NetNews Usenet Archive 1992 #20/NN_1992_20.iso/spool/comp/unix/question/10808

Internet Message Format | 1992-09-08 | 110.6 KB

Path: sparky!uunet!haven.umd.edu!mimsy!afterlife!adm!news
From: [e-mail] (SMTP MAILER)
Newsgroups: comp.unix.questions
Subject: Mail Delivery Problem
Message-ID: <[e-mail]>
Date: 7 Sep 92 21:31:20 GMT
Sender: [e-mail]
Lines: 3104


 ----Reason for mail failure follows----
Sending mail to recipient(s) woodfordm :
  Couldn't make final delivery.

 ----Transcript of message follows----
Received: from gw1.hanscom.af.mil by vd1.hanscom.af.mil with SMTP ;
          Mon,  7 Sep 92 14:18:38 EST
Date: 6 Sep 92 02:12:00 EST
From: [e-mail]
Subject: INFO-UNIX Digest  V15#141
To: "woodfordm" <[e-mail]>

Return-Path: <[e-mail]>
Received: from SEM.BRL.MIL by gw1.hanscom.af.mil with SMTP ;
          Sun,  6 Sep 92 02:10:55 EST
Received: by SEM.BRL.MIL id aa16373; 5 Sep 92 16:48 EDT
Received: from SEM.BRL.MIL by SEM.BRL.MIL id ab16025; 5 Sep 92 15:33 EDT
Received: from sem.brl.mil by SEM.BRL.MIL id aa16016; 5 Sep 92 15:16 EDT
Date:       Sat, 05 Sep 92 15:16:01 EST
From:       The Moderator (Mike Muuss) <[e-mail]>
To:         [e-mail]
Reply-To:   [e-mail]
Subject:    INFO-UNIX Digest  V15#141
Message-ID:  <[e-mail]>

INFO-UNIX Digest          Sat, 05 Sep 1992              V15#141

Today's Topics:
                    Re: Shell Scripts vs. C programs
                         kermit, where to ask?
                              RCS and make
                  Re: How to prevent a large core-dump
                 Re: reliable signals under BSD / SVR4
            Problem with "umount" on SCO-ODT (UNIX v. 5.3.2)
                  strange Digiboard problem (ISC UNIX)
                           Re: elm save file
                            awk with columns
                 evaluation of shell stuff in Makefiles
                          Re: terminfo capname
                      Re: Help: RCS from Purdue U?
                    Re: Fixing Corrupted Tar files?
                     Re: Need help from a make guru
             Re: need script to rename uppercase filenames
                         Problems using inittab
                         Extended chars in Unix
                      makefile with multiple dirs
                     Help wanted on hpterm problem
                       @ Re: Technical dictionary
                            a question on ls
                             Dumb Question
                           Re: Dear vi guru,
                             AWK and arrays
                      Question on WAKE UP call...
            What happens when all the inodes are exhausted?
     Distinguish between numeic & nonumeric char strings in shell ?
                     Re: Awk: Beginner needs help!
Re: How can I allow access to a subdirectory without risking other files?
                            help with script
                  Modem number access from UNIX system
                     Re: changing name of core file
                             two questions
               net.views -- USL/BSDI lawsuit -- responses
    Re: Question: which is faster 'find -exec' or 'find | xargs' ??
                net.views -- which vendor is most open?
-----------------------------------------------------------------

From: Martin McCormick <[e-mail]>
Subject: Re: Shell Scripts vs. C programs
Keywords: shell script, C
Date: 1 Sep 92 17:50:31 GMT
Sender: USENET News System <[e-mail]>
To:       [e-mail]


I always thought you should code your program in C because it executes so
much faster.  Since I am a beginner at C, I always code in shell scripts,
when possible, but that is mainly because it is easier to read, later.
I would like to see the answers to this question posted to the group.

Martin McCormick WB5AGZ   Stillwater, OK
O.S.U. Computer Center Data Communications Group

-----------------------------

From: Ray Jones <[e-mail]>
Subject: Re: Shell Scripts vs. C programs
Keywords: shell script, C
Date: 1 Sep 92 18:31:07 GMT
To:       [e-mail]

In <[e-mail]> [e-mail] (Ajay Shah) writes:

>nurban@tjhsst () writes:

>>I've heard that if at all possible, code a program as a shell script
>>rather than coding it in C.  Can someone explain the rationale behind
>>this philosophy?

>It's not so much a rule as a additional design alternative unique to
>the Unix environment.

>Shell scripts are frequently easy to write (esp. using sed/awk glue).
>They help in quickly getting a small system running.  They are
>tremendously powerful in automating mundane tasks.  In situations
>where you know you will throw away the code in a few weeks, the
>shell script approach is almost always better.

>Generally shellscripts are slower than a hand-tuned C program.
>When the script gets too large, it's easier to debug C instead
>of sh/sed/awk.  There are better debugging tools for C (a
>weakness of the sh/sed/awk paradigm which sounds fixable to me).

I agree with most of what you said, however, in some cases a shellscript
is actualy faster (not often, but sometimes).  They are much smaller in
terms of storage space.  As to debugging tools, try "sh -x scriptname".
This will show you every command as it execuites WITH THE SHELL
SUBSTITUIONS as they occure.  If this produces too much information, put
set -x
lines of code under question
set +x  # to turn off debuggingh


--
Onager Systems             Ray A. Jones
18710 NE 59th Ct.        UUCP .....uunet!camco!onager!ray
# 2053                   [e-mail]
Redmond, WA 98052               206-885-3568

-----------------------------

From: Don Steiny <[e-mail]>
Subject: Re: Shell Scripts vs. C programs
Keywords: shell script, C
Date: 2 Sep 92 20:23:56 GMT
To:       [e-mail]

[e-mail] (Stephen Harris) writes:

>In article <[e-mail]> [e-mail] (Don Steiny) writes:
>>    BZZZ - shell scripts are non-portable, there is no error checking,
>>and they have many other problems.    For very short programs shell
>>might be better, but in general C is better for many reasons.

>Of course!  Thats why the 1000+ lines of ksh script I have written and is
>running on three different systems (SunOs4.1.1, Convergent S/80 Ctix, Altos
>3068 SVR2.2) and used to run under Ultrix 4.1 was so hard to get working....
>NOT!

    This may be a matter of scale, but 1000 lines is not a very large
program.     Besides, there are still many systems that do not have ksh,
unfortunately, I have to use them all the time.   Even on Sun's, ksh is
a "local" program and is not always there.

    Not only that, but just doing line counts on programs tells very
little.

    Three systems is also not very many systems.    I ported AIM Resource
Accounting to dozens of vastly different systems.     If you are talking
about little programs of 1000 lines and a couple of systems, then it is
a different kettle of fish than 20,000+ line program that run on dozens of
systems.     Ask Larry Wall, the author of perl and rn.   He wrote perl
partially because of the portablity problems with the shell.

-don
--
Don Steiny
Don Steiny Software
Santa Cruz, CA 95060
[phone]

-----------------------------

From: Bryan Curnutt <[e-mail]>
Subject: Re: Shell Scripts vs. C programs
Keywords: shell script, C
Date: 2 Sep 92 21:28:04 GMT
To:       [e-mail]

In article <[e-mail]> nurban@tjhsst () writes:
>
>I've heard that if at all possible, code a program as a shell script
>rather than coding it in C.  Can someone explain the rationale behind
>this philosophy?

One rationale I haven't seen mentioned yet in this thread:  With a
shell script, the source code and the executable program are in the
same place.  It will be much easier for someone working on the system
several years down the road to find the source code, and to be sure
that the source code s/he has found is actually the code that is being
used in the executable program.
--
Any opinions above are mine, and do not necessarily reflect the views of SAI.
Bryan Curnutt
[e-mail]

-----------------------------

From: Christopher Davis <[e-mail]>
Subject: Re: Shell Scripts vs. C programs
Date: 3 Sep 92 04:33:23 GMT
Sender: NNTP News Poster <[e-mail]>
Nntp-Posting-Host: loiosh.eff.org
To:       [e-mail]

BC> == Bryan Curnutt <[e-mail]>

 BC> One rationale I haven't seen mentioned yet in this thread: With a
 BC> shell script, the source code and the executable program are in the
 BC> same place.

There are even perl scripts that are their own manual page nroff
sources, as well as the source and executable.  Now *that* is
integration...
--
Christopher Davis * [e-mail] * System Administrator, EFF * [phone]
            ``Ed Gruberman, you fail to grasp Ti Kwan Leep.
            Approach me that you might see.'' -- The Master

-----------------------------

From: John Navarra <[e-mail]>
Subject: Re: Shell Scripts vs. C programs
Keywords: shell script, C
Date: 3 Sep 92 07:20:33 GMT
Sender: "Usenet on news.acns" <[e-mail]>
To:       [e-mail]

In article <[e-mail]> [e-mail] (Martin McCormick) writes:
>
>I always thought you should code your program in C because it executes so
>much faster.  Since I am a beginner at C, I always code in shell scripts,
>when possible, but that is mainly because it is easier to read, later.
>I would like to see the answers to this question posted to the group.
>
    I think you will find, Martin, that many times most of the work
you do falls into fairly simple catagories:

        1) work aimed at solving an immediate problem
        2) work aimed at solving a recurring problem numerous times
        3) work you have time to sugar-coat.

I find that most of the work I do falls into the first two catagories.
In the first, I want a quick fix; I write a quick shell script; I disgard
it later. It would have taken me longer to write the C code. In the
second, I usually find it much quicker and easier to write and maintain
the shell script even though I may use it many times. When the combined
execution times of the shell script outdistance the time it takes to
write the equivalent C program , then I might have lost ground. However,
I find that once I have a working solution I am not so inclined to
rewriting that solution because other problems would be beating on my
door.

That brings us to the last case. And sure, if you have time to sugar-coat,
you probably have a life! (you know: a wife, kid, dog, car, home, credit
rating, etc etc.) Time to live the good life! ;-)


-tms
--
You can get further with a kind word | You can get further with a kind word
and a gun than a kind word alone.    | and a phaser than a kind word and a gun.
          --al capone                |           -- John Navarra
=======From the Lab of the MaD [e-mail]========

-----------------------------

From: Dr Ulrich Lauther <[e-mail]>
Subject: kermit, where to ask?
Date: 2 Sep 92 08:33:12 GMT
To:       [e-mail]

I have some problems with ckermit running under SCO XENIX/UNIX and Coherent,
running on 286 and 386 PCs, respectively.
What is the correct newsgroup (or maybe individual) to ask for help?

Thanks for help.

    -ulrich
[nosave]
 ----------------------------------------------------------------------------
Ulrich Lauther          ph: [phone] fx: ... 636 44950
Siemens ZFE BT SE 14    Internet: [e-mail]
                        from ems: lauther@ztivax:tcp-636-18:Mch P Siemens AG

-----------------------------

From: John VanHoozer <[e-mail]>
Subject: RCS and make
Date: 2 Sep 92 16:31:53 GMT
Sender: [e-mail]
Nntp-Posting-Host: titania
To:       [e-mail]

Greetings,

I was wondering if anyone had a reasonable method of using
RCS with the default make program (mine: SUN make v4.1)?

If my guess is correct,  it can't be done very easily.  I know
that RCS will work with GNU make but this may not be an option.


                jc

--
___________________________________________________________________________
John C. VanHoozer                 |
OptImage Interactive Services Company, L.P.  |  uunet!{optimage,aimla}!jc
11040 Santa Monica Blvd., Suite #300         |  Fax:  [phone]
Los Angeles, Ca. 90025                 |  Phone:[phone]
___________________________________________________________________________
Disclaimer:
   Opinions,  yew doone neeed my steenkeeng Opinions.

Carpe Diem:  A carp a day, that's all we ask.

-----------------------------

From: Lezz Giles <[e-mail]>
Subject: Re: How to prevent a large core-dump
Date: 2 Sep 92 17:52:59 GMT
Sender: USENET News System <[e-mail]>
Nntp-Posting-Host: fenric.dev.cdx.mot.com
To:       [e-mail]

In article <[e-mail]>, [e-mail] (Kenneth Herron) writes:
|>[e-mail] (Mark Brader) writes:
|>
|>>>     echo > core
|>>>     chmod 0 core
|>>> Then the program will not produce a core, since the file "core" already
|>>> exists, but is unwritable (see signal(2)).
|>
|>>Fine for most purposes, probably including the original query, but wait
|>>until the day that it's a setuid-root program whose core dumps you want
|>>to prevent!
|>>    mkdir core

I can't believe that nobody else has mentioned another solution yet....

ln -s /dev/null core

This won't work on all unixes since not everybody has symbolic links,
but it has the advantage that it works for setuid-root programs and it
will be reasonably obvious in the future to a unix-literate person
why the 'core' file was created.

BTW, I can't take credit for this solution - it is simply what we have
at our site and was thought up by some genius in the support dept.

lezz "The new infinite write-only memory device - /dev/null" Giles

-----------------------------

From: Robert Earl <[e-mail]>
Subject: Re: How to prevent a large core-dump
Date: 3 Sep 92 04:42:37 GMT
To:       [e-mail]

[e-mail] (Lezz Giles) writes:

| In article <[e-mail]>, [e-mail] (Kenneth Herron) writes:
| |>[e-mail] (Mark Brader) writes:
| |>
| |>>>     echo > core

Hmm, someone needs to look at the enlightening discussion of "how to
truncate a file" elsewhere on comp.unix.*.

| lezz "The new infinite write-only memory device - /dev/null" Giles

And I have bad news for you, /dev/null is readable too, and there are
lots of valid applications for feeding a program a persistent EOF.


--
"If you've got a pig that likes jumping fences, | robert earl
you have to make its fence a lot higher all at  | [e-mail]
once -- if you do it by increments, all you're  | [e-mail]
doing is training a jumping pig." -Henry Spencer|

-----------------------------

From: Terry Lambert <[e-mail]>
Subject: Re: reliable signals under BSD / SVR4
Keywords: signals, reliable, software interrupts
Date: 2 Sep 92 19:42:18 GMT
Sender: NetNews <[e-mail]>
Nntp-Posting-Host: thisbe.eng.sandy.novell.com
To:       [e-mail]

In article <[e-mail]> [e-mail] (Peter Wemm) writes:
>BSD/POSIX/SVR4 signals to not automatically "reset" to default on
>actication, and the signal is usually *masked* to prevent a recurring
>signal from within the handler.  The mask is cleared when the handler
>returns.  I dont seem to remember anything about reliable stacking or
>queueing of signals in this model.

    The POSIX 1003.4 draft currently includes queued signals, such that
signals are treated as discrete "1 signal/1 handler invocation" events.


                    Terry Lambert
                    [e-mail]
                    [e-mail]

---
Disclaimer:  Any opinions in this posting are my own and not those of
my present or previous employers.

-----------------------------

From: "W. Richard Stevens" <[e-mail]>
Subject: Re: reliable signals under BSD / SVR4
Keywords: signals, reliable, software interrupts
Date: 3 Sep 92 23:17:07 GMT
Sender: [e-mail]
Nntp-Posting-Host: gemini.tuc.noao.edu
To:       [e-mail]

>Has anyone investigated other models for signal delivery?

In the SVR4 sigaction(2) man page under SA_SIGINFO it says that the
signals are "reliably queued."  Apparently SVR4 can provide queued
signals, but someone-in-the-know told me that Sun made AT&T take this
feature out (when they were working on SVR4 together).  Perhaps POSIX.4
will have them put it back in ...

>Come to think of it, doesn't it make it somewhat "tricky" to use
> async. I/O at all?

I've yet to find a real application that actually uses BSD's SIGIO.
Has anyone seen one?

    Rich Stevens  ([e-mail])

-----------------------------

From: "Antony A. Courtney" <[e-mail]>
Subject: Re: reliable signals under BSD / SVR4
Keywords: signals, reliable, software interrupts
Date: 4 Sep 92 08:55:57 GMT
Sender: "NN required at ashe.cs.tcd.ie" <[e-mail]>
Nntp-Posting-Host: unix1.tcd.ie
To:       [e-mail]

In <[e-mail]> [e-mail] (W. Richard Stevens) writes:

>>Come to think of it, doesn't it make it somewhat "tricky" to use
>> async. I/O at all?

>I've yet to find a real application that actually uses BSD's SIGIO.
>Has anyone seen one?

Sure.  Back when I was at LBL, we had an animation server which I ported to X.
When you hit the "play" button on the interface, it would just render frames
as fast as it could.  I didn't want the overhead of a "check for any activity
on my UI" inside this loop, so I marked the socket connected to the X server
for asynchronous I/O.  That way it would actually stop when you hit the "stop"
button on the UI.  This is also applicable to any system which passes control
to some subroutine library for what could be "a long time" (e.g. doing a real
long image transformation of some sort).

The impression one gets from reading the docs on BSD's SIGIO is that it can
be used to write programs using interrupt-driven i/o.  IMHO, if no "real"
programs use SIGIO it shouldn't be provided.  But if it is provided, it should
work as advertised.

(begin long aside)
That said, it can be argued that any library which forces the programmer to
relinquish the flow of control to the library is badly designed.  Obviously
there can be only one main dispatch loop in a single-threaded program, so it is
at best quite difficult to interwork two libraries which want to provide the
main dispatch loop for the programmmer.  (e.g.  using an X toolkit in
conjunction with RPC) Further, such centralized dispatch loops use an
"event-callback" mechanism which clouds the natural control flow that
characterises the program.  This means that every "callback" function has to
check any data structure it operates on for consistency and validity.  For
instance, if I have a callback for a "zoom" button in a program that manipulates
images, my callback function has to check that the image data structure is
consistent, since it doesn't have any way of knowing whether or not the "load"
button has been pressed to load an image.  In a more conventional program, the
flow of control is governed by the program structure, and the abstract flow
graph it defines would never let a "zoom" function be called before a "load"
function.
(end long aside)

>    Rich Stevens  ([e-mail])

    -antony
--
********************************************************************************
* Antony A. Courtney                              Email: [e-mail] *
* Computer Science Department                            [e-mail] *
* Trinity College, Dublin, Ireland                Phone: 01+353+1-607389       *

-----------------------------

From: "Antony A. Courtney" <[e-mail]>
Subject: Re: reliable signals under BSD / SVR4
Keywords: signals, reliable, software interrupts
Date: 4 Sep 92 10:29:26 GMT
Sender: "NN required at ashe.cs.tcd.ie" <[e-mail]>
Nntp-Posting-Host: unix1.tcd.ie
To:       [e-mail]

In <[e-mail]> [e-mail] (Terry Lambert) writes:

>In article <[e-mail]> [e-mail] (Antony A. Courtney) writes:
>>There are some "hacks" to get around this, if you're environment lets you
>>write the centralized control loop for the program.  But usually SIGIO is
>>useful in cases where that isn't possible, e.g. when using an X toolkit.
>
>    Wrong.  In XTMainAppLoop() in the X toolkit, there are hooks for
>adding input sources to be "select()"ed on.
> [...]

    I didn't assert that it was totally impossible, just that it was a
"hack".  Adding extra input sources before passing control to XTMainAppLoop()
is a bit like saying "take over the control flow and tell me when something
happens that you understand (X events).  Oh, and also please tell me about
anything else that happens on these descriptors, even though you have no idea
what they are for and you don't know how to deal with them."  I humbly suggest
that this is hardly what one would call "elegant".

>    Read the fine manual.

    I have read the fine manual on select() for my Unix system.  Too bad X
toolkits hide "all that underlying system control flow stuff" from me...

>                    Terry Lambert
>                    [e-mail]

    -antony

--
********************************************************************************
* Antony A. Courtney                              Email: [e-mail] *
* Computer Science Department                            [e-mail] *
* Trinity College, Dublin, Ireland                Phone: 01+353+1-607389       *

-----------------------------

From: Jim Reid <[e-mail]>
Subject: Re: reliable signals under BSD / SVR4
Date: 4 Sep 92 12:04:31 GMT
Sender: [e-mail]
Nntp-Posting-Host: hunter
To:       [e-mail]

In article <[e-mail]> [e-mail] (Antony A. Courtney) writes:

   Has anyone investigated other models for signal delivery?  Since signals are
   meant to be analogous to interrupts, wouldn't it be useful to guarantee that
   all signals sent are processed?

Not always. For instance, how could a process handle a second KILL or
STOP signal sent to it? Or why would someone want to know how many
times a process has been sent a signal that it was ignoring?

In the Good Old Days, signals had one simple function to perform. This
was to inform a process that Something Bad had happened - memory
violation, arithmetic exception, etc - so that the process could do
some cleanup before it terminated. Everything that has been done to
signals since then has added complexity and features that the
mechanism was never originally designed or intended for.

        Jim

-----------------------------

From: Gokhale Rajendra V  <[e-mail]>
Subject: Problem with "umount" on SCO-ODT (UNIX v. 5.3.2)
Date: 2 Sep 92 20:27:16 GMT
Sender: news dummy user <[e-mail]>
Followup-To: poster
To:       [e-mail]

We have an 80486 machine running SCO-ODT (UNIX v.5.3.2).  We have two
hard discs, one containing the root partition while the other (designated
as /dev/user) is mounted as /usr2.  The problem is that we are unable to
unmount the latter.  umount complains "/dev/user busy" although the
file system is not really busy.  This behaviour is observed with
unfailing regularity.
As a result, we can never  do a normal shutdown of the machine.
Can anyone help?

Thanks,
Rajendra Gokhale,
ADA project,
IIT Bombay.

To indicate the state of the machine here is some (possibly relevant) info.

 ---------------------------------------------------------------------------
An " fsck -n /dev/user " says
"File system not modified"
I suppose this indicates that the file system  is corrupted but that it
cannot be repaired because of the -n flag.
 ----------------------------------------------------------------------------

 ----------------------------------------------------------------------------
Any normal file related actions (creating, deleting, copying,
searching files/directories ) get done without any problems whatsoever.
 ----------------------------------------------------------------------------

 ----------------------------------------------------------------------------
The edited output of ps -el (only process names retained) on a typical
occasion was :
   bdflush biod cpd cron getty inetd init lckclnt lockd logger lpsched mountd
   nfsclnt nfsd portmap ps rexd routed rwhod sched sed sh slink statd strerr
   vhand
No user other than root was logged in.  Root was logged in the root
directory when this output was obtained.
 ----------------------------------------------------------------------------

 ----------------------------------------------------------------------------
This is the file /etc/default/filesys on our machine :

bdev=/dev/root cdev=/dev/rroot \
    mountdir=/ mount=no fstyp=AFS \
    fsck=no fsckflags= \
    desc="The root filesystem" \
    rcmount=no rcfsck=no mountflags=
bdev=/dev/user cdev=/dev/rusr2 \
    mountdir=/usr2 mount=yes fstyp=S51K \
    fsck=no fsckflags= \
    desc="The usr2 filesystem" \
    rcmount=yes rcfsck=no mountflags=

-----------------------------

From: Todd Ogasawara <[e-mail]>
Subject: strange Digiboard problem (ISC UNIX)
Date: 3 Sep 92 02:15:26 GMT
Sender: News Service <[e-mail]>
Nntp-Posting-Host: uhunix.uhcc.hawaii.edu
To:       [e-mail]

I have a 386 box that's been running ISC UNIX 2.2 (SVR3) for nearly two
years now. It has a Digiboard Digichannel PC/16i (16-port serial board)
that has an unusual problem every now and then.

Every now and then (two or three times since the box was first configured)
accessing one of the ports either directly or through a dial-up results in
the following message during a login.

    Warning no access to tty (not a typewriter)
    Thus no job control in this shell

I found that the "not a typewriter" message is coming from /bin/login and
the rest of the warning message is generated by tcsh (/bin/csh will
generate a similar message). I have never been able to figure out what this
warning message means or what to do about it short of rebooting the system.

A few more clues: (1) "ps -ef" will not show an attached tty for processes
run from a session connected to the "misbehaving" port. (2) /etc/whodo
has the same problem (no tty reported attached to a process on that port).
(3) tty *does* report the correct /dev/ttyxx port when logged into the
misbehaving port. (4) 'su' will not run if logged into the bad port.

Any advice, hints, etc. would be appreciated...todd
--
Todd Ogasawara, U. of Hawaii, Psychology Department
BITNET:        todd@uhunix
INTERNET:    [e-mail]

-----------------------------

From: John Navarra <[e-mail]>
Subject: Re: elm save file
Date: 3 Sep 92 07:26:15 GMT
Sender: "Usenet on news.acns" <[e-mail]>
To:       [e-mail]

In article <[e-mail]> [e-mail] (Jeff Bowyer) writes:
>Gordon:
>
>When I set the following variables in elmrc
>
>     savemail = $HOME/axe    # for example
>     savename = ON
>
>then enter the 's' command, elm displays the login name of the
>current mail message, not the savemail directory, as I had hoped.
>
    I don't think the elmrc file likes the $HOME variable. I use
'~' in my elmrc. Try that. Otherwise, use the full path name.

-tms


--
You can get further with a kind word | You can get further with a kind word
and a gun than a kind word alone.    | and a phaser than a kind word and a gun.
          --al capone                |           -- John Navarra
=======From the Lab of the MaD [e-mail]========

-----------------------------

From: Esa Karell <[e-mail]>
Subject: awk with columns
Date: 3 Sep 92 06:30:46 GMT
To:       [e-mail]

Could someone tell me what would be the easisest
way to accomplish this with awk.

Input:

X    Y  T        P I
1    2  3        4 5
 .    .  .        . .

I'd need to swap the values of Y & T so output should look:

X    Y  T        P I
1    3  2        4 5
 .    .  .        . .

All the blanks between the columns should remain the same.
Thx,

Esa
--

Esa Karell [e-mail]
Tel. [phone]

-----------------------------

From: Luke Mewburn <[e-mail]>
Subject: evaluation of shell stuff in Makefiles
Date: 3 Sep 92 06:42:48 GMT
To:       [e-mail]


I have this slight problem with a Makefile. I have a shell expression
which I use to extract the version number of the program from a header
file, but make evaluates this expression everywhere the variable is
used, which slows things down a bit... Is there a way I can get make
to evaluate it once???

An extract of the makefile is below. The output of make tar is something
like: (wrapped for clarity)

--
    mkdir flon-`/bin/sed -ne 's/.*VERSION."\(.*\)".*/\1/p;/VERSION/q'
lon.h`
    cp Makefile lon.h getfile.c main.c init.c print.c flon.nro
Installation eg.friends tags Notes flon-`/bin/sed -ne
's/.*VERSION."\(.*\)".*/\1/p;/VERSION/q' lon.h`
    tar -cf flon-`/bin/sed -ne 's/.*VERSION."\(.*\)".*/\1/p;/VERSION/q'
lon.h`.tar flon-`/bin/sed -ne 's/.*VERSION."\(.*\)".*/\1/p;/VERSION/q' lon.h`
    compress flon-`/bin/sed -ne 's/.*VERSION."\(.*\)".*/\1/p;/VERSION/q'
lon.h`.tar
    rm -rf flon-`/bin/sed -ne 's/.*VERSION."\(.*\)".*/\1/p;/VERSION/q'
lon.h`
--

As you can see, this would be slow (and messy too :). So, any suggestions
how to clean this up?


Thanx in advance,
Luke.


-- extract from Makefile --
#[...]

#
# Version number
#
VERSION=`/bin/sed -ne 's/.*VERSION."\(.*\)".*/\1/p;/VERSION/q' lon.h`

#
# Executable name
#
EXEC=flon

#
# Archive name
#
ARCNAM=$(EXEC)-$(VERSION)

# [...]
tar:
    mkdir $(ARCNAM)
    cp $(ALLF) $(ARCNAM)
    tar -cf $(ARCNAM).tar $(ARCNAM)
    compress $(ARCNAM).tar
    rm -rf $(ARCNAM)

shar:
    mkdir $(ARCNAM)
    cp $(ALLF) $(ARCNAM)
    shar $(ARCNAM) > $(ARCNAM).shar
    rm -rf $(ARCNAM)
-- end extract --

-----------------------------

From: David Bath <[e-mail]>
Subject: Re: terminfo capname
Date: 3 Sep 92 06:58:55 GMT
Sender: Njuiz noveles nova newes <[e-mail]>
To:       [e-mail]

[e-mail] writes:

>What is the byte sequence for terminfo capnames 'cnorm', 'cvvis' and 'civis' for
>'ibm-pc' and 'vt100'?
i
It depends
cnorm is probably sgr0 ie \033[0m
civis is probably \033[8m
cvvis is probably \033[1m or \033[7m

========================================================================
David T. Bath
email:     [e-mail]
Work:     Global Technology Corporation, Carlton, Vic, OZ  [phone]
=======================================================================
--
========================================================================
David T. Bath
email:     [e-mail]
Work:     Global Technology Corporation, Carlton, Vic, OZ  [phone]

-----------------------------

From: David Bath <[e-mail]>
Subject: Re: Help: RCS from Purdue U?
Date: 3 Sep 92 07:01:13 GMT
Sender: Njuiz noveles nova newes <[e-mail]>
To:       [e-mail]

[e-mail] (Kuryan "Obi" Thomas) writes:

>In article <[e-mail]>,
><[e-mail]> writes:
>> Can anyone tell me about a configuration management program for UNIX called RCS
>> (Revision Control System).  I've heard that it is a shareware product that
>> recently got an upgrade to V5.7.  Please respond.

YEP - beats SCCS hands down for ease of use.
I've only seen 5.6
Look at symbolic version numbers !
Note : CVS sits on top of RCS - you may want that as well !
========================================================================
David T. Bath
email:     [e-mail]
Work:     Global Technology Corporation, Carlton, Vic, OZ  [phone]
=======================================================================
--
========================================================================
David T. Bath
email:     [e-mail]
Work:     Global Technology Corporation, Carlton, Vic, OZ  [phone]

-----------------------------

From: David Bath <[e-mail]>
Subject: Re: Fixing Corrupted Tar files?
Keywords: tar,corrupt
Date: 3 Sep 92 07:03:09 GMT
Sender: Njuiz noveles nova newes <[e-mail]>
To:       [e-mail]

[e-mail] (John Tillema) writes:

>Is there any way to do this? (fix corrupted tar files?)  Are
>there any utilties that will do this?

Try tarx from comp.sources.unix early volumes
--
========================================================================
David T. Bath
email:     [e-mail]
Work:     Global Technology Corporation, Carlton, Vic, OZ  [phone]

-----------------------------

From: Michael Salmon <[e-mail]>
Subject: Re: Need help from a make guru
Date: 3 Sep 92 07:18:21 GMT
Sender: [e-mail]
Nntp-Posting-Host: eos6c02.ericsson.se
To:       [e-mail]

In article <[e-mail]>, [e-mail] (alan dare) writes:
|>
|> In our application we wanted to keep the object and source separate from
|> each other in a directory structure as shown below:
|>
|>             <topdir>
|>                |
|>               / \
|>                obj  src
|>
|> My current makefile does this quite well. It also doesn't require a separate
|> condition for each source. Problem: Several users add modules to a
|> executive then compile it. It would would be nice if they didn't have to
|> modify the makefile. The makefile would just compile any source file in the
|> current directory. I have taken several stabs at it, but I am not having
|> any luck. All I need for it to work is a variable containing a list of
|> the source and a variable containing the objects with the directory path
|> "../obj/" prefixed to it.  Below is a makefile that would compile all *.c and
|> create a a.out file. However, I can't get make to execute the shell command
|> correctly on the OBJECTS line correctly (the line is a valid shell command).
|>
|> OBJDIR = ../obj
|>
|> SOURCE=*.c
|> OBJECTS=`for i in \`ls *.c |sed 's/\.c/.o/g'\`; do echo "../obj/$i"; done`
|> $(OBJDIR)/a.out:  $(OBJECTS)
|>     cc -o $@ $(OBJECTS)
|>
|> $(OBJECTS) : $SOURCE
|>     cc -c $(OPT)  $(@F:o=c) -o $@
|>
|> Can anybody tell me how I can get the OBJECT variable set correctly?

From the man page for make under the heading macros (SunOS 4.1.1):

     :sh =
          Define the value of a macro to be the output of a  com-
          mand (see Command Substitutions, below).

However I think that using % name matching rules is better than this
approach, particularly if you also want to use SCCS or RCS.

--

Michael Salmon

#include    <standard.disclaimer>
#include    <witty.saying>
#include    <fancy.pseudo.graphics>

Ericsson Telecom AB
Stockholm

-----------------------------

From: Alan Carter <[e-mail]>
Subject: Re: Need help from a make guru
Date: 3 Sep 92 10:55:54 GMT
Sender: News Administrator <[e-mail]>
Nntp-Posting-Host: bmdhh298
To:       [e-mail]

In article <[e-mail]>, [e-mail] (alan dare) writes:
|>
|> In our application we wanted to keep the object and source separate from
|> each other in a directory structure as shown below:
|>
|>             <topdir>
|>                |
|>               / \
|>            obj  src
|>
|> My current makefile does this quite well. It also doesn't require a separate
|> condition for each source. Problem: Several users add modules to a
|> executive then compile it. It would would be nice if they didn't have to
|> modify the makefile. The makefile would just compile any source file in the
|> current directory. I have taken several stabs at it, but I am not having
|> any luck. All I need for it to work is a variable containing a list of
|> the source and a variable containing the objects with the directory path
|> "../obj/" prefixed to it.  Below is a makefile that would compile all *.c and
|> create a a.out file. However, I can't get make to execute the shell command
|> correctly on the OBJECTS line correctly (the line is a valid shell command).
|>
|> OBJDIR = ../obj
|>
|> SOURCE=*.c
|> OBJECTS=`for i in \`ls *.c |sed 's/\.c/.o/g'\`; do echo "../obj/$i"; done`
|> $(OBJDIR)/a.out:  $(OBJECTS)
|>     cc -o $@ $(OBJECTS)
|>
|> $(OBJECTS) : $SOURCE
|>     cc -c $(OPT)  $(@F:o=c) -o $@
|>
|> Can anybody tell me how I can get the OBJECT variable set correctly?

Your problem is that you are trying to nest the stuff in `` command line
substitution braces. This is not allowed. Do it in two steps:

TEMP=`ls *.c |sed 's/\.c/.o/g'\`
OBJECTS=`for i in $TEMP; do echo "../obj/$i"; done`

   Alan

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
   Maidenhead itself is too snobby to be pleasant. It is the haunt of the
   river swell and his overdressed female companion. It is the town of showy
   hotels, patronized chiefly by dudes and ballet girls.

   Three Men In A Boat, Jerome K. Jerome, 1889
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

-----------------------------

From: Benjamin B Thomas <[e-mail]>
Subject: Re: need script to rename uppercase filenames
Date: 3 Sep 92 08:31:07 GMT
Sender: News system <[e-mail]>
Nntp-Posting-Host: e40-008-4.mit.edu
To:       [e-mail]

In article <[e-mail]> [e-mail] (Bill Miller) writes:
>
>I have some source code in DOS (many separate files) that I tarred under
>DOS and untarred under 386BSD.  The big problem is that all the files
>came through in UPPERCASE and I need a script to mv (rename) them all
>to lowercase quickly.

Try:

#!/bin/csh
foreach i ($argv)
setenv file $i
cat $file | tr -d '\015' > `echo foo | nawk '{print tolower(ENVIRON["file"])}'`
rm $file
end


     Now just type the script name and the filenames (or partial filenames
with wildcards) and you should be set.

                    -Benjy

-----------------------------

From: Rouben Rostamian <[e-mail]>
Subject: Re: need script to rename uppercase filenames
Date: 3 Sep 92 12:06:07 GMT
Sender: News posting account <[e-mail]>
To:       [e-mail]

In article <[e-mail]> [e-mail] (Benjamin B Thomas) writes:
>In article <[e-mail]> [e-mail] (Bill Miller) writes:
>>
>>I have some source code in DOS (many separate files) that I tarred under
>>DOS and untarred under 386BSD.  The big problem is that all the files
>>came through in UPPERCASE and I need a script to mv (rename) them all
>>to lowercase quickly.
>
>Try:
>
>#!/bin/csh
>foreach i ($argv)
>setenv file $i
>cat $file | tr -d '\015' > `echo foo | nawk '{print tolower(ENVIRON["file"])}'`
>rm $file
>end

I would suggest the quicker script:

#!/bin/sh
for file in *[A-Z]*
do
mv $file `echo $file | tr '[A-Z]' '[a-z]'`
done

This script changes all files names containing an uppercase letter
in the current directory to lowercase.  I would not strip the \015s
so casually without knowing the contents of the files.

--RR

-----------------------------

From: Samuel Kamens <[e-mail]>
Subject: Re: need script to rename uppercase filenames
Keywords: script,tar,msdos
Date: 3 Sep 92 13:50:42 GMT
Sender: [e-mail]
Nntp-Posting-Host: fork.bae.bellcore.com
To:       [e-mail]

In article <[e-mail]>, [e-mail] (Bill Miller) writes:
> Hi, everyone.
>
> I'm fairly new to unix, and I need a script or procedure to do the following:
>
> I have some source code in DOS (many separate files) that I tarred under
> DOS and untarred under 386BSD.  The big problem is that all the files
> came through in UPPERCASE and I need a script to mv (rename) them all
> to lowercase quickly.
>
> Since they're in DOS text format, I realize I also need to strip the
> extra carriage returns on each line.  I've been successful in doing this
> with:
>
>   cat (file) | tr -d '\015' > (newfile)
>
> It would be nice to combine both of these so that I could rename the
> files to uppercase and strip the extra newlines all in one fell swoop
> instead of doing it one file at a time.
>


Try this:

 -------------------- CUT HERE ----------------------
#!/bin/ksh
# Script to rename a list of files to its lowercase equivalent while
# stripping extra carriage returns
#

USAGE: $0 FILE1 FILE2 FILE3 ...

for i in $*
do
    lowfile=`echo $i | tr [A-Z] [a-z]`

    cat ${i} | tr -d '\015' > ${lowfile}
    # you could do rm $i here if you want
done
 -------------------- CUT HERE ----------------------

You could also do this with a one-liner:

    cat $1 | tr -d '\015' > `echo $1 | tr [A-Z] [a-z]`

Sam

--
Sam Kamens                    Bell Communications Research
[e-mail]                Phone: [phone]
444 Hoes Lane  Room RRC 1D-210
Piscataway, NJ 08854

-----------------------------

From: Bill Campbell <[e-mail]>
Subject: Re: need script to rename uppercase filenames
Keywords: script,tar,msdos
Date: 3 Sep 92 15:12:45 GMT
To:       [e-mail]

In <[e-mail]> [e-mail] (Bill Miller) writes:

>Hi, everyone.

>I'm fairly new to unix, and I need a script or procedure to do the following:
>
>I have some source code in DOS (many separate files) that I tarred under
>DOS and untarred under 386BSD.  The big problem is that all the files
>came through in UPPERCASE and I need a script to mv (rename) them all
>to lowercase quickly.
>
>Since they're in DOS text format, I realize I also need to strip the
>extra carriage returns on each line.  I've been successful in doing this
>with:
>
>  cat (file) | tr -d '\015' > (newfile)
>
>It would be nice to combine both of these so that I could rename the
>files to uppercase and strip the extra newlines all in one fell swoop
>instead of doing it one file at a time.

A simple perl solution (needs more error checking):

for $dos ( @ARGV ) {
    ( $unix = $dos ) =~ tr/A-Z/a-z/;    # convert to lower case
    if ( -f $unix ) {
        print STDERR "Cannot convert $dos -- $unix exists\n";
        next;
    }
    open(DOS, $dos);
    open(UNIX, "> $unix");
    while (<DOS>) {
        s/\s+$//;   # strip all trailing whitespace including ^M
        print UNIX;
    }
    close(DOS);
    unlink($dos); # remove old dos file
    close(UNIX);
}
--
INTERNET:  [e-mail]   Bill Campbell; Celestial Software
UUCP:   ...!thebes!camco!bill   6641 East Mercer Way
             uunet!camco!bill   Mercer Island, WA 98040; [phone]
SPEED COSTS MONEY -- HOW FAST DO YOU WANT TO GO?

-----------------------------

From: Michael Salmon <[e-mail]>
Subject: Re: need script to rename uppercase filenames
Keywords: script,tar,msdos
Date: 4 Sep 92 06:46:13 GMT
Sender: [e-mail]
Nntp-Posting-Host: eos6c02.ericsson.se
To:       [e-mail]

In article <[e-mail]>, [e-mail] (Bill Miller) writes:
|> Hi, everyone.
|>
|> I'm fairly new to unix, and I need a script or procedure to do the following:
|>
|> I have some source code in DOS (many separate files) that I tarred under
|> DOS and untarred under 386BSD.  The big problem is that all the files
|> came through in UPPERCASE and I need a script to mv (rename) them all
|> to lowercase quickly.
|>
|> Since they're in DOS text format, I realize I also need to strip the
|> extra carriage returns on each line.  I've been successful in doing this
|> with:
|>
|>   cat (file) | tr -d '\015' > (newfile)
|>
|> It would be nice to combine both of these so that I could rename the
|> files to uppercase and strip the extra newlines all in one fell swoop
|> instead of doing it one file at a time.
|>

A little plug for tcsh:

foreach fn ([A-Z]*)
    tr -d '\015' <$fn >$fn:al
end

Of course the original file name specification is a little simplistic.

--

Michael Salmon

#include    <standard.disclaimer>
#include    <witty.saying>
#include    <fancy.pseudo.graphics>

Ericsson Telecom AB
Stockholm

-----------------------------

From: Larry Wall <[e-mail]>
Subject: Re: need script to rename uppercase filenames
Keywords: script,tar,msdos
Date: 4 Sep 92 16:20:44 GMT
Sender: [e-mail]
Nntp-Posting-Host: scalpel.netlabs.com
To:       [e-mail]

In article <[e-mail]> [e-mail] (Bill Campbell) writes:
:         s/\s+$//;   # strip all trailing whitespace including ^M

Unfortunately, that also deletes the ^J.  Since the script unlinks the
DOS file, you've just thrown away information that's not easily
reconstructable.  (In fact, the files won't even be editable unless you
use GNU emacs.)  It would be safer to move the DOS file to a temporary
subdirectory until you're sure the translation works right.

I don't want to discourage anyone from helping people, but help is not
what booby traps are generally construed as...  :-)

Larry Wall
[e-mail]

-----------------------------

From: Benjamin B Thomas <[e-mail]>
Subject: Problems using inittab
Date: 3 Sep 92 08:49:41 GMT
Sender: News system <[e-mail]>
Nntp-Posting-Host: e40-008-4.mit.edu
To:       [e-mail]


    I was wondering if someone could help me clear something up?  The
documentation I have leads me to believe that if I want to run a script from
/etc/inittab at run level 3 and insure that it remains alive, I need to add
this to /etc/inittab:

foo :3:respawn:/bin/sh -c 'exec /usr/local/bin/my_script'

The script gets executed (ps gives:  ...../usr/local/bin/my_script /usr/local/)
but doesn't work right.

The script writes to a tmp file, sleeps, checks some stuff, and then writes
over the tmp file (ls -t > /tmp/foo) This works fine when I execute from the
command line, but won't write anything when started from inittab.However, it
will write a blank /tmp/foo if I place anything in the file while the process
is sleeping.

Any ideas?   BTW It's a csh script.

                    -Benjy

-----------------------------

From: Scott Barnhart <[e-mail]>
Subject: Extended chars in Unix
Date: 3 Sep 92 11:28:36 GMT
To:       [e-mail]



I am using Unix SystemVR4, and want to find a way to generate the
extended ASCII chars from the keyboard.  The reason for this is
to generate the appropriate foreign characters (such as the vowels
with the accents, etc) needed in the various different languages
(specifically French, German and other European languages).

I first thought it could be done somehow with the LOCALE stuff
in Unix, setlocale(), chrtbl(), etc.  But the man pages don't go
into how to generate or translate these characters.

If anyone has any idea how to do this sort of thing, please
e-mail me at [e-mail], or post a reply.
I would really appreciate any help in this area.

-----------------------------

From: David Pape <[e-mail]>
Subject: makefile with multiple dirs
Keywords: make,makefile
Date: 3 Sep 92 13:59:02 GMT
Sender: [e-mail]
To:       [e-mail]


        Heres my (highly simplified) project directory setup:

proto:  makefile, resultant executable

proto/x:        x.c, x.h, x.o
proto/y:        y.c, y.h, y.o

How can I run "make" in the "proto" dir, and have it put "x.o" and
"y.o" in the correct directories? I also want to be able to run
"makedepend" from the makefile in "proto" so that it creates the
dependencies correctly, i.e:

# DO NOT DELETE THIS LINE -- make depend depends on it.

proto/x/x.o: proto/x/x.h
proto/y/y.o: proto/y/y.h

Even when I have
 ---------------------------------
X_PATH = proto/x

depend:
        makedepend $(X_PATH)/x.c
 ---------------------------------
in my makefile, I always get

x.o: proto/x/x.h        #notice missing path

Any helpful hints out there??

--
David Pape                 | "Justice is inci- |            ///
[e-mail] |  dental to law    |     \\\ UNI_DO
Dortmund, Germany             |  and order"       |      \\\///
tel: [phone]             | - J. Edgar Hoover |       \\//

-----------------------------

From: Jack versloot <[e-mail]>
Subject: Help wanted on hpterm problem
Keywords: X, hpterm
Date: 3 Sep 92 14:58:22 GMT
Followup-To: poster
To:       [e-mail]


Help wanted on the following problem.

First some information:
Language: FORTRAN
Machine : HP9000
Terminal: hp

I have a program doing the following:
   - set up protected fields on the terminal
   - let the user enter some data in these fields
   - reading lines from the terminal (with the entered information) with:
        READ (*, '(A)') line

To read lines ESCd is sent to the terminal. The character*512 variable
line is searched for ESC[ (begin protected field) and ESC] (end
protected field). All information between these two codes will be the
information entered by the user.

This works FINE for a real hp-terminal. On a X-window with hpterm it
doesn't work however! The last field on each line is misformed: the
code for 'end protected field' is missing.

My question:
   Why?
and more important:
   What can I do about it?

Is hpterm not completely compatible with a hp-terminal or should I (maybe)
give hpterm some options or what?

If you have the answer please send it by E-mail to the
following address:

             [e-mail]

-----------------------------

From: Taed Nelson <[e-mail]>
Subject: @ Re: Technical dictionary
Date: 3 Sep 92 15:55:51 GMT
Sender: UseNet News account <[e-mail]>
To:       [e-mail]

In article <[e-mail]>, [e-mail] (Michael G. Bruce) writes:
> Does anyone know of a quality CS dictionary?  While I have a working knowledge
> of terms such as MIPS, FLOPS, IPC, etc,  I have occasion to need a proper,
> concise definition.

Although not "concise", I would highly recommend _The Encyclopedia of Computer
  Science_.  It's about 2000 pages, and has 1/2 - 4 page articles on
  everything about CS that you would ever want to know.  Particularly
  interesting is the history and details of the "classic" machines.  I don't
  know who publishes/ed it, but it goes for about $50 used.

-----------------------------

From: scott williams <[e-mail]>
Subject: a question on ls
Date: 3 Sep 92 17:50:42 GMT
To:       [e-mail]


In article <6786#tekig7.PEN.TEK.COM> [e-mail] (Naren Bala)
writes:

| >What command would I type to display all files that do not have a .Z
| >extension ?

In sh, my preference would be to combine ls with grep, as in:

   ls | grep -v '\.Z$'
---
 ~ DeLuxe} 1.12 #8626 ~ Scott Williams, Editor, Open Systems Letter
--
Canada Remote Systems  - Toronto, Ontario/Detroit, MI
World's Largest PCBOARD System - 416-629-7000/629-7044

-----------------------------

From: "Subramanian Venkataraman (TV" <[e-mail]>
MMDF-Warning:  Parse error in original version of preceding line at BRL.MIL
Subject: Dumb Question
Date: 3 Sep 92 18:56:42 GMT
Sender: Usenet <[e-mail]>
Nntp-Posting-Host: doowop.jpl.nasa.gov
To:       [e-mail]

Hello.

I have two dumb questions.

(i) I am given to understand that there exists a public domain
version of the BSD386 system (net2?). How do I get hold of the
source code?

(ii) Is there a SUN version of net2?

Please note that I am note a system guru, or even a system expert!
Just a novice trying to get hold of info that may be useful for a
project I am involved in.

Thanks a lot.

S.T. Venkataraman.
Jet Propulsion Laboratory
[e-mail]

-----------------------------

From: Afsar Husain <[e-mail]>
Subject: Re: Dear vi guru,
Date: 3 Sep 92 21:48:12 GMT
Sender: Usenet News <[e-mail]>
Nntp-Posting-Host: ad-22.naitc.com
To:       [e-mail]

In article <[e-mail]> [e-mail] (Brian D. Schieber) writes:
>Dear vi guru,
>
> I like to use tabs alot in just about every session of vi I
>work with, It would be great if I had a macro or whatever that
>could replace the tabs with the appropriate number of spaces
>in a file WHILE in vi. Do you have such a macro?
>
>Most graciously,
> Brian
>
>p.s. please send all suggestions to switch to EMACS to /dev/null

What I have done is to map a free key such as a "w" or a "q"
with a macro. This macro will replace all the tabs with 4 spaces
and then writes the file and quits the editor. You can include
such a mapped macro in the ".exrc" file

    map w :1,$s/     /    /g

The above map can be accomplished in a "vi" session by the following series of key
strokes. Ofcourse you have to substitute the "tab" character and the
"spaces" with the desired number of spaces.

    :map w :1,$s/tab/spaces/g^V^M:wq^V^M

The control M sequence can be generated by preceding it with
a control V i.e. ^V^M.


Hope this helps.

--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
#include    <standard.disclaimer>
email: [e-mail]

-----------------------------

From: Anthony P Scott <[e-mail]>
Subject: AWK and arrays
Date: 4 Sep 92 03:07:33 GMT
Sender: [e-mail]
Originator: aps583@huxley
To:       [e-mail]

Suppose I want to read a bunch of letters one to a line
and output the position in the alphabet that they occupy.

ie

a
b
c
d
e

would be output as

1
2
3
4
5

etc.

One way I visualized doing this is to write an array called alph[letter]
such as:

alph[a] =1
alph[b] = 2
alph[c] = 3
 ...

 ...
alph[z] = 26

and then read the lines of text as another array called charact[NR]

ie

{
charact[NR] = $1
}

and the use the charact array to subscript the alph array ie

{ for (i=1; i<=NR; i++) {
char_num[i] = alph[charact[i]] }
}

This doesn't work. Can anyone point me in the right direction

Another way I thought of doing the job was to get the subscript number of
an array item.
 Something like:

for (charact[i] in alph) {
    print alph[i]
}

This doesn't work either.

I hope this is not a silly question.

Regards

Tony Scott

-----------------------------

From: "M. P. Capistrano" <[e-mail]>
Subject: Question on WAKE UP call...
Date: 4 Sep 92 04:34:52 GMT
To:       [e-mail]


What should happen if the kernel attempts to wake up all processes sleeping
on an event, but no processes are asleep on the event at the time of the
wakeup?

Thanks.

-MPC-

-----------------------------

From: "M. P. Capistrano" <[e-mail]>
Subject: What happens when all the inodes are exhausted?
Date: 4 Sep 92 04:39:25 GMT
To:       [e-mail]


When a filesystem is created, a fixed number of inodes are allocated
for the inode list.  So, what happens when all the inodes are
exhausted then?

Thanks.
-MPC-

-----------------------------

From: "S.Ramakrishnan" <[e-mail]>
Subject: Distinguish between numeic & nonumeric char strings in shell ?
Date: 4 Sep 92 07:03:57 GMT
Sender: [e-mail]
To:       [e-mail]


 How does one distinguish between a number and a nonnumeric
 character using shelltools (sh/csh, awk, sed, expr, etc) ?
 I am trying to validate the IP address typed in by the user.
 I isolate the four components and check each one to see if lies
 within the legal range (0..255). In the event the user types, say,
 tom.jack.joe.jim, I want to be able to flag an error. However the
 IF statement, marked (+) in the fragment below, fails even if a 4
 component string such as "tom.jack.joe.jim"  is entered.
 Any suggestions ?

 Encl:
    #!/bin/sh
        ...
    while test "$ip" = ""
    do
          echo "    IP Address = "
      read ip
          if [ "$ip" != "" ]
      then
              cnt=`echo $ip | awk -F'.' '{ print NF }'`
              one=`echo $ip | awk -F'.' '{ print $1}'`
              two=`echo $ip | awk -F'.' '{ print $2}'`
              thr=`echo $ip | awk -F'.' '{ print $3}'`
              fou=`echo $ip | awk -F'.' '{ print $4}'`
#(+)
              if [ $cnt -ne 4 -o $one -gt 255 -o $two -gt 255 -o $thr -gt 255 -o $fou -gt 255 ]
              then
            echo ""
                echo "*** Invalid IP Address Format: Please Retype"
            echo ""
            ip=""
              fi
      fi
        done
    ...

---
S Ramakrishnan, CS Dept, McBryde Hall, VaTech, Virg.

-----------------------------

From: Michael Salmon <[e-mail]>
Subject: Re: Awk: Beginner needs help!
Keywords: awk unix
Date: 4 Sep 92 07:32:13 GMT
Sender: [e-mail]
Nntp-Posting-Host: eos6c02.ericsson.se
To:       [e-mail]

In article <[e-mail]>,
[e-mail] (Steven C. Velasco) writes:
|>
|> HI,
|> I am new to unix, and really new to awk, so hopefully somebody can
|> answer this question real easily.  I have hundreds of SAS program
|> files written for MVS, that we now need to run on UNIX.
|> The files from MVS look like this :
|>
|> COMMAND? list unnumbered
|> //MVS JCL statements
|> //MVS JCL statements
|> //MVS JCL statements
|> DATA _NULL_ ;
|>    SET SASLIB.DATASET ;
|> MORE SAS STATEMENTS
|> RUN ;
|>
|> I can get awk to print out the lines that don't contain the string  "//"
|> or "COMMAND?" in $1, but I would like to have
|> something that replaces the entire line  "COMMAND? list unnumbered"
|> (this is always the first line, and occurs just once)
|> with the string: "libname saslib /usr2/username/datasets"  or, to
|> somehow put the "libname" string in the first line of each file.
|> So the resulting file would look like this:
|>
|> libname saslib '/usr2/username/datasets' ;
|> DATA _NULL_ ;
|>    SET SASLIB.DATASET ;
|> MORE SAS STATEMENTS ;
|> RUN ;
|>
|> below is my feeble attempt to do this.  When I run this program, I
|> get a message saying something like 'awk: cant set $0 at record 1':
|>
|> $1 !~ /\/\// { if ( $0 ~ /COMMAND? list unnumbered/ )
|> $0 = "libname saslib /usr2/username/datasets ;"
|> print >FILENAME }
|>
|> What am I doing wrong?  I would appreciate any help or suggestions.

I would solve this in this way:

#!/usr/bin/awk
BEGIN { print "libname saslib /usr2/username/datasets ;" }

/^\/\// { next }
/^COMMAND?/ { next }
{ print }

The first line is only useful if you system supports this. I would also
guess that the COMMAND? line could be used for something useful but I
don't know anything about SAS other than that it is Sweden's
international airline ;^)

--

Michael Salmon

#include    <standard.disclaimer>
#include    <witty.saying>
#include    <fancy.pseudo.graphics>

Ericsson Telecom AB
Stockholm

-----------------------------

From: Andre Hoekstra <[e-mail]>
Subject: Re: How can I allow access to a subdirectory without risking other files?
Date: 4 Sep 92 12:58:22 GMT
Sender: [e-mail]
Nntp-Posting-Host: waal
To:       [e-mail]

In article <[e-mail]> [e-mail] (Leslie Mikesell) writes:
>In article <[e-mail]> [e-mail] writes:
>>I know how to give group/other access to a subdirectory without risking my
>>other files, however, I want people to be able to read/write to ANY file in
>>the subdirectory. That includes files put there by other people.  The only way
>>I know of to do this is to make sure that everyone has their umask set to 000
>>when they are working in that subdirectory.
>
>It sounds fairly dangerous to do this without any arbitration to keep
>several people from writing to the same file at once.  Why don't you
>just insist that each person copy a file before making changes?  Or
>write a suid root program that does nothing but change the modes on
>the files in that directory and make the users run that before writing
>to any files.  Better yet, make the program also perform some kind of
>file locking while each person is writing.

SCCS(1) ?
>
>Les Mikesell
>  [e-mail]

Andre Hoekstra

-----------------------------

From: Mark Fugazzotto <[e-mail]>
Subject: help with script
Date: 4 Sep 92 14:03:11 GMT
Sender: [e-mail]
Originator: [e-mail]
To:       [e-mail]


 A while ago I had asked for help with posting the
motd on the server to all remote workstations. I got
a big response back. Thanks to everyone that responded.
 Some suggetions were to use rdist(unsupported on this
version of Unix), use rcp (would have worked but a
security hazard to make root equivilent on from the
server to all workstations). The one I went with was
a script on the server to ftp  via anonymous the motd
to all workstations. then have a cron job on each
workstation copy the motd into the etc directory.
 This works well except for one flaw with cron I can't
seem to overcome. Cron simply executes a script on each
workstation at the specified times, and it does. Only it
does not echo to the screen the line in the script which
tells the user that an updated motd has arrived. Where is
this output being sent to? If I execute the script manually
it works fine. What can I do to get cron to echo the line to
the screen?
 Also while I'm here, Is there something I can put into my
script that will check first for a logon, if one is found then
echo to the screen, if one is not found, do nothing??

                          thanks in advance

                                             Mark...

-----------------------------

From: Michael Gleibman <[e-mail]>
Subject: Modem number access from UNIX system
Date: 4 Sep 92 15:15:49 GMT
Sender: [e-mail]
To:       [e-mail]

Hello!
Does anybody here knows how can I access the modem number from VAX (BSD4.3)
Thanx in advance. Any help is appreciated.

--
                        Michael Gleibman,
Dept. of Mathematics, Technion, Haifa, Israel,
e-mail: [e-mail]
Phone: [phone]

-----------------------------

From: Brian Fitzgerald <[e-mail]>
Subject: Re: changing name of core file
Date: 4 Sep 92 15:26:09 GMT
Nntp-Posting-Host: mml0.meche.rpi.edu
To:       [e-mail]

Dan Stanger writes:
>does anyone know how to change the name of the core file that will be
>created if a program fails.  we have several programs running in the

Not without kernel sources.

>same directory so if more than one of them fails the core file from
>one overwrites the other.  we would like to change the name to
>program.core or somthing like that.

Run the program as:

program ; mv core program.core

When "program" exits, then if "core" exists, it is renamed.

This is the idea, anyway.  You want a fancier shell script.

#!/bin/sh
program || test -f core && mv core program.core

is a little better, but there's no assurance that you are renaming
the right core file.

Brian

-----------------------------

From: Upkar Singh Kohli <[e-mail]>
Subject: two questions
Date: 4 Sep 92 15:38:13 GMT
Sender: Usenet News <[e-mail]>
To:       [e-mail]


1. What is the best way to disable echo from within a C program?
   The purpose is read a password.

2. What is the best unique identifier for machines to use for
   software licensing? I have heard that even the "effective"
   ethernet address can be changed. I know the hostids can easily
   be changed on certain architectures like DEC and SGI.

Thanks!
Upkar

(if possible please email [e-mail]... followup over
 news will be ok too, except I read it less)

-----------------------------

From: Mitch Wagner <[e-mail]>
Subject: net.views -- USL/BSDI lawsuit -- responses
Date: 4 Sep 92 19:24:41 GMT
Followup-To: comp.unix.questions
To:       [e-mail]

Here we go with the responses to the latest net.views. Thanks to all
who responded to my recent plea for help; and my apologies to those
who replied the first time and whose responses I lost because of my
foolish lack of backups of the net.views file. You can certainly rest
assured that I have a good backup policy in place now.

Thanks also to those who ignored my invitation to flame me for not
making backups.

 =-=-=-=-=-=-=-=-=-=-

On another topic, if anyone would like to suggest a question for us to
use for net.views, I'd love to hear it. You can find information on
how to reach me in my sig, below.

 =-=-=-=-=-=-=-=-=-=-

And now, net.views.

The question was:

     Unix Systems Laboratories recently sued software company
     BSDI and the Regents of the University of California at
     Berkeley, charging that BSDI's commercial Unix workalike
     and UCB's free software violated USL's intellectual property
     rights. Was USL justified in filing the suit?

The responses are attached, below.

Thanks again to all---participants and readers---and apologies again
for the administrative snafu, earlier.

                         -- mitch w.
--
Mitch Wagner, senior editor, Open Systems Today
2353 Massachusetts Ave. Suite 47, Cambridge, MA 02140
[e-mail]  [phone]  CIS:70212,51  GEnie:MITCH.WAGNER
For subscription information, please call 516/562-5882





From: Berk Walker <elroy.jpl.nasa.gov!techsys![e-mail]>

I am a manager for Hamada of America, I speak only for myself.

Whereas I believe in our justice system (in general), and it's not fair
to try someone in the press, or in this case, usenet, if USL does not
come forth with /specific/ allegations, I feel I need to act.

I have 3 phone lines, and minor influence over several others, in this
and other states.  I will be forced to remove AT&T as my LD supplier,
with written notice as to why. (this includes changing cellular handler).
 APPLE never should have won, and XEROX should have sued apple.  Of
course, if UC Regents are considered a co-conspirator, and do have the
funding to do war with USL, and I feel that BSDI isn't seriously
threatened, I will just sit and watch....... but USL will probably file
for an injunction against BSDI shipping or offering for sale product.

And if BSDI /did/ do a direct port of (C) code?.... well they deserve to
be hanged.  Unfortunately, I will probably make the decision to protest
or not long before substantial proof is shown.  Then I will have struck
against the righteous..... maybe.  I use AT&T in thanks for everything
Bell Labs had done.

lots of words - sorry the meaning is so thin.

Berk Walker

From: Bob Pendelton <hal.com![e-mail]>



Bob Pendleton
Senior Staff Engineer
HAL Computer Systems

Tough question.

The answer is that I really don't know. But I care very much. I've had
access to USL sources, even if I haven't looked at them. Does this
mean that USL can claim ownership on anything I write that has a
function similiar to anything in the USL UNIX distribution?  Maybe it
does.

The thought that I can't claim (or at least can't prove) ownership of
my own ideas is chilling.

            Bob P.

From: CJ Canon <arp.anu.edu.au![e-mail]>




RESPONSE:    No - I don't believe that AT&T/USL is justified in filing
against UCB Regents (CSRG) and BSDI. After reading the views of many people,
some for, and others against, I find it completely petty and underhand that
USL should, in effect, snap at the hand which gave it so many features (UCB) to
put into it's own UNIX (whose TM?). On the BSDI front, I find it reasonable
for USL to request BSDI to remove the 1-800-ITS-UNIX phone number, but only
because that number could actually be misleading advertising. Maybe the number
1-800-ITS-BSD would have been a better choice.

    However, the request that anyone who has seen AT&T source should
be barred from working in the field for a competitor is absolutely ludicrous.
I am reminded by the flavour of this request of the suit that broke up Ma Bell
and also of the suit launched against IBM - both anti-trust cases. It makes
one wonder whether or not any of USL's lawyers have thought about the
implications for them if this case succeeds - would success not leave them
open to an anti-trust suit?

    And then on the matter of intellectual property, if USL is so
concerned about their rights, why did they not sue UC Berkeley earlier in
their (USL's) existence?

    To sum up: I hope that whoever presides over the suit dismisses
USL's case, and that complete costs are awarded to BSDI and UC Berkeley.
It might be a purely selfish motive on my part, but I certainly don't want
to be forced out of a preferred career path merely because I've seen AT&T
source and might possibly at some later date remember that code and base
further work on what I remember.


CJ Canon

   Internet: [e-mail]

   Australian EST Daytime phone: [phone]  (also fax)
   Student, (2nd year Maths --> B.Sc), ANU

  I read my email every day, if further comment/whatever is needed.




From: CJ Canon <arp.anu.edu.au![e-mail]>


C. James Canon, and the ANU is the Australian National University, located in
Canberra, the Australian Capital Territory (A.C.T.)

CJ

From: Chaitanya Nagappa <menudo.uh.edu![e-mail]>
Organization: University of Houston

Hi,
Posting this from a friend's account because of some strong feelings (about
the subj.). I have had my share of DOS and VMS, and even though each (seriously!)
had its share of good points the whole idea of it being controlled be one
single entity with its limited vision (mo' money, mo' money) as the only
impetus, I finally found myself liking Unix. Inspite of all its idiosyncracies
and difficulties, Unix (NO *TM*) has moved ahead offering its benefits to
the outside world not because the brilliant minds at AT&T found new ways of
doing things (and not releasing it / or releasing it with undocumentation),
but because the good folks at BSD released it gratis. I recently went to a
talk by the local USL manager, about SVR4.2. The feeling I got was that
this is a pure business minded enterprise now; if possible they'll put a
charge on each character that you type into the machine. The talk itself was
that Unix belongs to AT&T, and that it did not matter whose version that you
use. It seems that USL has become a larger version of Microsoft, a bigger
bully. If USL manages to place a leash on outside enhancements of Unix,
I can guarantee that this is the last major version of Unix that we will
see. Having extensively used software from people such as GNU, kermit, etc.
I wish we could move to an Operating System with the same amount of ties.

About myself - I work as a System Analyst doing programming on X.25 &
Unix, and some system administration among other things. The company
I work for is US Videotel which is in the electronic information field
(a la Prodigy, Compuserve, etc.). If you need to send email to me, you
can either to this address, with the first line addressing it to "Ravi"
or send to "[e-mail]". Phone # is (713)-877-4272.

Thanks,
--Ravi.
(Ravindran Ramachandran)

From: Contr Karl Vogel <c-17igp.wpafb.af.mil![e-mail]>
Organization: Control Data Systems Inc.

wrote:

Mitch> Unix System Laboratories recently sued software company BSDI and the
Mitch> University of California at Berkeley, charging that BSDI's commercial
Mitch> Unix workalike and UCB's free software violated USL's intellectual
Mitch> property rights. Was USL justified in filing the suit?

       The first thing that came to mind when I learned about the lawsuit was
       "We're the phone company.  We don't care.  We don't have to."  I've
       followed the BSDI efforts in "Dr. Dobbs", and applaud them for their
       ingenuity and generosity.

       [ I know, I know; USL is not exactly the same as AT&T, but they've
       inherited some of AT&T's poor attitude.]

       If USL can prove beyond a shadow of a doubt that BSDI ripped them off
       by pointing to a piece of code and saying "THIS came from us, and we
       can demonstrate that to be true", I'll listen.  I have no respect for
       a thief.  However, I haven't heard about any such evidence being
       introduced, and until I do, USL is guilty of mis-using the legal
       system by trying to kill BSDI through the "chilling effect" of a
       lawsuit.  Innocent until proven guilty.

       I can't do anything to USL to persuade them to come up with a clear
       rights violation before they start suing someone, but I was thinking
       about writing a letter to AT&T and telling them I was switching
       long-distance carriers because of what USL is doing.  I have a
       question for the net:  would this hurt USL, or are they totally
       separate corporate entities?  How much influence does AT&T have over
       USL?

---
Karl Vogel              Internet:   [e-mail]  [129.48.19.253]
Analyst/Programmer          UUCP:   ...!uunet!c-17igp.wpafb.af.mil!vogel
Control Data Sys. Inc.              [e-mail]

Sometimes, you're the windshield .... sometimes, you're the bug.
                            --Mark Knopfler

From: "Daryl V. McDaniel" <illian.mnet.com![e-mail]>



Mitch,

Briefly, here are my views on the USL/BSDI suit.  I have read a summary of
the complaint, ftpd from uunet, and base my comments partly on that.  This
year, our company decided to go with SVR4 to replace our aging BSD4.2 and
BSD4.3 systems.  I also base part of my response upon my experiences
bringing up SVR4.

USL claims that the NET2 release discloses USL and AT&T trade secrets.  The
strongest argument against this is that every aspect of the Unix operating
system is taught in schools today.  This was partly the original intent
when Unix was licensed to Universities.  One can not maintain a trade
secret while facilitating the wide disclosure of its details.

USL also claims that Bostic, et al, engaged in a conspiracy for the last
several years to force the regents to release USL proprietary source code.
It is difficult to prove or disprove a persons intent.  Experience leads me
to believe that this scenario is highly unlikely.  Where I have seen this
type of argument used before has always been a case of "grasping for
straws."

For the last several years, AT&T and USL have been promoting Unix as an
operating system for business use.  My experiences with their product span
from V6 in the early seventies, to SVR4 today.  I have watched the quality
of AT&T/USL Unix rise and fall over the years.  Product quality has been in
steady decline since SVR3.  It appears that AT&T/USL believes that the
addition of features is what makes an operating system "business ready."
During this same period, I have seen the Berkeley variant rise steadily in
quality and robustness.  From the earliest I can remember, a goal of the
group at Berkeley has been to produce an AT&T free version of Unix.

When our company decided that it was necessary to upgrade our old
(1985-1987 vintage) machines running BSD4.2 and BSD4.3 Unix, we decided
that we should switch to USL's SVR4 for the commercial support, robustness,
and modern-ness of a mainstream product.  It took two months to get SVR4
running to a point we could begin commercial operations using it.  Within
these two months, we identified and reported 163 different bugs of which 14
were fatal and 3 were the fault of the software vendor (not USL).  The 14
fatal bugs and remaining 146 non-fatal ones are directly attributable to
flaws in USL's product.

It is my belief that USL saw the availability of a low-cost product such as
BSDI's as a serious threat to their market.  With a large community of
users with access to source code, a natural maturation of the product will
take place.  The result will be a Unix product from BSDI that is robust,
performs as claimed, is well supported, and universally accepted.  Witness
the success of the GNU offerings.  The GNU C compiler is now accepted as
the "compiler of choice" on many platforms.

Due to the deficiencies of the USL product, only a large software
organization can afford to generate and maintain bug fixes and merge them
into subsequent USL releases.  At some point, the amount of code fixes
will exceed the size of original code.  Then we will see companies thinking
about replacing all of the USL code and having a product which they
completely control.

If USL wins, the world will be much worse off.  Universities and research
labs will be afraid to accept "Educational" and "Research" licenses for
companies technologies.  We will end up with many companies duplicating
effort and using valuable resources fixing and maintaining other companies
products instead of developing and maintaining their own products.  In
fact, it will significantly inhibit innovation, since the majority of new
developments are "derived" from previous developments, and USL claims that
new ideas derived from old ideas are the property of the owner of the
original idea.

Unfortunately, it is the opinion of our corporate lawyer that USL will win.
Not because of technical or legal merit, but because they have the money.

Daryl V. McDaniel        [e-mail]
Principal Engineer        [e-mail]
Micronetics
Aloha Research Group
10700 SW B-H Hwy.  #420
Beaverton, OR 97005

From: "David B. Teague" <wcuvax1.wcu.edu![e-mail]>


I am David Teague, Associate Professor of Computer Science at
Western Carolina University. I feel strongly that AT&T has
overstepped the bounds of decency in suing BSDI and others.

U Cal Berkeley submitted the NET 2 tape to AT&T for their perusal
and comment. They declined to do so, and ignored the NET 2 tapes
until someone actually began to develop an operating system by
filling in the gaps in that software. Then they sue, damaging
CMU's Mach effort, the FSF's Hurd effort, Jolitz's effort, and the
one commercial effort, BSDI.

All responsible people should boycott AT&T, changing any computer
support to another company, changing the long distance phone
service from AT&T to MCI, or Sprint, and at least writing to them
of their displeasure. (that's my opinion)



  David B Teague   [e-mail]

  Boycott AT&T. Evil be to the Evil Empire in same measure as they
  meet out to users of NET-2.

  Western Carolina University     BAN USER INTERFACE COPYRIGHTS &
  Room 308 Stillwell Building     SOFTWARE PATENTS. For info email
  Computer Science Dept           the League for Programming Freedom
  Cullowhee, NC 28723                   [e-mail].

  Justice William O. Douglas wrote: "As the night fall does not come
  at once, neither does oppression. It is in such twilight that we
  must all be aware of change in the air - however slight - lest we
  become unwitting victims of the darkness.


From: David Muir Sharnoff <tfs.com![e-mail]>


Sigh, when will people learn to do backups.

(Answer: it takes about five years on the average
I think, because you have to loose a few things
before the message gets through.)

I have no idea what I sent to you the first time.
I now keep copies of my outgoing mail.

Anyway...

     Unix Systems Laboratories recently sued software company BSDI and the
     Regents of the University of California at Berkeley, charging that BSDI's
     commercial Unix workalike and UCB's free software violated USL's
     intellectual property rights. Was USL justified in filing the suit?

At one time, AT&T ran an office that you could submit your source code
to for evaluation.  They would tell you if they thought that it
infringed on their intellectual property.  The decision was made by
computer scientists rather than lawyers.  One can only theorize about
why they shut down the office, but my guess is that they felt that they
could keep ownership of more code if the decisions were made by
the courts.

USL's actions are reprehensible because they are using legal tactics
desiged solely to expend the limited resources of the much smaller
company that they are fighting.  If they were interested in resolving
fairly the real issues in the case they would be pursuing it differently.

Technically, they are light-years behind if you ignore the
75% of sysVR4 is derrived from BSD.  Their approach shows them to be
operating under the "if you can't innovate, litigate" creed.

Assuming that BSDI does not suffocate in the legal slime and that
UC Berkeley actually fights, the suit will probably resolve around
creating a new legal definition for derivative work.  If USL/AT&T
wins completly then they may gain retroactive ownership of most of
the operating systems research done in the last seventeen years.  This
will be an indirect death-blow to UNIX and possibly the US software
industry.  The suit has already caused projects that would have
benefitted many to be cancelled or delayed.

To protest AT&T's attemted enforcement of their backing-store patent,
I switched to MCI.  To protest this suit, I've dropped my AT&T Universal
Card.  Please join my boycott.



David Muir Sharnoff
Systems Engineer
TRW Financial Systems

Organization: eklektix - Boulder, Colorado
From: Dick Dunn <raven.eklektix.com![e-mail]>


Well, just consider it this way:  I sent a response and kept a copy.  After
a few days, I got an acknowledgment, at which point I deleted my copy.  If
you think I'm going to write another one, think again.

I expect you've lost the more thoughtful responses.  The folks who will
write again, except for those who actually kept copies of their mail all
this time, are more likely to be the rabid ones.  I don't know how you can
patch around this one; I don't think you can.  It's not like starting over
from scratch on the question.

It's OK for me...other trade rags (and a newspaper) have phone-interviewed
me about the suit, so my opinions will show up here and there.
---
Dick Dunn    [e-mail]   -or-   raven!rcd    Boulder, Colorado
                Cats!
    by duke.cs.duke.edu (5.65/2.22G/4.1.1)

From: "Gregory G. Woodbury" <wolves.durham.nc.us![e-mail]>


Gregory G. Woodbury
Duke University Center for Demographic Studies
Systems Programmer

    I find myself vary confused about this issue.  Based on the
discussions on the net, and the readings that I have done on the event
in various places, it seems to me that there is some sort of violation
of the spirit of the intellectual property rights that AT&T/USL hold in
the Unix brand Operating System, but (IMO) not a technical violation of
the letter of the licenses and laws.
    The filings by USL in the case (as published) are so vague and
unclear as to which parts of the code they are claiming infringement on,
and the possibilities that they may be attempting to compare a modern
UNIX source code version against the BSD source code where there may be
large swaths of the BSD enhancements integrated into the USL code, make
me very uncomfortable with USL's intentions in the case.
    The "politics" of the suit, also, cannot be ignored.  There are
a number of "Edition 7" derived "unix-like" products out there on the
market, that AT&T and USL could potentially go after, but they decide to
pick on BSDI and UCB-CSRG (while ignoring the "volunteer effort" OS also
derived from the Net-2 sources.)  This certainly appears to be intended
to have a "chilling" effect on developing systems based on the BSD Net-2
sources.
--
Gregory G. Woodbury @ The Wolves Den UNIX, Durham NC
UUCP: ...dukcds!wolves!ggw   ...duke!wolves!ggw           [use the maps!]
Domain: [e-mail]  [e-mail]  [e-mail]
[The line eater is a boojum snark! ]           <standard disclaimers apply>



From: John Hensley <usenix!clinicom.com![e-mail]>
Organization: CliniCom Incorporated, Boulder, CO


Justified or not, it's the absolute wrong thing to do when trying to sell
an infant UNIX product to users who have been enjoying CSRG's work in one
form or another for years. Destiny/UNIXware/whatever was looking like a strong
competitor to Windows NT, Solaris, Next Step and crew, but now I hope this
will be the last we hear of USL.

 -------------------------------------------------------------------------------

John Hensley
Systems Administrator
CliniCom, Incorporated
[e-mail]

(P.S. Thanks for screwing up -- I missed the original request!)

--
John Hensley           |
[e-mail]      | "You think slower when you graze." -- Holling
[e-mail] |

From: Jon Peatfield <cus.cam.ac.uk![e-mail]>

Well I was just about to delete my copy of that mail...

I wrote:
--include-the-text--

Well, I've read the initial complaint, but havn't had a chance to look
at the extended suit yet, and this is the big one since it names UCB
and all the Regents of UC for claiming that NET/2 is free of any AT&T
code... so this is entirely based on the first complaint and what I've
read on the net.

It would appear that USL are attempting to force a small startup
company into failing, by making them fight a legal battle which
appears to be totally without cause.  In particulat the NET/1 and
NET/2 releases have been made avaialable by the UCB for several years
without complaint by USL.  Indeed in neither of the complaints have
USL explained what they believe to be proprietary to USL in the NET/2
release.  They are now demanding some form of comparison between
versions of code -- and here I'll assume they want a comparison
between the last version AT&T shipped to UCB many years ago and the
NET/2 -- yet they have had access to NET/2 since it's release by UCB,
and can't point to the offending bits...

While I would not want to deny a company the right to pursue it's
rights to prevent unauthorised use of proprietary material, this does
not seem to be the case here.  Perhaps USL are just worried about how
Destiny will fair against the much cheaper offerings made possible by
the release of the BSD NET/2 code.  Unix and Destiny are trademarks of
AT&T and USL (I think.)


My name:    Jon Peatfield
My job:        Computer Officer
My university:    Cambridge University (Department of Applied Maths and
        Theoritical Physics) U.K.
My phone number: [phone] (work) [phone] (home)
(answering machine at home, since this is GMT +1 (BST actually), my
home number is probably more useful for leaving messages in your
daytime.  I assume you know how to dial international calls...)
--end-of-include--

Since then I've read the second complaint too, and I can't see any
reasonable interpretation which should stand up in a court of law.
However, courts being what they are once can never tell, since the
court can be easily pursuaded by good lawyers, rather than on the
technical issues.  I'm hoping for it to get thrown out, but I've heard
of worse things getting accepted by the courts.

-- Jon Peatfield

From: Jonathan Eunice <cs.pitt.edu![e-mail]>


    Unix Systems Laboratories recently sued software company BSDI and the
    Regents of the University of California at Berkeley, charging that
    BSDI's commercial Unix workalike and UCB's free software violated USL's
    intellectual property rights. Was USL justified in filing the suit?

Of course USL is justified in taking any and all legal actions its
management deems necessary to protect its property and its rights.
Whether our laws properly define property and intellectual rights
relative to software is a completely different question, as is whether
attacking BSDI or UCB will best serve USL's ends and interests.  But
the law gives "the rules," and everyone (including USL) has a right to
play as agressively as possible within those rules.

Where USL is clearly unjustified is its continual hyping of the
"openness" of its key product, UNIX System V Release 4.  The recent
legal action simply confirms an oft-hidden message delivered several
years ago with the AT&T-internal slogan, "UNIX(r) is a five-letter
word"--ie, that UNIX System V is proprietary and USL has a selfish
interest in it no weaker than IBM's interest in MVS, or DEC's in VAX/VMS.

We are deluded if we believe that UNIX System V (and its companion
"standard," SVID) is magically "open," and that USL cares primarily
about advancing peace, goodwill, and good operating systems among
humanity.  USL is a business, with a bottom line.  SRV4 and SVID are
only slightly less proprietary than SAA or the Macintosh, and equal on
the proprietary/open continuum to AIX, A/UX, Coherent, DG/UX, DYNIX,
ES/IX, HP-UX, ODT, Solaris, ULTRIX or any of the other
standards-oriented systems based on what began as "Unix" two decades
ago inside BTL.

Jonathan Eunice
Research Manager/Software
D.H. Brown Associates

From: Karl Lehenbauer <sugar.neosoft.com![e-mail]>
Organization: NeoSoft Communications Services --

I think that as long as USL saw the NET-2 release as nothing more than a
grab bag of software technology, they didn't have a problem with it.
When they discovered that talented hackers had been able to cobble together
the missing pieces and produce a credible 4.3 BSD system (allegedly) free of
their copyrights, they totally freaked.

Ironically, many of the similarities between the two systems came from
AT&T/USL adopting code from Berkeley.  For a GUI, both systems run X-windows,
which of course originated at MIT.

Karl Lehenbauer
Member of Technical Staff
NeoSoft
--
-- Email [e-mail] for info on getting interactive Internet access.
You will now awaken feeling relaxed and refreshed, remembering everything
you've read except the details of the Omega contingency plan.
---
-- Email [e-mail] for info on getting interactive Internet access.
You will now awaken feeling relaxed and refreshed, remembering everything
you've read except the details of the Omega contingency plan.
From: Karl Lehenbauer <sugar.neosoft.com![e-mail]>



Forgive me.  We are located in Houston, Texas.

--
-- Email [e-mail] for info on getting interactive Internet access.
You will now awaken feeling relaxed and refreshed, remembering everything
you've read except the details of the Omega contingency plan.


From: Michael Bentley <nuchat.sccsi.com![e-mail]>
Organization: /etc/organization


I am no lawyer.  As a non-lawyer, I feel confident in saying that USL
is doing the Wrong Thing.  I read through all the documents (from uunet)
related to the matter, and have concluded that the AT&T/USL position is
one of random flameage (and wholly content free).

If USL insists on trying to keep the world from using the Unix name (the
only specific complaint that I saw in the aforementioned documentation), I
think we can all learn to say GNU!

Michael Bentley
no company (unemployed)
no title (usually programmer)

--
+============================================================+
|  Michael S. Bentley (Looking for a way back to California) |
|  VoiceNet :                                                |
|  E-mail   : [e-mail]                       |
+============================================================+

From: Multics <acm.rpi.edu![e-mail]>
Organization: The Voice of Fate

Since UNIX is based so strongly on Multics, shouldn't Honeywell consider
sueing AT&T for the same reasons AT&T is sueing BSDI and UCB?

I've heard that the older UNIX manuals even acknowledge Multics as the
parent of UNIX.

Richard Shetron, John Grady Inc.
I'm not availibe by phone during the day, though I will be home this
monday packing for my vacation.  [phone].  I'll be in and out
most of the day.
--
Richard Shetron  [e-mail]  [e-mail]
What is the Meaning of Life?      There is no meaning,
It's just a consequence of complex carbon based chemistry; don't worry about it
The Super 76, "Free Aspirin and Tender Sympathy", Las Vegas Strip.
From: Multics <acm.rpi.edu![e-mail]>

Richard Shetron, John Grady Inc., 358 Broadway suite 402,
Saratoga Springs, NY 12180.

NOTE:we do not use unix at work, I only have access at work.  My employer
knows nothing about anything except Bluebird SuperDos and Wang 2200.

--
Richard Shetron  [e-mail]  [e-mail]
What is the Meaning of Life?      There is no meaning,
It's just a consequence of complex carbon based chemistry; don't worry about it
The Super 76, "Free Aspirin and Tender Sympathy", Las Vegas Strip.
From: Multics <acm.rpi.edu![e-mail]>

It was written originally to run Data General Business basic on a pc in
multi-user mode.  It also runs Niakwa Basic-2C a wang 2200 emulator,
HIGH-C, and I believe a DEC DIBOL compiler of some sort.

There are no directories, instead there are 64 user groups per disk.

--
Richard Shetron  [e-mail]  [e-mail]
What is the Meaning of Life?      There is no meaning,
It's just a consequence of complex carbon based chemistry; don't worry about it
The Super 76, "Free Aspirin and Tender Sympathy", Las Vegas Strip.
From: Multics <acm.rpi.edu![e-mail]>

Its neither DOS compatable or better the DOS.  MS-DOS 2.0 and maybe even
1.0 are better, except SuperDos is multiuser.  It uses technology that I
always thought was proven obsolete by IBM when they invented the methods.
All users get 1 fixed size partition allocated when the system boots.
The older versions designed to run on the PC limits the partition size to
remaining memory in the base partitioin or 64K in expanded memory.  the
newer version allows use of expanded memory on 386+ machines and partition
size is limited to memory size, which I think is limited to 16MB total.

There is no command processor, i/o redirection, etc.

Whatever I want, JGI is a small software firm, John is the president and
holds all officer's positions and I'm the programming staff.  Thee is a
part time secretary.  We are listed in D&B's business credit ratings listings.
When John wants me to sound important, my title is sometime listed as
Manager of software R&D or similier types of titles.  I just tend to call
myself a programmer/analyst/software engineer.

--
Richard Shetron  [e-mail]  [e-mail]
What is the Meaning of Life?      There is no meaning,
It's just a consequence of complex carbon based chemistry; don't worry about it
The Super 76, "Free Aspirin and Tender Sympathy", Las Vegas Strip.
    by himnariki.cs.UMD.EDU (5.64/UMIACS-0.9/04-05-88)


From: 'Olafur Gudmundsson <cs.umd.edu![e-mail]>


Forwarded old message (slightly edited), with an addition
 ------- Forwarded Message

From: ogud (Olafur Gudmundsson)


You asked about my oppinion on the USL suit.

It stinks!!
After reading the brief filed with the court, my reaction one of shock.
These USL people are out of sync with the real world.
BSD has contributed a lot to the evolution of UNIX while the
contributions by USL are marginal (at best). V32 is a 14 year old system
that was outdated by BSD 4.1 and AT&T own Release 7. All the ideas of
UNIX have been well documented in books and articles over the last decade
so what is the fuzz about.  The fuzz is about control USL wants to become
single source of UNIX (impossible dream given who works there), they must
think that they can run BSDI out of business and that is a prerequisite
for gaining monopoly on UNIX, and guarantees them and their lawyers
lifetime employment!

Or their goal is to gain 100% controll of where people like me,
that have seen UNIX source code, work. If they want to dictate to me and
othere where we can work and what we are permitted to do. They will end their
life the same way as the pigs in "Animal farm".

- -    -    -    -    -    -    -    -    -
Olafur Gudmundsson  Dept. of Computer Science University of Maryland
Internet: [e-mail]               UUCP: {...!}uunet!mimsy!ogud
UPS: College Park MD. 20742
Research Faculty
 ------- End of Forwarded Message

From: Paul Nash <frcs.alt.za![e-mail]>



For a number of reasons, most of which other have stated on the net, I
feel that USL are _not_ justified.  They certainly give the impression
of being yet another monolith that is getting upset at seeing someone
else trying to introduce competition.  Can you say "phone company"?
They are acting just like IBM :-(.

Paul Nash, M.Eng
Network Development Consultant, CSIR
PO Box 395, Pretoria, 0001 South Africa

 ---=---=---=---=---=---=---=---=---=---=---=---=---=---=---=---=---
 Paul Nash                                         [e-mail]
 Box 12475, Onderstepoort, 0110 South Africa         [phone]

      "You don't want to get locked into open systems" -- IBM
From: Paul Nash <frcs.alt.za![e-mail]>


CSIR == Council for Scientific and Industrial Research (sort-of NSF
clone [-ish]).  I am a "Network Consultant" == resident guru.

    paul

 ---=---=---=---=---=---=---=---=---=---=---=---=---=---=---=---=---
 Paul Nash                                         [e-mail]
 Box 12475, Onderstepoort, 0110 South Africa

      "You don't want to get locked into open systems" -- IBM
          Thu,  3 Sep 92 14:34:20 PST
From: ssd0.laafb.af.mil![e-mail]

Oh, sorry
Los Angeles AirForce Base  Space & Missle Center / Something or other I've
never been able to figure out the rest of it.

Chuck
 ------------------------------------------------------------------------------

Organization: Taronga Park BBS
From: Peter da Silva <taronga.com![e-mail]>


Smooth. Let's see if I can recall...

                                                         ^^^^^^^^^ filing?

No.

If they'd done so three years ago when Berkeley started shipping large
subsets of "freed" code, before they'd established an expectation that
this code was indeed free, maybe. As it is, they've encouraged a lot of
people to invest a lot of resources in this code, and are clearly trying
to use FUD to hurt a potential competitor rather than simply protect
their legitimate rights. The fact that they waited until the 11th hour
to include the Net/2 release in the suit bears this out... if their
trade secrets were what they were worrying about they'd have hit THAT
first, and hit Net/1 as well.

Peter da Silva
Lead Software Engineer
Ferranti International Controls Corporation
---
                                                                `-_-'
                         Have you hugged your wolf today?        'U`

Peter da Silva, Taronga Park BBS, Houston, TX


From: Phil Nelson <henson.cc.wwu.edu![e-mail]>


This is a hard question to answer.  My "reaction" is no.  If they are
claiming actual copyright violation, they I can see it as "justified."
On the other hand, if they are claiming a "look-and-feel" type
violation, then I think they are not justified in filing suit.

From what I have heard (not know), USL is not claiming that BSDI or UCB
is distributing AT&T (or USL) copyrighted materials, just that it looks
similar.  IF this really is the case, then I think that USL is wrong,
and if USL does not withdraw their suit, then the entire UN*X community
should join a boycott against AT&T long distance service.  Such a
boycott should be publicized.


Phil Nelson
Assistant Professor of Computer Science
Western Washington University
Bellingham, WA 89226-9062

Member of the League for Programming Freedom


From: Russell Nelson <crynwr.com![e-mail]>
Organization: Crynwr Software



Moreover, was it wise?  With NT nipping at Unix's heels, should the
"real" Unix vendors be arguing with each other?  Better 10% of a size
N^2 market than 90% of a size N market...

--
-russ <[e-mail]>  I'm proud to be a humble Quaker!
Crynwr Software            Crynwr Software sells packet driver support.
11 Grant St.
Potsdam, NY 13676

From: Russell Nelson <crynwr.com![e-mail]>
Organization: Crynwr Software



Oh, why, President, of course.  Unless you want me to be Head of R&D,
or maybe, Clerk, or perhaps just Head Janitor.  :)

--
-russ <[e-mail]>  I'm proud to be a humble Quaker!
Crynwr Software            Crynwr Software sells packet driver support.
11 Grant St.
Potsdam, NY 13676

From: Ted Lemon <lupine![e-mail]>


It is clear from reading the two USL complaints (which I obtained from
uunet) that at least one motivation of this lawsuit is to drive BSDI out
of business by exposing them to a long and involved legal process.    My
reason for believing this is that the suit makes a very large number of
very broad, vague claims, and also proposes a rather bizarre conspiracy
theory that makes no sense based on my knowledge of the people involved.
Barristry of this kind is a shameful misuse of our legal system.

It is possible that there may be some code in the Net2 release which is
derived from the AT&T Unix 32V sources that Berkeley received from AT&T
many years ago, and if that is the case, Berkeley should certainly be
required to remove that code from any public BSD distribution.   However,
at this late date it's hard to imagine what that code might be, and it's
even harder to imagine how the release of that code could cause significant
harm to USL.

Of course, it's quite obvious how the use of Net2 to produce a UNIX-like
system without requiring a UNIX license could harm USL, but removing any
remaining 32V code from the Net2 distribution probably wouldn't significantly
impede the implementation of a UNIX-like system from Net2.   USL's claims
regarding trade secrets are clearly nonsense - very little about UNIX as
it is today is in any way based on UNIX as it was at the time of 32V, and
that part of the interface that is the same is well-known to the computing
community and cannot be covered by any sort of trade secret protection.

I am a senior software engineer at NCD, and the opinions expressed above
are my own and have no known (by me) relationship to the opinions of NCD.

                _MelloN_

From: George W Herbert <lurnix.com![e-mail]>


    As anyone who's been reading the Usenet
discussions can attest, the biggest problem with
the USL/BSDI/UCB suit is confusion over exactly
what it is.
    Is it a claim that there is AT&T code
left in the Berkeley Net/2 release?
    Is it an attempt to claim that anything
that walks, talks, and swaps like UNIX is USL
property?
    Or is it a fishing expedition by USL
trying to harrass BSDI and Berkeley?
    We can't tell.  While the suit makes some
explicit claims (that Net/2 infringes on USL's
trade secrets, as does BSDI's work) it doesn't
tell us wether they refer to bits of code or the
functionality or the concept.  USL hasn't
come clean and explained exactly what they're
referring to, and Berkeley and BSDI apparently
don't know.
    Personally, I dislike the suit.  It seems,
and I admit this is just a feeling, like USL
is trying to claim that anything that works
like USL Unix and looks sort of like AT&T Unix
is theirs.  It seems like a nasty turnabout from
the previous AT&T position of letting unencumbered
sections of BSD releases be freely released
to the world.  But, like just about everyone else
I don't really know.  And that's what worries me
the most, right now.

-George William Herbert
Analyst
Lurnix
Berkeley, CA

From: Doug Carter <htbawd.rain.com![e-mail]>



As in the past, AT&T seems only to care about UNIX after someone
does something meaningful with it. The fine efforts of the people
at Berkeley turned a bunch of development tools into a useful
operating system. It was that same effort that got AT&T seeing
enough dollar signs for them to produce System V. Now that Berkeley
has finished the job, the only way AT&T can figure to get a cut is
through litigation. AT&T is to be commended for the concepts and
methods that UNIX has brought to all computing environments. They
are to be criticized for their continued mindless manangement of
UNIX, a global success, despite their folly.


    Doug Carter
    A satisfied BDSI customer
    Beaverton, Oregon.


From: ssd0.laafb.af.mil![e-mail]

Gosh, I would love to have a free (UCB), or low cost (BSDI), full blown Unix
on my desktop.  Then I wouldn't need OS/2.  Unfortunately, since USL owns the
Unix name and source code, and both UCB and BSDI had access to that source
code, it is likely the code in the UCB and BSDI products infringes on the USL
copyright.  It's not nice to build a house with stolen lumber.  And though I
hate lawyers, it is nice to have snakes in the grass when the gophers are
nibbling at your garden.

Chuck Paussa
Sr Systems Engineer
LAAFB SMC/SCTMA
Los Angeles, CA


From: watson.ibm.com![e-mail]

Hello,
    To answer your question - USL had absolutely no
    moral or technical rights to file their
    completely unjustified lawsuit
    against BSDI.

    This opinion is mine and has nothing to do with
    my employer.

    Uri Blumenthal
    Research Staff Member.


From: Dave Stokes <ucsd.edu!comtec![e-mail]>


I glad that when the wheel was invented that nobody claimed
intellectual 'property' or we would all still be walking or
riding horses!  I guess U$L is now free to go after MINIX
since that also is a work alike.  It seems like U$L is picking
off the 'commerical easy targets' and maybe they'll go after
the CP/M folk next!

David Stokes
ORINCON Corp


From: merlin <usc.edu![e-mail]>


The copyright law is intended to protect the right to profit from one's
writings by forbidding plagiarism and/or wholesale duplication without
just compensation to the author.  In the present case, USL is claiming
a property interest in software designed and written by a large number
of contributors working in government laboratories, public universities,
and private organizations throughout the world.  The 4.3BSD-NET2 source
code is a copyrighted collection of programs owned by the U. C. Regents,
publically certified as containing no source code belonging to AT&T/USL,
and licensed for use, modification, and redistribution for any purpose.
U. C. Regents carefully screened all contributions to the 4.3BSD-NET2 in
order to avoid any possible infringement of code derived from AT&T UNIX.
As a consequence, USL appears to have no legitimate legal basis for the
present lawsuit against U. C. Regents and Berkeley Software Design, Inc.

 ------------------------------------------------------------------------------
Alexander-James Annala
Principal Investigator
Neuroscience Image Analysis Network
HEDCO Neuroscience Building, Fifth Floor
University of Southern California
University Park
Los Angeles, CA 90089-2520



From: [e-mail]

The current legal situation (patents, copyrights) re software is a mess.
This is an area the US is the world leader in. It won't be if we keep
extracting money from businesses to have lawyers debate things in court,
particularly since no-one seems to have the wit or grace to admit that
*anything* is obviously right or wrong today (example: politicians).
Nor will the US succeed if small software businesses are excluded by forcing
legal costs (e.g. patent searches) up them.

ATT vs BSDI is not a clear-cut issue either way, but is certainly part of
this obnoxious trend towards litigation as the basis for business competition.
Whatever happened to making a better product or doing a better job of
marketing?
 ----------------------------------------------------------------------
  Dr. Peter J. Welcher            EMAIL: [e-mail]
  Mathematics Department, M/S 9E
  572 Holloway Road
  U.S. Naval Academy
  Annapolis, MD  21402-5002
 ----------------------------------------------------------------------


From: [e-mail]


Professor. Didn't my signature have it ?
I do/did s/w consulting, which my have some influence on why I think
the current s/w patent situation is beyond absurd. It
in principle puts small guys
out of business writing new software!
 ----------------------------------------------------------------------
  Dr. Peter J. Welcher            EMAIL: [e-mail]
  Mathematics Department, M/S 9E
  572 Holloway Road
  U.S. Naval Academy
  Annapolis, MD  21402-5002
 ----------------------------------------------------------------------


From: apollo.hp.com![e-mail]

Take two.

And do better backups in the future!

 ------- Forwarded Message

From: [e-mail]

My name is Bill Sommerfeld; I'm a software engineer employed by, but
by no means speaking for, Hewlett Packard. I can be reached at
1-508-436-4352 or 1-617-396-5661.

       Unix System Laboratories recently sued software company BSDI
       and the University of California at Berkeley, charging that
       BSDI's commercial Unix workalike and UCB's free software
       violated USL's intellectual property rights. Was USL justified
       in filing the suit?

No.  USL and/or AT&T once ran a service by which universities could
determine whether or not source code was "contaminated" with source
from UNIX; however, they stopped doing this several years ago.

As a result, persons connected with CSRG made a good faith effort to
separate out the code in BSD not derived from AT&T code, and made it
freely available.  Now, this is an eminently tricky process, so it's
entirely likely that a few files of AT&T code may have slipped through
the cracks and made it into NET/2 (just as AT&T code is occasionally
posted to USENET by accident).  However, instead of filing a lawsuit,
USL should have made specific complaints about specific infringements,
and UCB would most likely have taken prompt, reasonable action to
correct the problems.

 ------- End Forwarded Message


From: warren.mentorg.com![e-mail]


I'd be surprized if every vendor in the world selling a backup scheme
doesn't call you to ask if you want to buy the product. :-)



We really can't know until USL reveals more details.  For example,
which lines of source are they talking about?  Until then, everything
is just hype.  Including this.

Tom Limoncelli
Mentor Graphics Corp -- IC Group
System Admin.
(not speaking for my company)
--
Tom Limoncelli -- [e-mail] (work) -- [e-mail] (play)
"Oh! I thought it was one of those useless demos of everything that
    a GUI builder could do." -Anonymous person watching demo of
         Solaris 2.0's graphical tool for managing NIS+

-----------------------------

From: Barry Margolin <[e-mail]>
Subject: Re: Question: which is faster 'find -exec' or 'find | xargs' ??
Date: 4 Sep 92 21:49:11 GMT
NNTP-Posting-Host: telecaster.think.com
To:       [e-mail]

In article <[e-mail]> [e-mail] (John Navarra) writes:
>I was experimenting with find and xargs. I issued the command
>  find . -exec ls -ld {} \;
>and the command
>  find . -print | xargs ls -ld

>Question: why is the xargs command MUCH faster?

The -exec version executes a separate ls command for each file that is
found.  The xargs version accumulates a bunch of arguments and then
executes one ls command with all of them.  The difference is all the time
spent starting up new shell and ls processes.  It's the difference between:

    sh -c "ls file1"
    sh -c "ls file2"
    sh -c "ls file3"
    sh -c "ls file4"
    sh -c "ls file5"

and

    sh -c "ls file1 file2 file3 file4 file5"
--
Barry Margolin
System Manager, Thinking Machines Corp.

[e-mail]          {uunet,harvard}!think!barmar

-----------------------------

From: Mitch Wagner <[e-mail]>
Subject: net.views -- which vendor is most open?
Date: 4 Sep 92 22:59:46 GMT
Followup-To: comp.unix.questions
To:       [e-mail]


       Which computer and/or operating-vendor has the greatest
       commitment to open systems? Which has the least? Why?


     This question is being posted to gather responses for a regular
 opinion column in OPEN SYSTEMS TODAY called "net.views."

     By posting a response to this topic, you are granting
 permission for OPEN SYSTEMS TODAY us to publish your response. A
 compilation of the responses to this post will be posted here about
 two weeks from today.

     Please include in your response your name, your employer or
 university, your job title or class standing (that is to say, if
 you identify yourself as a student, are you a freshman, sophomore,
 junior, senior, graduate student?), and a telephone number where
 you can be reached during the daytime (overseas readers note:
 that's daytime in North America).

     Please be brief; try to limit your responses to two screens
 (24x80) of text, not including headers and other administrivia.

     Thanks!

                         -- mitch w.
--
Mitch Wagner, senior editor, Open Systems Today
2353 Massachusetts Ave. Suite 47, Cambridge, MA 02140
[e-mail]  [phone]  CIS:70212,51  GEnie:MITCH.WAGNER
For subscription information, please call 516/562-5882

-----------------------------


End of INFO-UNIX Digest
***********************