIE.MIT.EDU>; Wed, 25 Nov 87 20:25:29 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA07234@EDDIE.MIT.EDU>; Wed, 25 Nov 87 20:25:00 EST
- Received: by june.cs.washington.edu (5.52.1/6.11)
- id AA25784; Wed, 25 Nov 87 15:20:18 PST
- Return-Path: <pyramid!prls!philabs!trotter!bill@EDDIE.MIT.EDU>
- Message-Id: <8711252320.AA25784@june.cs.washington.edu>
- Date: 23 Nov 87 17:08:10 GMT
- From: pyramid!prls!philabs!trotter!bill@eddie.mit.edu (Bill Gunshannon)
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: Re: Let's calm down the CONS/CLNS flames just a bit?
- Summary: N2WX Level 3 Switch
- References: <8711201629.AA23107@decwrl.dec.com>
-
-
- I have a new question. Why is it with all the arguments about COSI,
- TCP-IP, NETROM, etc. I never hear anyone mention the N2WX Level 3 Switch
- software. I just put up a PAC-COMM DR200 and am impressed by what I have
- seen so far.
-
- I would be interested in hearing from anyone else who has experience with
- this software. I would like to learn more about the statistics and also
- if there are any major or minor networks set up using it. I am trying to
- get something going on 6 meters here on the east coast. Maybe this would
- be a good way to start. If others are interested we can try and set up an
- engineered network rather than just letting it grow out of control on it's
- own (145.01).
-
- Any information or comments would be appreciated.
-
- And by the way how does one get in touch with N2WX??? Is he on the net
- anywhere??
-
- 73's
-
-
- bill gunshannon
-
-
- UUCP: {philabs}\ US SNAIL: Martin Marietta Data Systems
- {phri } >!trotter.usma.edu!bill USMA, Bldg 600, Room 26
- {sunybcs}/ West Point, NY 10996
- RADIO: KB3YV PHONE: WORK (914)446-7747
- AX.25: KB3YV @ K3RLI PHONE: HOME (914)565-5256
-
-
- 25-Nov-87 22:13:00-EST,5134;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 25 Nov 87 22:12-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA06863@EDDIE.MIT.EDU>; Wed, 25 Nov 87 20:03:12 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA06859@EDDIE.MIT.EDU>; Wed, 25 Nov 87 20:02:59 EST
- Received: by june.cs.washington.edu (5.52.1/6.11)
- id AA16560; Wed, 25 Nov 87 12:06:45 PST
- Return-Path: <delni.dec.com!goldstein@DECWRL.DEC.COM>
- Message-Id: <8711252006.AA16560@june.cs.washington.edu>
- Date: 23 Nov 87 18:02:00 GMT
- From: delni.dec.com!goldstein@DECWRL.DEC.COM (Fred R. Goldstein dtn226-7388)
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: trying to throw water on protocol flames
-
- Paul Flaherty replies to me,
-
- IO>True enough at the moment; DDN Suite provides the most popular network
- IO>functions on a public domain multi-vendor basis. It does provide an
- IO>open systems interconnection (lower case) function! ISO OSI is a more
- IO>ambitious project of which X.25-1984 is a small piece.
-
- FZX> The key operative here is "public domain". Most of the TCP/IP
- FZX>implementations running today were hacked out on the govenment dole, and
- FZX>are henceforth free. Some individuals have even been known to use all of
- FZX>their "recreation time" to do this.
-
- FZX> This, of course, is entirely unacceptable to certain vendors, who
- FZX>woudl stand to make tons of money selling networking software, if the
- FZX>public domain stuff wasn't there. So, don't be suprised to see them
- FZX>hawking their wares here.
-
- Real Live OSI is "public domain" too -- ISO doesn't really like
- to bless things that cost significant money to license (since ISO
- represents many parties, most of whom by definition would be licensees).
- Implementations are a different story -- somebody could sell commercial
- TCP/IP code (Hmmm, don't Wollangong, NRC, etc.?) or OSI code.
-
- FZX> Frankly, all of this reminds me of the US T1 vs. CCITT T1 battles;
- FZX> the US poured tons of money to develop the concept, and the CCITT made
- FZX> subtle changes, so that the US couldn't sell its technology overseas.
-
- As one who lives in that space, I'll disagree: AT&T DS-1 was
- a first stab; CEPT-1 (CCITT style) is a hell of a lot better. Like
- PAL and SECAM vs. NTSC for color TV: The US innovates, Europe debugs
- and comes up with a better standard(s), and other countries manufacture!
-
- IO> If you like TCP you shouldn't
- IO>dislike TP4, though you might not like TP0 (the "do nothing" Transport)!
-
- FZX> And, speaking of fast, what about UDP?
-
- ISO is beginning to look at connectionless applications stacks.
- They're behind.
-
- FZX> Okay, I'll bite. Why is IP addressing unsuitable to the ampr?
-
- IP addressing is unsuitable because it 1) requires call to IP
- address mapping which is difficult in practice as nets get big (I guess
- I don't have faith in on-air nameservers); 2) doesn't allow callsigns in
- the datagram level headers; and 3) doesn't provide geographic
- information unless you rule out portable/mobile operation. And probably
- other reasons. IP procedures are not the problem, but to be in favor of
- IP-style connectionless networks, one doesn't have to clone the wireline
- Internet hook, line and sinker.
-
- FZX> Of course TCP/IP isn't practical for inter - vendor communication,
- FZX>because they can't make any money off of it.
-
- Priced any Wollangong software lately? BTW, DECnet and SNA
- protocols (most of 'em) are "free", but implementations aren't.
-
- IO>Implementations aren't the key to protocol beauty contests like this
- IO>one.
-
- FZX> Wrong. Certain features of protocols can make them immune to
- FZX>software faults; conversely, some features make protocols error prone.
-
- Come on. It may be easier to implement a simpler protocol
- (how about Kermit for everything? :-) ) but we're not at the stage yet
- where we can dismiss DDN-IP/TCP _or_ OSI-IP/TP4 as inherently buggy.
-
- FZX> This is NOT a beauty contest. We are trying to make a serious
- FZX>comparison of two protocol families, before we lock ourselves into one.
- FZX>That way, we can hopefully avoid expensive mistakes like AX.25.
-
- I agree on the goal! But OSI is not one protocol stack like
- DDN; it is a family whose members are not all in agreement (i.e., the
- X.25-TP0 clan vs. 8473-TP4). BTW, the term "protocol beauty contest"
- comes from Eric K3NA in his agendas for ANSI T1D1.2 a couple years
- back; I confess to having stolen his term.
-
- FZX> The de facto international open systems interconnect is TCP/IP.
-
- True. Today. If it accomodates your "style of computing";
- it won't fit in all the mainframers who are playing OSI. It does
- handle AMPR nicely, except for the addressing and concomitant routing
- problems.
-
- FZX>The burden of proof clearly lies with those in COSI - land.
-
- OSI and COSI aren't the same thing! Connectionless OSI is alive
- and well. There's lots more work to be done, and TCP/IP is a great
- thing for today, but we can't rest on our laurels and halt any progress
- that isn't directly upward-compatible with it.
- fred k1io
-
-
- 25-Nov-87 23:40:13-EST,2093;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 25 Nov 87 23:40-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA07197@EDDIE.MIT.EDU>; Wed, 25 Nov 87 20:23:05 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA07193@EDDIE.MIT.EDU>; Wed, 25 Nov 87 20:22:44 EST
- Received: by june.cs.washington.edu (5.52.1/6.11)
- id AA26180; Wed, 25 Nov 87 15:29:37 PST
- Return-Path: <bellcore!faline!karn@EDDIE.MIT.EDU>
- Message-Id: <8711252329.AA26180@june.cs.washington.edu>
- Date: 25 Nov 87 03:09:48 GMT
- From: bellcore!faline!karn@eddie.mit.edu (Phil R. Karn)
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: Re: Repeater advice anyone?
- Summary: problems with implicit acks
- References: <7446@eddie.MIT.EDU>
-
- > DARPA packet radios use a technique called pacing, in which the onward
- > transmission of a packet from the digipeater to the next hop is heard
- > by the original transmitter as the ACK for its transmission. Since the
- > digipeater will delay until the channel is quiet, this forms an automatic
- > sensing mechanism for hidden transmitters and without the channel overhead
- > of additional ACKs. The sender, of course, should delay once his relayed
- > transmission is heard to avoid collision on the next hop, just as you
- > describe.
-
- "Implicit acks" (using the retransmission of your packet by a digipeater
- as an acknowledgement that your packet was received by the digipeater)
- is an appealing idea. It does have its pitfalls, though. The protocol
- must be designed to allow the digipeater to filter out duplicates
- generated when the sending station doesn't hear the retransmission of a
- packet even though the digipeater relayed it successfully to the next
- station down the chain. Otherwise it's possible to get a geometric
- explosion of duplicate packets.
-
- Hidden transmitters are still a problem, since there may be "hidden"
- traffic on frequency that doesn't involve the digipeater. And there can,
- of course, be collisions by user stations out of direct range with each
- other that try to access the digipeater simultaneously.
-
- Phil
-
-
- 26-Nov-87 00:00:05-EST,4851;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 26 Nov 87 00:00-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA07221@EDDIE.MIT.EDU>; Wed, 25 Nov 87 20:24:08 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA07204@EDDIE.MIT.EDU>; Wed, 25 Nov 87 20:23:27 EST
- Received: by june.cs.washington.edu (5.52.1/6.11)
- id AA26038; Wed, 25 Nov 87 15:27:37 PST
- Return-Path: <bellcore!faline!karn@EDDIE.MIT.EDU>
- Message-Id: <8711252327.AA26038@june.cs.washington.edu>
- From: bellcore!faline!karn@EDDIE.MIT.edu (Phil R. Karn)
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Newsgroups: rec.ham-radio.packet
- Subject: Re: trying to throw water on protocol flames
- Summary: ip addressing
- Date: 25 Nov 87 03:00:17 GMT
- References: <8711231507.AA07190@decwrl.dec.com>
-
- > IP addressing is unsuitable because it 1) requires call to IP
- > address mapping which is difficult in practice as nets get big (I guess
- > I don't have faith in on-air nameservers); 2) doesn't allow callsigns in
- > the datagram level headers; and 3) doesn't provide geographic
- > information unless you rule out portable/mobile operation. And probably
- > other reasons. IP procedures are not the problem, but to be in favor of
- > IP-style connectionless networks, one doesn't have to clone the wireline
- > Internet hook, line and sinker.
-
- We've been over this one many times, in many different forms. You will
- need one of the following in ANY amateur network:
-
- 1. callsign to address mapping
-
- -or-
-
- 2. "flat" routing tables large enough to hold an entry for each and
- every station in the network.
-
- The fundamental problem is that callsigns do not contain any information
- about geographic location. Even if they did, they would not necessarily
- tell you anything useful about routing. Hence you will always need
- *some* routing tables, flat or hierarchical. Given (for the moment) that
- purely "flat" routing tables are impractically large, we will have to
- incur the administrative burden of assigning addresses to stations with
- the topology of the network in mind, and of mapping callsigns to these
- addresses.
-
- We have no lack of proposals for addressing schemes. Grid squares,
- telephone area codes, airport identifiers, postal zip codes and even
- congressional districts have been proposed. Each has ardent proponents
- who argue that theirs is *the* easiest to use. True enough, almost
- everyone knows their own telephone number, mailing address and home
- airport. Most hams know, or can easily find out, what grid square they
- live in. Some even know the name of their Representative (though not
- necessarily his district number). One problem with all these schemes,
- however, is that they are based on geography, politics and historical
- accidents, not network topology. Another, more serious one: you may know
- your own address well enough, but how will you find that of your
- destination party if all you know is his callsign?
-
- Worse, consider a station operating portable or mobile. (Amateur packet
- voice is not that far away). Must he change his address and notify
- everyone who might try to reach him when operating away from home, even
- temporarily? (I thought we were trying to avoid the "frequently updated
- distributed database", i.e., "nameserver", problem).
-
- Clearly, you want address assignments to be relatively stable, whether
- or not you happen to believe in nameservers. (SOME mechanism will have
- to be found to publish address assignments, and even if this is done
- through the network itself you want to minimize the frequency of
- changes). Any attempt to "hardwire" geographic location fields into addresses
- is therefore a Bad Idea, not only because it makes life hard for those
- trying to communicate with stations in motion, but because it makes
- addresses so much larger without commeasurate benefit.
-
- Surely our network should be able to allow at least a small number of
- stations to move around without immediately forcing them to change
- addresses. What's needed is a hybrid between hierarchical and flat
- addressing that combines the efficiency of the former for the majority
- of stations that remain fixed with the flexibility of the latter for
- those relatively few stations in motion. I believe I have laid the
- foundation for this in the way I've implemented IP address routing in my
- TCP/IP package; we simply don't need the kind of address expansion Fred
- proposes.
-
- One final, radical, thought: flat addressing isn't all that unthinkable
- anymore. Consider that there are only about 300,000 hams in the US (in
- very round numbers). Dynamic RAM is now cheap enough that I could store
- an IP routing table entry for each and every US ham in perhaps $125
- worth of chips, even without any clever storage compression tricks. And
- that price is of course only going to go down.
-
- Phil
-
-
- 26-Nov-87 02:00:00-EST,1342;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 26 Nov 87 01:59-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA11189@EDDIE.MIT.EDU>; Thu, 26 Nov 87 00:47:30 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA11177@EDDIE.MIT.EDU>; Thu, 26 Nov 87 00:47:13 EST
- Received: from huey.udel.edu by Louie.UDEL.EDU id aa12294; 26 Nov 87 0:45 EST
- Date: Thu, 26 Nov 87 0:42:09 EST
- From: Mills@UDEL.EDU
- To: "Phil R. Karn" <bellcore!faline!karn@eddie.mit.edu>
- Cc: PACKET-RADIO@eddie.mit.edu
- Subject: Re: Repeater advice anyone?
- Message-Id: <8711260042.aa00464@Huey.UDEL.EDU>
-
- Phil,
-
- All that is quite true, but pacing sure improves performance of the system.
- Next bauble to play is tier-based routing, a variant of distributed
- Bellman-Ford (min-hop) algorithms in which a relayed packet can go
- "forward," "sideways" (by one hop), but not "backwards."
-
- Gently, gently the DARPA secrets are exposed to the teeming throng. See
- Proc. IEEE November 1978 and subsequent articles scattered widely and
- thinly so you may never find them. Unfortunately, the SURAN (Survivable
- Radio Network) program has not published on the traditional RFC (Request
- for Comments) frequencies, so the technology has fallen below the noise
- level of most receivers. Maybe something can be done about that.
-
- Dave
- 26-Nov-87 03:42:16-EST,1115;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 26 Nov 87 03:42-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA11075@EDDIE.MIT.EDU>; Thu, 26 Nov 87 00:37:01 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA11061@EDDIE.MIT.EDU>; Thu, 26 Nov 87 00:36:42 EST
- Received: from huey.udel.edu by Louie.UDEL.EDU id aa11754; 26 Nov 87 0:34 EST
- Date: Thu, 26 Nov 87 0:32:20 EST
- From: Mills@UDEL.EDU
- To: "Phil R. Karn" <bellcore!faline!karn@eddie.mit.edu>
- Cc: PACKET-RADIO@eddie.mit.edu
- Subject: Re: IP and roundtables/nets?
- Message-Id: <8711260032.aa00434@Huey.UDEL.EDU>
-
- Phil,
-
- I recommend the NCC Proceedings 1979 (Washington, DC) in which the DARPA
- SATNET system "coming out party" was honked to the world precisely on
- the merits you suggest and subsequently forgotten. Even INTELSAT forgot,
- in spite of my hooting and hollering at the time. The broadcast nature
- of the satellite/packet-radio channel is an exquisitely precious feature
- too valuable to ignore. Virtual-circuit fanatics should really learn to
- understand and exploit this feature.
-
- Dave
- 26-Nov-87 12:19:17-EST,528;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 26 Nov 87 12:19-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA19194@EDDIE.MIT.EDU>; Thu, 26 Nov 87 11:38:06 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA19186@EDDIE.MIT.EDU>; Thu, 26 Nov 87 11:37:56 EST
- Date: Thu, 26 Nov 87 11:36 EST
- From: McAuliffe@RADC-MULTICS.ARPA
- Subject: Removal
- To: packet-radio@EDDIE.MIT.EDU
- Message-Id: <871126163602.385105@RADC-MULTICS.ARPA>
-
- Please remove me from the packet-radio list also.
- 26-Nov-87 22:59:34-EST,5278;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 26 Nov 87 22:59-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA25289@EDDIE.MIT.EDU>; Thu, 26 Nov 87 22:03:12 EST
- Received: from mc.lcs.mit.edu by EDDIE.MIT.EDU via Chaosnet with SMTP with sendmail-5.45/4.7 id <AA25282@EDDIE.MIT.EDU>; Thu, 26 Nov 87 22:02:58 EST
- Resent-Message-Id: <8711270302.AA25282@EDDIE.MIT.EDU>
- Received: from SIMTEL20.ARPA (TCP 3200000112) by MC.LCS.MIT.EDU 26 Nov 87 22:09:11 EST
- Date: Wednesday, 25 November 1987 13:53-MST
- Message-Id: <KPETERSEN.12353831510.BABYL@SIMTEL20.ARPA>
- Sender: deneb.ucdavis.edu!ccruss@UCDAVIS.UCDAVIS.EDU (Russ Hobby)
- From: deneb.ucdavis.edu!ccruss@UCDAVIS.UCDAVIS.EDU (Russ Hobby)
- To: tcp-ip@SRI-NIC.ARPA
- Subject: SLIP software available
- Resent-From: KPETERSEN@SIMTEL20.ARPA
- Resent-To: packet-radio%eddie.mit.edu@mc.lcs.mit.edu
- Resent-Date: Thu 26 Nov 1987 20:03-MST
-
- Yes, we have a SLIP server implementation now ready. The reason I have
- not posted it sooner is because we have been working with the Sun NFS
- people on the method of establishing a dialin slip connection and have
- been refitting some of their suggestions into our version.
-
- The software allows PCs to make dialup SLIP connections to the campus IP
- network. We are also have worked out a method of abbreviated serial line
- IP (ASLIP) packeting that makes dialup IP networks more efficient. The
- ASLIP software is new to this version and Sun has not seen it yet.
-
- Here is how the system works. The user logs on to the host that is
- acting as the gateway, a 4.3 bsd system. He then types in the command
- "slip" and he becomes a host on the network. He can then use all the
- programs that come with the CMU/MIT PC/IP or Phil Karn's system. To make
- connecting to the network a little easier, we have written a program
- that will do the complete login automatically. The program has a user
- configurable script file that specifies a sequence of strings to send
- out the serial line and responses to wait for coming back. It has its
- own simple language with branching depending on if the correct response
- came back. The net result is that after the script has been set up, the
- user types in one command on the PC to connect to the network.
-
- Unlike some PC/IPs, our system assumes that each PC (or actually each
- user) has its own, permanent IP address. Security is provided by logon
- security on the gateway host. The IP address of the PC is associated
- with the usercode on the gateway host. The network connection for that
- PC's IP address can only be started from a user logged in with the
- correct usercode. The system also makes sure that the IP address is not
- already connected before making a new connection.
-
- The way we have set up IP address for the PCs is to have a separate
- subnet that contains all the PCs. This way the gateway host needs only
- to advertise that it is a route to that subnet and all the PCs are
- covered. In essence the gateway host is the network for all SLIP
- connections on that subnet.
-
- The abbreviated packets work on the assumption that the connecting
- computer is an endnode and will not be doing any routing. In this case
- many of the fields in the IP packet are unnecessary. ASLIP uses the
- minimum header size based on this assumption. With ASLIP the header size
- is either 8 or 4 bytes, much smaller than the standard 40 bytes. The
- host that is acting as the ASLIP gateway rebuilds the outgoing packets
- to legal IP packets before sending them out the network and abbreviates
- the incoming packets from the network. The same server software handles
- both SLIP and ASLIP. It only goes into ASLIP mode once it has received
- an ASLIP packet from the PC, thus if only SLIP packets are sent, it will
- stay in regular SLIP mode. I will post more details on the ASLIP
- protocol later.
-
- Also there have been some terminal server vendors interested in this
- project. It should not be much work to turn a terminal server into an
- ASLIP or SLIP server and that would make it cheaper than using a VAX as
- the gateway. Plus there would not be as much maintenance and downtime
- with a simple server box.
-
- The software is available via anonymous FTP from ucdavis.edu and is in
- the directory dist/slip/. This includes all software to run the
- SLIP/ASLIP server on a 4.3 bsd system, and the modifications to CMU
- PC/IP software to implement ASLIP and the auto-login program, plus fix a
- couple bugs. See the README file in this directory for a discription of
- what the various tar files give you.
-
- The next thing we want to add to the system is BOOTP so that the PC
- software does not have to be configured for IP address, but rather get
- it from the server.
-
- Russell Hobby
- Data Communications Manager
- U. C. Davis
- Computing Services BITNET: RDHOBBY@UCDAVIS
- Davis Ca 95616 UUCP: ...!ucbvax!ucdavis!rdhobby
- (916) 752-0236 INTERNET: rdhobby@ucdavis.edu
-
- 27-Nov-87 04:28:50-EST,660;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 27 Nov 87 04:28-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA28207@EDDIE.MIT.EDU>; Fri, 27 Nov 87 03:29:04 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA28200@EDDIE.MIT.EDU>; Fri, 27 Nov 87 03:28:50 EST
- Posted-Date: 27 Nov 87 9:08 +0100
- Received: by tor.nta.no (5.54/3.21)
- id AA08916; Fri, 27 Nov 87 09:08:28 +0100
- Date: 27 Nov 87 9:08 +0100
- From: Tore J Berg <tore%tor.re.nta.uninett@TOR.nta.no>
- To: <packet-radio@EDDIE.MIT.EDU>
- Message-Id: <270*tore@tor.re.nta.uninett>
- Subject: Removal
-
- Please remove me from the packet-radio list.
-
- Tore/NDRE
- 28-Nov-87 12:06:23-EST,1067;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 28 Nov 87 12:06-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA19275@EDDIE.MIT.EDU>; Sat, 28 Nov 87 11:17:03 EST
- Received: from mc.lcs.mit.edu by EDDIE.MIT.EDU via Chaosnet with SMTP with sendmail-5.45/4.7 id <AA19252@EDDIE.MIT.EDU>; Sat, 28 Nov 87 11:16:34 EST
- Resent-Message-Id: <8711281616.AA19252@EDDIE.MIT.EDU>
- Received: from SIMTEL20.ARPA (TCP 3200000112) by MC.LCS.MIT.EDU 28 Nov 87 11:19:01 EST
- Date: Wednesday, 25 November 1987 14:39-MST
- Message-Id: <KPETERSEN.12354237431.BABYL@SIMTEL20.ARPA>
- Sender: ZAKAR@A.ISI.EDU
- From: ZAKAR@A.ISI.EDU
- To: info-hams@SIMTEL20.ARPA
- Subject: AX.25
- Resent-From: KPETERSEN@SIMTEL20.ARPA
- Resent-To: packet-radio%eddie.mit.edu@mc.lcs.mit.edu
- Resent-Date: Sat 28 Nov 1987 09:13-MST
-
- Folks,
- i'm just getting started in packet radio. I understand that I can
- get a copy of the AX.25 spec from ARRL, but does anyone know if there
- is an electronic copy available over Arpanet?
-
- Thanks,
- Joe Zakar
- Zakar at A.ISI.EDU
- .
-
- 30-Nov-87 17:10:52-EST,2927;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 30 Nov 87 17:10-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA00658@EDDIE.MIT.EDU>; Mon, 30 Nov 87 15:52:37 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA00652@EDDIE.MIT.EDU>; Mon, 30 Nov 87 15:52:19 EST
- Received: by june.cs.washington.edu (5.52.1/6.11)
- id AA26067; Sun, 29 Nov 87 09:27:27 PST
- Return-Path: <uunet!nuchat!sugar!splut!jay@EDDIE.MIT.EDU>
- Message-Id: <8711291727.AA26067@june.cs.washington.edu>
- Date: 28 Nov 87 21:55:19 GMT
- From: uunet!nuchat!sugar!splut!jay@EDDIE.MIT.edu (Jay Maynard)
- To: PACKET-RADIO@EDDIE.MIT.EDU
- Subject: Re: trying to throw water on protocol flames
- References: <8711231507.AA07190@decwrl.dec.com> <1574@faline.bellcore.com>
-
- Phil answered most of my questions about IP address assignment versus
- callsigns, except for one: how can we keep the FCC happy? (NOTE: I'm _NOT_
- flaming, but asking for info.) How can they (or we, in our time-honored role
- as self-policing spectrum users) know that 44.93.6.129 is K5ZC? (No, that's
- _not_ my address; I haven't gotten an answer yet from E-mail to WB6JPR about
- getting a Texas block assignment.)
-
- In article <1574@faline.bellcore.com>, karn@faline.bellcore.com (Phil R. Karn) writes:
- > Surely our network should be able to allow at least a small number of
- > stations to move around without immediately forcing them to change
- > addresses. What's needed is a hybrid between hierarchical and flat
- > addressing that combines the efficiency of the former for the majority
- > of stations that remain fixed with the flexibility of the latter for
- > those relatively few stations in motion. I believe I have laid the
- > foundation for this in the way I've implemented IP address routing in my
- > TCP/IP package; we simply don't need the kind of address expansion Fred
- > proposes.
-
- Huh? Please expand on this; you allow a very nice hierarchical address
- structure, but what happens when I take my Houston-area address with me when
- I move to Dallas? or San Diego? The high order bits don't apply any more.
-
- > One final, radical, thought: flat addressing isn't all that unthinkable
- > anymore. Consider that there are only about 300,000 hams in the US (in
- > very round numbers). Dynamic RAM is now cheap enough that I could store
- > an IP routing table entry for each and every US ham in perhaps $125
- > worth of chips, even without any clever storage compression tricks. And
- > that price is of course only going to go down.
-
- Transferring that data file would be a real pain, though; talk about network
- loading!
-
- --
- Jay Maynard, K5ZC (@WB5BBW)...>splut!< | GEnie: JAYMAYNARD CI$: 71036,1603
- uucp: {uunet!nuchat,academ!uhnix1,{ihnp4,bellcore,killer}!tness1}!splut!jay
- Never ascribe to malice that which can adequately be explained by stupidity.
- The opinions herein are shared by none of my cats, much less anyone else.
-
-
- 30-Nov-87 21:52:31-EST,1567;000000000000
- Received: from EDDIE.MIT.EDU by DEEP-THOUGHT.MIT.EDU via Chaosnet; 30 Nov 87 21:52-EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA08248@EDDIE.MIT.EDU>; Mon, 30 Nov 87 20:44:08 EST
- Received: by EDDIE.MIT.EDU with sendmail-5.45/4.7 id <AA08210@EDDIE.MIT.EDU>; Mon, 30 Nov 87 20:42:19 EST
- Message-Id: <8712010142.AA08210@EDDIE.MIT.EDU>
- Received: from DHVRRZN1.BITNET by wiscvm.wisc.edu ; Mon, 30 Nov 87 17:31:13 CDT
- Date: Mon, 30 Nov 87 19:32:21 MEZ
- To: packet-radio@eddie.mit.EDU
- From: MAMI%DHVRRZN1.BITNET@wiscvm.wisc.edu
- Subject: searching TNC2C-Terminal prg for ATARI & MAC
-
- hello ob's,
-
- I am new with packet radio and I am looking for a special terminal
- program to work with TNC2C with WA8DED firmware in host mode for
- two kinds of computers, the MAC and the Atari ST. If any body has
- an idea, where I can get that adopted programespecially for the MAC,
- please let me know. I have full access to all BITNET servers and I
- wonder if there is any availibilty for BITNETTER's getting prog's
- and sources of such specified code??? I subscribe the packet-radio
- bulletin since 3 month, but I don't remember anything in that connection.
-
- Second kind of problems is the solution of TCP/IP for both of these
- computers. Atari is well known to much hams here in Hannover but MAC
- is not. Therefore system support for MAC with public domain software
- with packet-radio application is closed to ZERO.
-
- It would be wonderful to get some information on that 2 problems...
- Thank you very much, 73, cul
-
- Michael Hartje, DK5HH, (MAMI@DHVRRZN1)
-