home *** CD-ROM | disk | FTP | other *** search
/ Cream of the Crop 26 / Cream of the Crop 26.iso / bbs / fnewse24.zip / FIDO1424.NWS
Text File  |  1997-06-16  |  117KB  |  2,545 lines

  1.      F I D O N E W S --       Volume 14, Number 24          16 June 1997
  2.      +----------------------------+-----------------------------------------+
  3.      |  The newsletter of the     |   ISSN 1198-4589 Published by:          |
  4.      |    FidoNet community       |   "FidoNews"                            |
  5.      |          _                 |        1-904-409-7040    [1:1/23]       |
  6.      |         /  \               |                                         |
  7.      |        /|oo \              |                                         |
  8.      |       (_|  /_)             |                                         |
  9.      |        _`@/_ \    _        |                                         |
  10.      |       |     | \   \\       |   Editor:                               |
  11.      |       | (*) |  \   ))      |        Christopher Baker  1:18/14       |
  12.      |       |__U__| /  \//       |                                         |
  13.      |        _//|| _\   /        |                                         |
  14.      |       (_/(_|(____/         |                                         |
  15.      |             (jm)           |     Newspapers should have no friends.  |
  16.      |                            |                    -- JOSEPH PULITZER   |
  17.      +----------------------------+-----------------------------------------+
  18.      |               Submission address: FidoNews Editor 1:1/23             |
  19.      +----------------------------------------------------------------------+
  20.      |  MORE addresses:                                                     |
  21.      |                                                                      |
  22.      |    submissions=> cbaker84@digital.net                                |
  23.      +----------------------------------------------------------------------+
  24.      |    For  information,   copyrights,   article   submissions,          |
  25.      |    obtaining copies of FidoNews or the internet gateway FAQ          |
  26.      |    please refer to the end of this file.                             |
  27.      +----------------------------------------------------------------------+
  28.  
  29.  
  30.                      NO NEWS IS GOOD NEWS?
  31.  
  32.  
  33.                         Table of Contents
  34.      1. EDITORIAL  ................................................  1
  35.         How hard IS it to format a text file to 70 columns?  ......  1
  36.      2. LETTERS TO THE EDITOR  ....................................  2
  37.         What to do for more articles  .............................  2
  38.      3. ARTICLES  .................................................  3
  39.         SHARING-the FidoNet World Class, Global Communication E  ..  3
  40.      4. GETTING TECHNICAL  ........................................  7
  41.         FSC-0082 - Proposed New Packet Type  ......................  7
  42.         FSC-0083 - Proposed standard for message IDs  ............. 18
  43.      5. COORDINATORS CORNER  ...................................... 37
  44.         Nodelist-statistics as seen from Zone-2 for day 164  ...... 37
  45.      6. NET HUMOR  ................................................ 38
  46.         Java the Hutt?  ........................................... 38
  47.      7. NOTICES  .................................................. 40
  48.         A_THEIST Echo is on the Backbone!  ........................ 40
  49.         Future History  ........................................... 40
  50.      8. FIDONEWS PUBLIC-KEY  ...................................... 42
  51.         FidoNews PGP public-key listing  .......................... 42
  52.      9. FIDONET BY INTERNET  ...................................... 43
  53.      10. FIDONEWS INFORMATION  .................................... 45
  54.      FIDONEWS 14-24               Page 1                   16 Jun 1997
  55.  
  56.  
  57.      =================================================================
  58.                                  EDITORIAL
  59.      =================================================================
  60.  
  61.  
  62.      Even the most primitive word processing [EDLIN anyone?] programs can
  63.      count the columns live and in color as one is typing an article or
  64.      notice. Netmail and email is harder to control but text files are
  65.      difficult to make conform to ARTSPEC.DOC? Nah.
  66.  
  67.      If anyone is being held back by the formatting constraints of the
  68.      ARTSPECs, please be assured that I will reformat [and even spell check
  69.      if desired] your submission for you. ARTSPEC exists to provide
  70.      directions to make a file that an automated MAKENEWS operation will
  71.      not choke on during a weekly run to produce FidoNews. Since I don't
  72.      run FidoNews as an automatic function, the poorly formatted stuff
  73.      doesn't get stuck in the dead letter file.
  74.  
  75.      A couple promised articles never arrived [and i still have the email
  76.      link open behind me as i type this, JIC] so here's what we have for
  77.      this week.
  78.  
  79.      BTW, what ever happened to the ASCII_ART Echo? It still isn't in the
  80.      BACKBONE.NA list and I would have sworn it got the REC requests.
  81.  
  82.      And anybody heard any news on the resumption of ops at 1:13/10?
  83.  
  84.      My Father's Day ran a little late and so is this Issue. [grin]
  85.  
  86.      C.B.
  87.  
  88.      -----------------------------------------------------------------
  89.  
  90.      FIDONEWS 14-24               Page 2                   16 Jun 1997
  91.  
  92.  
  93.      =================================================================
  94.                            LETTERS TO THE EDITOR
  95.      =================================================================
  96.  
  97.  
  98.      From: steve@gen.lcrnet.org
  99.      Date: 13 Jun 97 22:11:07 -0700
  100.      Subject: Fidonews - what to do for more articles
  101.      To: cbaker84@digital.net
  102.      Organization: Don't Mistake Lack Of Talent For Genius
  103.  
  104.      >From steve steffler, 1:342/1022
  105.  
  106.      Hi Chris, feel free to publish this in Fidonews if you wish..
  107.  
  108.      I see in the newest Fnews that you ask what could be done to make
  109.      readers write articles for Fidonews.  I know that for me, the one
  110.      thing that has kept me from contributing is that I don't like having
  111.      to carefully follow ARTSPEC.DOC when writing.  If you eliminated that
  112.      document and formatted everything yourself before feeding it into that
  113.      MKNEWS program (or whatever it's called) then I am confident that more
  114.      people would contribute to the publication.
  115.  
  116.      I'm also a firm believer that there is too much filler content in it -
  117.      if a 10k Fidonews was released without all the stuff like the PGP key
  118.      and the FSC documents, perhaps people would open their eyes and see
  119.      that an important link between individuals in the Fidonet community is
  120.      on the brink of extinction.  Plus, the stuff that never changes from
  121.      issue to issue is really a deterrent to reading it, but I'll save that
  122.      for another rant, as I'm sure you've already all heard it all before.
  123.      ;-)   <= (Suggestion: Put the copyright notice, etc, etc, stuff that
  124.      never changes or rarely changes, into a separate file in the Fidonews
  125.      archive?)
  126.  
  127.      steve@gen.lcrnet.org * http://www.geocities.com/SoHo/3755 * team os/2
  128.  
  129.       -30-
  130.  
  131.      -----------------------------------------------------------------
  132.  
  133.      FIDONEWS 14-24               Page 3                   16 Jun 1997
  134.  
  135.  
  136.      =================================================================
  137.                                  ARTICLES
  138.      =================================================================
  139.  
  140.  
  141.      Christopher Baker
  142.      Rights On!, 1:18/14
  143.      Edgewater_FL_USA
  144.  
  145.      There's a new Echo in town dedicated to the REAL spirit of FidoNet
  146.      Sysoping. It's called SHARING. Here's the blurb from the EListing:
  147.  
  148.      What Sysops will see and learn in SHARING will make it "the place to
  149.      be." What they will see will be up to them but there will be laughter
  150.      and wonders and mysteries of FidoNet technology will be explained.
  151.      SHARE the wonderful experience of our hobby.  This Echo will be
  152.      Awesome. It's the Swiss Army knife of information!  There's plenty to
  153.      learn. Sysops can SHARE everything!  Every Sysop in the FidoNet
  154.      phonebook [Nodelist] is invited to participate but we will have a few
  155.      simple rules. POLITENESS to each other is EXPECTED! It is REQUIRED!
  156.      The few off-topic topics are contained in the accompanying SHARING.RUL
  157.      file. The Moderators reserve the right to declare off-topic any
  158.      subject getting out of the bounds of politeness and cooperation. The
  159.      Backbone or any backbone is off-topic automatically except as noted in
  160.      the rules.
  161.  
  162.      mod  Christopher Baker, 1:18/14
  163.      mod  jim barchuk, 1:141/355
  164.      mod  Debra Milner, 1:112/285
  165.      mod  Emeritus-Don Dawson, 1:150/730
  166.  
  167.      Backbone status has been in effect for months so Areafix a link from
  168.      your local Backbone feed or contact any of the Moderators for a direct
  169.      link via Netmail.
  170.  
  171.      Here's the SHARING Echo Guidelines for those without a copy of the
  172.      current ELRUL file:
  173.  
  174.      --- Following message extracted from SHARING @ 1:18/14 ---
  175.          By Christopher Baker on Sat Apr 20 23:38:03 1997
  176.  
  177.      From: Christopher Baker
  178.      To: All
  179.      Date: 15 Apr 97  00:52:32
  180.      Subj: SHARING Echo Guidelines - regular repost
  181.  
  182.      From: [by Don Dawson]
  183.      To: Y'all
  184.      Date:  4 Aug 95  22:27:48
  185.      Subj: Da Rulz
  186.  
  187.      Sorry, it's dirty work but someone has to do it.  :-)
  188.  
  189.      What is this echo?
  190.      ------------------
  191.      FidoNet is a World Class, Global, Communications Network of, by and
  192.      FIDONEWS 14-24               Page 4                   16 Jun 1997
  193.  
  194.  
  195.      for FidoNet Sysops around the Globe.  There are other FTN networks but
  196.      none are of the size of FidoNet.
  197.  
  198.      What Sysops will see and learn in SHARING will make it "the place to
  199.      be".  What might they see?  We don't have an agenda, hidden or
  200.      otherwise, but I'm sure we'll laugh at ourselves, cry for each other
  201.      and everyone will be "just themselves".
  202.  
  203.      The wonders and mysteries of FidoNet Technology will be explained!.
  204.      Learn how and where files fall into dishes.  Learn how and where files
  205.      pop out of the InterNet tunnel.  Learn how to eat the Echomail
  206.      Elephant rather than have it eat you! Learn how to send e-mail to
  207.      anywhere on the Globe often with a local phone call, including to AOL,
  208.      Prodigy, C$erve, and perhaps that office down the street.  Learn
  209.      there's no such thing as FREE!
  210.  
  211.      SHARE the wonderful experience of our hobby.  This echo is Awesome,
  212.      it's the Swiss Army knife of information!  There's plenty to learn,
  213.      learn, learn and it doesn't matter what hardware, operating system,
  214.      mailer or BBS Software a sysop uses.
  215.  
  216.      Sysops can share those funny experiences we all have.  One of mine is
  217.      a brief story about NERF.BAT.  We can laugh with each other and cry
  218.      with each other. Yes, tragedy, sometimes simple, sometimes bizarre,
  219.      hits us all.  Why not share it?  There's good news too: someone is a
  220.      new parent for the first time, a new grandparent even.
  221.  
  222.      Sysops find good deals on all sorts of goods and services.  Why not
  223.      SHARE where, how much and how good so we can all get the most out of
  224.      our hobby budget?
  225.  
  226.      Please!
  227.      --------
  228.      What won't you see?  No anger.  No threats.  No intimidation. No
  229.      inappropriate language.  And none of that politistrivial junk so
  230.      common in other Sysop Echos.  No Zeroes. No Binary addresses allowed.
  231.      If you're a 0 or a 1, stay away!  No a.k.a.'s allowed!  Use your real,
  232.      primary address or stay out, especially those 0's and 1's!  Simple
  233.      enough?
  234.  
  235.      Every Sysop in any FTN phonebook is invited to participate, invited to
  236.      just be themselves and SHARE the FidoNet Technology Experience.
  237.      FidoNet Technology Sysops are *very* special people, this echo is by,
  238.      for and about them.
  239.  
  240.      Politeness to each other is EXPECTED!  It IS REQUIRED.  Almost no
  241.      topic is "off topic" except as noted below.  The Moderators reserve
  242.      the right to BAN any topic.  I reserve the right to "suspend
  243.      discussion" of a topic for a time certain.  Be CERTAIN of your FACTS,
  244.      NAME names, don't take the cowards way out by using "They said, He
  245.      said".  Be SPECIFIC.
  246.  
  247.      There are three Moderators. They are:
  248.  
  249.      Christopher Baker;
  250.      jim barchuk; and
  251.      FIDONEWS 14-24               Page 5                   16 Jun 1997
  252.  
  253.  
  254.      Debra Milner.
  255.  
  256.      They are the only ones who will make Moderatorial pronouncements to
  257.      anyone else. Leave any Moderating to them. We don't expect to have to
  258.      do much in the way of moderating in a friendly and cooperative Echo.
  259.  
  260.      Definitions:
  261.      ------------
  262.  
  263.      Phonebook:   The Nodelist
  264.  
  265.      FidoGawd:    (or its variations) are prohibited from this Echo.
  266.  
  267.      Fight-o-Net: (or its variations) are also prohibited.
  268.  
  269.      Policy:      Policy4 as indicated in the nodelist. The ONLY policy.
  270.                   If your zone/region/net also has a local policy, be
  271.                   specific.  If you use Policy by itself, POLICY4 is
  272.                   assumed.
  273.  
  274.      EchoPol:     It does not exist, never did.  Maybe never will.
  275.  
  276.      CRP:         Cost Recovery Plan/Program.  I prefer Cost Sharing but
  277.                   you call it what you wish.
  278.  
  279.      CRaP:        Something that sometimes accompanies participation in a
  280.                   CRP.  It is and is intended to be an unflattering term.
  281.  
  282.      Grunt Sysop: All 30,000+ PEOPLE in the FidoNet phonebook and or any
  283.                   FTN phone book.
  284.  
  285.      backbone:    or Backbone, whichever you prefer is OFF Topic [except we
  286.                   will share how to get an Echo on the Backbone if asked.]
  287.  
  288.      As Mr. Bartles and Mr. James are known to say: Thank you for your
  289.                                                     support!
  290.  
  291.      Short and sweet (?)
  292.  
  293.      Moderator Emeritus, former 1:150/730
  294.  
  295.      B-)  Don
  296.  
  297.       QOFM.
  298.       Chris
  299.  
  300.       Christopher Baker
  301.       jim barchuk
  302.       Debra Milner
  303.       Moderators, SHARING
  304.  
  305.       -30-
  306.  
  307.      We invite you to join us in a friendly and cooperative Echo for
  308.      FidoNet Sysops who don't need someone standing over them with a club
  309.      to behave like adults and who would prefer a Sysop Echo with less tar.
  310.      FIDONEWS 14-24               Page 6                   16 Jun 1997
  311.  
  312.  
  313.      [grin]
  314.  
  315.      QOFM.
  316.      Chris
  317.  
  318.      -----------------------------------------------------------------
  319.  
  320.      FIDONEWS 14-24               Page 7                   16 Jun 1997
  321.  
  322.  
  323.      =================================================================
  324.                              GETTING TECHNICAL
  325.      =================================================================
  326.  
  327.  
  328.      [This is part of the continuing FidoNet History series of FTSC
  329.       Standards and Proposals. These docs have been reformatted to 70
  330.       columns where required. Tables may be askew. Node numbers and phone
  331.       numbers may be out of date.] Ed.
  332.  
  333.        | Document: FSC-0082
  334.        | Version:  001
  335.        | Date:     14 May 1995
  336.        |
  337.        | Stephan Slabihoud, 2:2446/110.6@fidonet.org
  338.  
  339.                            A Proposed New Packet Type
  340.                                 Stephan Slabihoud
  341.                             2:2446/110.6@fidonet.org
  342.                               90:400/410@nest.ftn
  343.                     slabih00@marvin.informatik.uni-dortmund.de
  344.                                1.Rev: Sep 20, 1994
  345.  
  346.           Status of this document
  347.           =======================
  348.  
  349.           This FSC suggests a proposed protocol for the FidoNet(r)
  350.           community, and requests discussion and suggestions for
  351.           improvements. Distribution of this document is unlimited.
  352.  
  353.           Fido and FidoNet are registered marks of Tom Jennings and
  354.           Fido Software.
  355.  
  356.           Purpose
  357.           =======
  358.  
  359.           This document should introduce a widely used standardised
  360.           extension to FTS-0001, like FTS-0006, 0007 and 0008 are, and
  361.           provides a new way to switch to a new more confortable bundling
  362.           method. I call this method XType-1. This is also more convenient
  363.           than FSC-0014 (an earlier binary-style msg proposal) and allows
  364.           multimedia extensions for further support (e.g. samples and
  365.           pictures like World-Wide-Webb). An example how to implement MM
  366.           extensions can be found at the end of this document. Note: This
  367.           proposal does not suggest how to implement MM extensions, it
  368.           should only demonstrate the flexibility of XType-1.
  369.  
  370.           Prologue
  371.           ========
  372.  
  373.           The new bundling method (XType-1) that document is introducing is
  374.           NOT backward compatible. So only new software packages may offer
  375.           this bundling method.
  376.  
  377.           Why introducing a new bundle format?
  378.           ====================================
  379.      FIDONEWS 14-24               Page 8                   16 Jun 1997
  380.  
  381.  
  382.           Well, FSC-0001, 0039, 0048 and 0045 are not very comfortable to
  383.           handle. Software must be very complex to process a Type-2 packet
  384.           and looking for control lines like SEEN-BYs, MSGIDs, REPLYs and
  385.           so on slows down the importing, processing and exporting of every
  386.           mail.
  387.  
  388.           How can I recognize a new XType-1 bundle?
  389.           =========================================
  390.  
  391.           XType-1 bundles are using a new extension "*.PKX" and not longer
  392.           "*.PKT". So software can recognize a reveived XType-1 packet in a
  393.           very easy way. Older software that do not know the XType-1
  394.           bundling method will not touch the file. But it is highly
  395.           recommended to send the XType-1 bundles only to nodes you know
  396.           about that they can process this new bundling method.  Filename
  397.           naming is the same as in FTSC-0001 explained. Only the extension
  398.           has been changed from "PKT" to "PKX".  For older software it is
  399.           possible to convert the XType-1 format in one of the older
  400.           formats like FSC-0001, 0039, 0048 and 0045.
  401.  
  402.           Packet Header
  403.           =============
  404.  
  405.             Offset
  406.            dec hex
  407.                    .-----------------------------------------------------.
  408.              0   0 | HeaderVersion      ($01) | I/M-Format           [1] |
  409.                    [2] +--------------------------+------------------------
  410.                    --+
  411.              2   2 | ProductCode          (*) | ProductCode          (*) |
  412.                    +--------------------------+--------------------------+
  413.              4   4 | Revision         (major) | Revision         (minor) |
  414.                    +--------------------------+--------------------------+
  415.              6   6 | origZone             (*) | origZone             (*) |
  416.                    +--------------------------+--------------------------+
  417.              8   8 | origNet              (*) | origNet              (*) |
  418.                    +--------------------------+--------------------------+
  419.             10   A | origNode             (*) | origNode             (*) |
  420.                    +--------------------------+--------------------------+
  421.             12   C | origPoint            (*) | origPoint            (*) |
  422.                    +--------------------------+--------------------------+
  423.             14   E | destZone             (*) | destZone             (*) |
  424.                    +--------------------------+--------------------------+
  425.             16  10 | destNet              (*) | destNet              (*) |
  426.                    +--------------------------+--------------------------+
  427.             18  12 | destNode             (*) | destNode             (*) |
  428.                    +--------------------------+--------------------------+
  429.             20  14 | destPoint            (*) | destPoint            (*) |
  430.                    +--------------------------+--------------------------+
  431.             22  16 |                      password                       |
  432.                    |                8 bytes, null padded                 |
  433.                    +--------------------------+--------------------------+
  434.             30  1E | Date/Time in POSIX 1003.1 format                (*) |
  435.                    | (4 bytes)                                           |
  436.                    [5] +--------------------------+------------------------
  437.                    --+
  438.      FIDONEWS 14-24               Page 9                   16 Jun 1997
  439.  
  440.  
  441.             34  22 | CapabilWord          (*) | CapabilWord          (*) |
  442.                    +--------------------------+--------------------------+
  443.             36  24 |          length of origNetwork (in bytes)       (*) |
  444.                    [3]
  445.                    +-----------------------------------------------------+
  446.             38  26 |  origNetwork, zero when "length of origNetwork"=0   |
  447.      [4]
  448.                    |           null padded to an even length             |
  449.                    +-----------------------------------------------------+
  450.             ~~  ~~ |          length of destNetwork (in bytes)       (*) |
  451.                    [3]
  452.                    +-----------------------------------------------------+
  453.             ~~  ~~ |  destNetwork, zero when "length of destNetwork"=0   |
  454.      [4]
  455.                    |           null padded to an even length             |
  456.                    +-----------------------------------------------------+
  457.             ~~  ~~ |                    zero or more                     |
  458.                    ~                        packed                       ~
  459.                    |                       messages                      |
  460.                    +--------------------------+--------------------------+
  461.             ~~  ~~ |      0      |      0     |      0      |      0     |
  462.                    '-----------------------------------------------------'
  463.  
  464.          (*) high-low-byte or low-high-byte according to I/M-Format-Flag
  465.              (see [1]).
  466.  
  467.          [1] This flag defines Intel ($00) or Motorola ($01) format.
  468.              Intel-Format stores low-byte first, Motorola-Format stores
  469.              high-byte first.
  470.          (2) HeaderVersion $01 means XType-1 ($02 means XType-2 and so on).
  471.          (3) Length of network domain (max. 64k characters). Zero, when
  472.              no network name is used, not known or your software does not
  473.              allow a 5D address. When this field is $0000 the next field
  474.              (the domain itself) will not be stored.
  475.          (4) Domain names are not case sensitive.
  476.          (5) POSIX 1003.1 format: Long integer containing the number of
  477.              seconds since the 1st of January 1970 (00:00:00).
  478.  
  479.          Packet       = PacketHeader  { PakdMessage }  $00 $00
  480.  
  481.          PacketHeader = $01            /* $01 means XType-1 header
  482.      */
  483.                         I/M-Format     /* $00=Intel format, $01=Motorola
  484.      format*/
  485.                         productCode    /* 0 for Fido, write to FTSC for
  486.      others */
  487.                         revision       /* revision or 0
  488.      */
  489.                         origZone       /* zone of pkt sender (otherwise
  490.      null)  */
  491.                         origNet        /* of packet, not of messages in
  492.      packet */
  493.                         origNode       /* zone of pkt sender (otherwise
  494.      null)  */
  495.                         origPoint      /* zone of pkt sender (otherwise
  496.      null)  */
  497.      FIDONEWS 14-24               Page 10                  16 Jun 1997
  498.  
  499.  
  500.                         destZone       /* zone of pkt receiver (otherwise
  501.      null)*/
  502.                         destNet        /* of packet, not of messages in
  503.      packet */
  504.                         destNode       /* of packet, not of messages in
  505.      packet */
  506.                         destPoint      /* of packet, not of messages in
  507.      packet */
  508.                         password       /* session pasword  (otherwise null)
  509.      */
  510.                         date           /* of packet creation, binary coded
  511.      */
  512.                         time           /* of packet creation, binary coded
  513.      */
  514.                         CapabilWord    /* bitvector of XType versions known
  515.      by */
  516.                                        /* orig. software
  517.      */
  518.                         origLength     /* length of orig domain
  519.      */
  520.                         origNetwork    /* network of pkt sender
  521.      */
  522.                         destLength     /* length of dest domain
  523.      */
  524.                         destNetwork    /* network of pkt receiver
  525.      */
  526.  
  527.  
  528.  
  529.                          msb           Capability Word               lsb
  530.           Node Supports  ------------FTSC Type Supported **)------------
  531.  
  532.                           U  S 14 13 12 11 10  9  8  7  6  5  4  3  2  1
  533.  
  534.           Type-N,XType-1  0  1  0  0  0  0  0  0  0  0  0  0  0  0  0  1
  535.  
  536.                           ^  ^
  537.                           |  +-- "S" Indicates nodes able to process type
  538.                           2,
  539.                           |      type 2+ or stone age style packets
  540.                           +----- "U" Indicates nodes able to process RFC-
  541.                                  822 bundles.
  542.                          ** - In the example bit definitions only XType-1
  543.                               is defined now. The rest are to be concidered
  544.                               "reserved by FTSC".
  545.  
  546.           Generating XType-1 bundles
  547.           ==========================
  548.  
  549.            Do we have a CW              Does CW indicate
  550.           stored for dest?  YES ---->   higher packets  YES ---> Generate
  551.      higher
  552.                 NO                       we support?                packet
  553.                 |                            NO
  554.                \|/                           |
  555.                 +-----<----------------------+
  556.      FIDONEWS 14-24               Page 11                  16 Jun 1997
  557.  
  558.  
  559.                 |
  560.            Fill header with all info
  561.                 |
  562.            Add Messages
  563.                 |
  564.            Terminate packet
  565.                 |
  566.            Send packet
  567.  
  568.  
  569.           Receiving bundles
  570.           =================
  571.  
  572.             Receiving a PKX? NO -------------> Old style (PKT) format
  573.                YES
  574.                 |
  575.             HeaderVersion = $01 NO -------------> Process XType-Other
  576.                YES
  577.                 |
  578.             Store CW
  579.                 |
  580.             Process
  581.  
  582.  
  583.           Packed Messages in the XType-1 bundle
  584.           =====================================
  585.  
  586.           To conserve space and eliminate fields which would be meaningless
  587.           if sent, messages are packed for transmission in a binary style.
  588.  
  589.           XType-1 uses two different styles, a netmail style and an
  590.           echomail style.
  591.  
  592.  
  593.                                   Packed Netmail Message
  594.             Offset
  595.            dec hex
  596.                    .-----------------------------------------------------.
  597.              0   0 |     0      |     1       | I/M-Format           [1] |
  598.                    +--------------------------+--------------------------+
  599.              2   2 | origZone             (*) | origZone             (*) |
  600.                    +--------------------------+--------------------------+
  601.              4   4 | origNet              (*) | origNet              (*) |
  602.                    +--------------------------+--------------------------+
  603.              6   6 | origNode             (*) | origNode             (*) |
  604.                    +--------------------------+--------------------------+
  605.              8   8 | origPoint            (*) | origPoint            (*) |
  606.                    +--------------------------+--------------------------+
  607.             10   A | destZone             (*) | destZone             (*) |
  608.                    +--------------------------+--------------------------+
  609.             12   C | destNet              (*) | destNet              (*) |
  610.                    +--------------------------+--------------------------+
  611.             14   E | destNode             (*) | destNode             (*) |
  612.                    +--------------------------+--------------------------+
  613.             16  10 | destPoint            (*) | destPoint            (*) |
  614.                    +--------------------------+--------------------------+
  615.      FIDONEWS 14-24               Page 12                  16 Jun 1997
  616.  
  617.  
  618.             18  12 | Attribute            (*) | Attribute            (*) |
  619.                    +--------------------------+--------------------------+
  620.             20  14 | cost                 (*) | cost                 (*) |
  621.                    +--------------------------+--------------------------+
  622.             22  16 | Time/Date string (20 characters)                    |
  623.                    [2]
  624.                    +-----------------------------------------------------+
  625.             42  2A |          length of origNetwork (in bytes)       (*) |
  626.                    [3]
  627.                    +-----------------------------------------------------+
  628.             44  2C |  origNetwork, zero when "length of origNetwork"=0   |
  629.                    |           null padded to an even length             |
  630.                    +-----------------------------------------------------+
  631.             ~~  ~~ |          length of destNetwork (in bytes)       (*) |
  632.                    [3]
  633.                    +-----------------------------------------------------+
  634.             ~~  ~~ |  destNetwork, zero when "length of destNetwork"=0   |
  635.                    |           null padded to an even length             |
  636.                    +-----------------------------------------------------+
  637.             ~~  ~~ |                  variable fields                    |
  638.                    ~                                                     ~
  639.                    |                                                     |
  640.                    `-----------------------------------------------------'
  641.  
  642.                                   Packed Echomail Message
  643.             Offset
  644.            dec hex
  645.                    .-----------------------------------------------------.
  646.              0   0 |     0      |     2       | I/M-Format           [1] |
  647.                    +--------------------------+--------------------------+
  648.              2   2 | Attribute            (*) | Attribute            (*) |
  649.                    +--------------------------+--------------------------+
  650.              4   4 | cost                 (*) | cost                 (*) |
  651.                    +--------------------------+--------------------------+
  652.              6   6 | Time/Date string (20 characters)                    |
  653.                    [2] +--------------------------+------------------------
  654.                    --+
  655.             26  1A |                  variable fields                    |
  656.                    ~                                                     ~
  657.                    |                                                     |
  658.                    `-----------------------------------------------------'
  659.  
  660.          (*) high-low-byte or low-high-byte according to I/M-Format-Flag
  661.              (see [1]).
  662.          [1] This flag defines Intel ($00) or Motorola ($01) format.
  663.              Intel-Format stores low-byte first, Motorola-Format stores
  664.              high-byte first. Date/Time always stored in the format above!
  665.          [2] Time/Date string (ascii format) Format (see FTS):
  666.              DAY [ ] MONTH [ ] JEAR [ ][ ] HOUR [:] MINUTE [:] SECOND [0]
  667.  
  668.                 DAY: [00] ... [31]
  669.               MONTH: [Jan|Feb|Mar|Apr|May|Jun|Jul|Aug|Sep|Oct|Nov|Dec]
  670.                JEAR: [00] ... [99]
  671.                HOUR: [00] ... [23]
  672.              MINUTE: [00] ... [59]
  673.              SECOND: [00] ... [59]
  674.      FIDONEWS 14-24               Page 13                  16 Jun 1997
  675.  
  676.  
  677.          (3) Length of network domain (max. 64k characters). Zero, when
  678.              no network name is used, not known or your software does not
  679.              allow a 5D address. When this field is $0000 the next field
  680.              (the domain itself) will not be stored.
  681.  
  682.  
  683.            Due to routing, the origin and destination net and node of a
  684.            packet are often quite different from those of the messages
  685.            within it, nor need the origin and destination nets and nodes of
  686.            the messages within a packet be homogenous.
  687.  
  688.            PakdMessage  = $01             /* $01 indicates packed netmail
  689.      message*/
  690.                           I/M-Format      /* $00=Intel, $01=Motorola-Format
  691.      */
  692.                           origZone        /* of message */
  693.                           origNet         /* of message */
  694.                           origNode        /* of message */
  695.                           origPoint       /* of message */
  696.                           destZone        /* of message */
  697.                           destNet         /* of message */
  698.                           destNode        /* of message */
  699.                           destPoint       /* of message */
  700.                           AttributeWord   /* as described in FTS-0001
  701.                           */
  702.                           cost            /* in lowest unit of originator's
  703.                           */
  704.                                           /* currency
  705.      */
  706.                           Date/Time       /* message body was last edited
  707.      */
  708.                           origLength      /* length of orig domain
  709.      */
  710.                           origNetwork     /* network of pkt sender
  711.      */
  712.                           destLength      /* length of dest domain
  713.      */
  714.                           destNetwork     /* network of pkt receiver
  715.      */
  716.  
  717.  
  718.            PakdMessage  = $02             /* $02 indicates packed echomail
  719.      message*/
  720.                           I/M-Format      /* $00=Intel, $01=Motorola-Format
  721.      */
  722.                           AttributeWord   /* as described in FTS-0001
  723.      */
  724.                           cost            /* in lowest unit of originator's
  725.      */
  726.                                           /* currency
  727.      */
  728.                           Date/Time       /* message body was last edited
  729.      */
  730.  
  731.  
  732.           AttributeWord   bit       meaning
  733.      FIDONEWS 14-24               Page 14                  16 Jun 1997
  734.  
  735.  
  736.                           ---       --------------------
  737.                             0  +    Private
  738.                             1  + s  Crash
  739.                             2       Recd
  740.                             3       Sent
  741.                             4  +    FileAttached
  742.                             5       InTransit
  743.                             6       Orphan
  744.                             7       KillSent
  745.                             8       Local
  746.                             9    s  HoldForPickup
  747.                            10  +    unused
  748.                            11    s  FileRequest
  749.                            12  + s  ReturnReceiptRequest
  750.                            13  + s  IsReturnReceipt
  751.                            14  + s  AuditRequest
  752.                            15    s  FileUpdateReq
  753.  
  754.                                  s - need not be recognized, but it's ok
  755.                                  + - not zeroed before packeting
  756.  
  757.           Bits numbers ascend with arithmetic significance of bit position.
  758.  
  759.           What is a variable field:
  760.           =========================
  761.  
  762.           A variable field consists of a header of four bytes length:
  763.  
  764.                    .-----------------------------------------------------.
  765.              0     | DATA LENGTH          (*) | DATA LENGTH          (*) |
  766.                    |               $0000 when last field                 |
  767.                    +--------------------------+--------------------------+
  768.              2     | FIELD-ID                                            |
  769.                    |           "ND" (end data) when last field           |
  770.                    +--------------------------+--------------------------+
  771.              4     | FIELD-DATA                                          |
  772.                    ~                                                     ~
  773.                    | zero padded to an even length                       |
  774.                    `-----------------------------------------------------'
  775.  
  776.  
  777.           Defined FIELD-ID's:
  778.           ===================
  779.  
  780.           FIELD-ID  -  synonym to
  781.           --------------------------------------------------------------
  782.                 FR     "From" user
  783.                 TO     "To" user
  784.                 SJ     "Subject"
  785.                 AR     AREA               (only used in echomails)
  786.                 PD     ^PID
  787.                 TD     ^TID
  788.                 ED     ^EID
  789.                 MD     ^MSGID
  790.                 RP     ^REPLY
  791.                 RT     ^REPLYTO           (used by uucp gateways)
  792.      FIDONEWS 14-24               Page 15                  16 Jun 1997
  793.  
  794.  
  795.                 RA     ^REPLYADDR         (used by uucp gateways)
  796.                 SN     ^SEEN-BY           (only used in echomails)
  797.                 VA     ^VIA               (only used in netmails)
  798.                 RN     ^REALNAME
  799.                 SP     ^SPLIT
  800.                 CS     ^CHARSET or ^CHRS
  801.                 OR     Origin             (only used in echomails)
  802.                 TL     Tearline
  803.                 ML     Mailtext follows
  804.                 ND     End of data fields
  805.           --------------------------------------------------------------
  806.            multimedia extensions (explanation follows):
  807.                 VO     audio data VOC format
  808.                 WA     audio data WAV format
  809.                 MI     MIDI data
  810.                 GF     bitmap data GIF
  811.                 TI     bitmap data TIFF
  812.                 JP     bitmap data JPEG
  813.                 AV     video data AVI
  814.           --------------------------------------------------------------
  815.            write to St.Slabihoud for more...
  816.  
  817.           All fields must have an even length. An odd field length must
  818.           be aligned to an even one with a padded 0.
  819.  
  820.            Field  = dataLength       /* of field data (incl. 0)   */
  821.                     fieldID          /* see table                 */
  822.                     fieldData        /* Field data                */
  823.  
  824.           Example (NetMail):
  825.           ==================
  826.  
  827.           From: Stephan Slabihoud on 2:2446/110.6
  828.           To  : Guenther Paczia on 2:2446/110
  829.           Subj: This is a testmail
  830.           -----------------------------------------
  831.           ^PID: AVALON 3.72
  832.           ^MSGID: 2:2446/110.6@fidonet.org a3dbcfe5
  833.           ^MYCTRL nothing interest
  834.           This is the message body
  835.  
  836.  
  837.                    .-----------------------------------------------------.
  838.                    | MESSAGE-HEADER                                      |
  839.                    ~                                                     ~
  840.                    |                                                     |
  841.                    +--------------------------+--------------------------+
  842.                    | PACKED NETMAIL MESSAGE HEADER                       |
  843.                    ~                                                     ~
  844.                    |                                                     |
  845.                    +--------------------------+--------------------------+
  846.                    |           18             |           0              |
  847.                    +--------------------------+--------------------------+
  848.                    |           'F'            |           'R'            |
  849.                    +--------------------------+--------------------------+
  850.                    |             'Stephan Slabihoud', $00                |
  851.      FIDONEWS 14-24               Page 16                  16 Jun 1997
  852.  
  853.  
  854.                    +--------------------------+--------------------------+
  855.                    |           16             |           0              |
  856.                    +--------------------------+--------------------------+
  857.                    |           'T'            |           'O'            |
  858.                    +--------------------------+--------------------------+
  859.                    |             'Guenther Paczia', $00                  |
  860.                    +--------------------------+--------------------------+
  861.                    |           18             |           0              |
  862.                    +--------------------------+--------------------------+
  863.                    |           'S'            |           'J'            |
  864.                    +--------------------------+--------------------------+
  865.                    |             'This is a testmail'                    |
  866.                    +--------------------------+--------------------------+
  867.                    |           12             |           0              |
  868.                    +--------------------------+--------------------------+
  869.                    |           'P'            |           'D'            |
  870.                    +--------------------------+--------------------------+
  871.                    |             'AVALON 3.72', $00                      |
  872.                    +--------------------------+--------------------------+
  873.                    |           34             |           0              |
  874.                    +--------------------------+--------------------------+
  875.                    |           'M'            |           'D'            |
  876.                    +--------------------------+--------------------------+
  877.                    |             '2:2446/110.6@fidonet.org a3dbcfe5\0'   |
  878.                    +--------------------------+--------------------------+
  879.                    |           50             |           0              |
  880.                    +--------------------------+--------------------------+
  881.                    |           'M'            |           'L'            |
  882.                    +--------------------------+--------------------------+
  883.                    | '^MYCTRL nothing interest', $0A                     |
  884.                    | 'This is the message body', $0A                     |
  885.                    +--------------------------+--------------------------+
  886.                    |           0              |           0              |
  887.                    +--------------------------+--------------------------+
  888.                    |           'N'            |           'D'            |
  889.                    +--------------------------+--------------------------+
  890.                    | more messages or zero                               |
  891.                    ~                                                     ~
  892.                    |                                                     |
  893.                    +--------------------------+--------------------------+
  894.                    |      0      |      0     |      0      |      0     |
  895.                    '-----------------------------------------------------'
  896.  
  897.           Unknown control lines are stores as usual in the message body. So
  898.           it is possible to receive a XType-1 packet and convert it into an
  899.           old style Type-2+ packet to send to it to another systems that do
  900.           not recognize the new Xtype-n bundles.
  901.  
  902.           Messages can be longer than 65535 bytes. Just use the 'ML' fields
  903.           more than once. When importing such a mail the importer can
  904.           easily split the mail into smaller parts. All 'ML' fields can be
  905.           added to one big mail, or each 'ML' text can be stored in its own
  906.           message.  According to older software each 'ML' field should not
  907.           be longer than 8 kbyte (but it is allowed to use longer fields!).
  908.  
  909.           All fields are unsigned integer.
  910.      FIDONEWS 14-24               Page 17                  16 Jun 1997
  911.  
  912.  
  913.           Example: How to implement MultiMedia extensions (draft version):
  914.           ================================================================
  915.  
  916.           Graphics and sounds are coded in one of the following fields:
  917.              Audio: VO,WA,MI
  918.              Bitmap: GF,TI,JP
  919.              Video: AV
  920.  
  921.           Each field-data starts with a multimedia header:
  922.  
  923.                     .------------------------.
  924.              0   0  | Name (Title)           |
  925.                     | 16 chars (zero padded) |
  926.                     +------------------------+
  927.             16  10  | ID                     |
  928.                     | 32bit Random Number    |
  929.                     +------------------------+
  930.             20  14  | Flags                  |
  931.                     | 16bit bitfield         |
  932.                     +------------------------+
  933.             22  16  | 42 reserved bytes      |
  934.                     |                        |
  935.                     +------------------------+
  936.             64  40  | start of data          |
  937.                     ~                        ~
  938.                     |                        |
  939.                     '------------------------'
  940.  
  941.             Flags:
  942.              Bit 0/1  - 1 = align left
  943.                         2 = align right
  944.                         3 = center
  945.                         0 = reserved
  946.              Bit 2-15 - reserved
  947.  
  948.  
  949.           There are some possibilties for a mail editor to show/play the
  950.           multimedia extensions:
  951.  
  952.            1. It shows the mail in the first window and a list of all
  953.               available fields in an extra (selection) window. The user
  954.               selects the picture/sound from the selection window.
  955.  
  956.            2. Pictures will be put together with the mailtext in ONE window
  957.               (a button will be shown when it is an audio field).  To
  958.               define the place where a picture (or other multimedia
  959.               extension) is shown put following ^A-control line into the
  960.               mailbody:     ^MMEDIA: <field> <id> [<infotext>]
  961.                 <field> is the "variable field" shortcut.
  962.                 <id> is the 32bit ID in hex from the multimedia header.
  963.                 <infotext> can be used as infotext for buttons.
  964.  
  965.           Example of ML field (mailbody):
  966.           ------------------------------------------------------------
  967.             Welcome to\n
  968.             ^MMEDIA: GF 5417fde6\n\n
  969.      FIDONEWS 14-24               Page 18                  16 Jun 1997
  970.  
  971.  
  972.             Please select:\n\n
  973.             To hear my voice click on the button:\n
  974.             ^MMEDIA: VO 2f4dca67 Say it\n
  975.             I am watching you ;-):\n
  976.             ^MMEDIA: GF 5627320f Click here\n
  977.           ------------------------------------------------------------
  978.  
  979.           This mail could be shown as follows:
  980.           ------------------------------------------------------------
  981.             Welcome to:
  982.             +------------------------------+
  983.             | GIF-Picture                  |
  984.             +------------------------------+
  985.  
  986.             Please select:
  987.  
  988.             To hear my voice:
  989.             +--------+
  990.             | Say it |
  991.             +--------+
  992.             I am watching you ;-):
  993.             +-------------+
  994.             |             |
  995.             | GIF-Picture |
  996.             |             |
  997.             +-------------+
  998.           ------------------------------------------------------------
  999.           Note: All pictures can be shown as button as well. This should
  1000.                 be switchable in the mail editor.
  1001.  
  1002.           Credits
  1003.           =======
  1004.  
  1005.           Thanx to Jonathan de Boyne Pollard, Peter Dreuw, Daniel Roesen
  1006.           and Rowan Crowe for their good ideas.
  1007.  
  1008.           Epilog
  1009.           ======
  1010.  
  1011.           That's all, now it's up to you to decide whether or not to
  1012.           implement it.
  1013.  
  1014.       -30-
  1015.  
  1016.  
  1017.      -----------------------------------------------------------------
  1018.  
  1019.  
  1020.        | Document: FSC-0083
  1021.        | Version:  001
  1022.        | Date:     17 June 1995
  1023.        |
  1024.        | Jonathan de Boyne Pollard, FIDONET#2:440/4.0
  1025.  
  1026.                  A proposed standard for message IDs on FTN systems.
  1027.  
  1028.      FIDONEWS 14-24               Page 19                  16 Jun 1997
  1029.  
  1030.  
  1031.                                         by
  1032.  
  1033.                     Jonathan de Boyne Pollard, FIDONET#2:440/4.0
  1034.  
  1035.                              Version 0.02, Sun 19950507
  1036.  
  1037.              This document is (c) Copyright 1995 Jonathan de Boyne Pollard,
  1038.              all rights reserved.  Originally written on Tuesday 19950131.
  1039.  
  1040.              Permission is hereby granted to copy and use this document
  1041.              without modification in any way that you see fit, provided
  1042.              that you do not attempt to make money from it, and that you
  1043.              understand that I take no responsibility whatsoever for any
  1044.              effect that it may have on your machine, data, marital status,
  1045.              or cat.
  1046.  
  1047.              Especial permission to freely use and redistribute this
  1048.              document in its original form is given to developers of FTN
  1049.              softwares and whatever FIDONET Technical Standards bodies may
  1050.              exist from time to time.
  1051.  
  1052.              -----------------------
  1053.       0.0 Definition of terms
  1054.              -----------------------
  1055.  
  1056.              This document assumes familiarity with several terms in common
  1057.              use in discussion of mail systems, such as `User Agent',
  1058.              `Message Transport Agent', and so forth.
  1059.  
  1060.              Robot mail programs qualify as UAs, incidentally.
  1061.  
  1062.              0.1 Knackered Backward Form
  1063.              ---------------------------
  1064.  
  1065.              This specification uses a modified BNF notation for discussion
  1066.              of textual representation of message IDs.
  1067.  
  1068.              Literal syntax elements (terminal nodes of the grammar) are
  1069.              enclosed in single quotes.
  1070.  
  1071.                      'MSGID:'        '@'    '<'        '"'
  1072.  
  1073.              Non-terminal nodes are enclosed in angle brackets (greater
  1074.              than and less then signs).
  1075.  
  1076.                      <quoted-text>           <hex-text>              <q-p-
  1077.      site-identifier>
  1078.  
  1079.              Production rules comprise a non-terminal, followed by
  1080.              productions.  Alternate productions for the same non-terminal
  1081.              are separated by a vertical bar.
  1082.  
  1083.                      <qtext-chars> ::=
  1084.                                        '"' '"'
  1085.                                      | <any-character-except-quotes-NUL-or-
  1086.      CR>
  1087.      FIDONEWS 14-24               Page 20                  16 Jun 1997
  1088.  
  1089.  
  1090.              Optional sequences within a production are indicated in two
  1091.              ways.  Square brackets enclose a sequence that may occur
  1092.              exactly once or not at all.
  1093.  
  1094.                      [ '@' <dns-name> ':' ]
  1095.  
  1096.              Curly braces enclose a sequence that may be repeated any
  1097.              number of times.  A leading numeric prefix (usually 0 or 1)
  1098.              indicates the minimum number of repetitions.
  1099.  
  1100.                      1*{ <hex-character> }
  1101.  
  1102.              0.1.1 Some standard production rules
  1103.              ------------------------------------
  1104.  
  1105.                      <whitespace-char> ::= <tab> | <space>
  1106.  
  1107.                      <whitespace> ::= 1*{ <whitespace-char> }
  1108.  
  1109.                      <hex-character> ::=
  1110.                              '0'|'1'|'2'|'3'|'4'|'5'|'6'|'7'|'8'|'9'|
  1111.                              'A'|'B'|'C'|'D'|'E'|'F'|
  1112.                              'a'|'b'|'c'|'d'|'e'|'f'
  1113.  
  1114.                      <upper-hex-char> ::=
  1115.  
  1116.      '0'|'1'|'2'|'3'|'4'|'5'|'6'|'7'|'8'|'9'|'A'|'B'|'C'|'D'|'E'|'F'
  1117.  
  1118.                      <qtext-char> ::=
  1119.                                      '"' '"'
  1120.                                | <any-ASCII-character-except-quotes-NUL-or-
  1121.      CR>
  1122.  
  1123.                      <quoted-text> ::= '"' 0*{ <qtext-char> } '"'
  1124.  
  1125.                      <quoted-char> ::=
  1126.                                      <any-ASCII-character-except-quotes-
  1127.      backslash-NUL-or-CR>
  1128.                                | '\' <any-ASCII-character-execpt-NUL-or-CR>
  1129.  
  1130.                      <quoted-string> ::= '"' 0*{ <quoted-char> } '"'
  1131.  
  1132.                      <word> ::= 1*{ <any-ASCII-character-above-SPACE-and-
  1133.      below-DEL> }
  1134.  
  1135.              Note the difference between the two forms of quoting.
  1136.              <quoted-text> is a string with embedded quotation marks
  1137.              represented by double quotation marks (the way that most BASIC
  1138.              languages do).  However, <quoted-string> is a string with all
  1139.              quotation marks and backslashes (and, indeed, any other
  1140.              character) escaped by the backslash character, in the style of
  1141.              the C and C++ languages.
  1142.  
  1143.              -------------------------------------
  1144.      1.0 Definition and use of message IDs
  1145.              -------------------------------------
  1146.      FIDONEWS 14-24               Page 21                  16 Jun 1997
  1147.  
  1148.  
  1149.              For the purposes of this document, the network is considered
  1150.              to form a vast distributed database of messages, which uses
  1151.              replication and store and forward distribution to ensure that
  1152.              all carriers of the database are kept up to date.  Every
  1153.              message, whether netmail or echomail, carries a primary
  1154.              message ID that uniquely identifies it, and zero or more
  1155.              reference message IDs that uniquely identify any messages that
  1156.              it refers to.
  1157.  
  1158.              A primary message ID is a globally unique key that is used for
  1159.              uniquely identifying any single given mail message in the
  1160.              database (that is, counting all replicas of a message over all
  1161.              of the network as "one").  The reference message IDs are used
  1162.              by user agents to form a reply graph, allowing the the user to
  1163.              easily navigate the messagebase.
  1164.  
  1165.              Message transport protocols may require the data in a message
  1166.              ID to be encoded so that it may be safely transported.  This
  1167.              standard distinguishes between the "underlying" message IDs
  1168.              and the encoded forms.  This chapter discusses the underlying
  1169.              message IDs and the concepts behind them without reference to
  1170.              a particular encoding, and subsequent chapters discuss the
  1171.              various encoded forms.
  1172.  
  1173.              1.1 Components of a message ID
  1174.              ------------------------------
  1175.  
  1176.              A message ID comprises two parts, namely a site identifier and
  1177.              a local part.  Both of these parts are arbitrary 8-bit binary
  1178.              data, that implementations are free to store in any way they
  1179.              choose, but which they should never alter.  There are no
  1180.              distinguished characters in either the site identifier or
  1181.              local part, especially not terminating characters.  So
  1182.              implementations must usually store an additional length count
  1183.              for both.
  1184.  
  1185.              The "minimum maximum" lengths for the site ID and local part
  1186.              are 64 octets each, and conforming implementations may not
  1187.              impose shorter maximum length restrictions.  In fact,
  1188.              implementations are encouraged to impose no length
  1189.              restrictions on message IDs whatsoever (for example, it is not
  1190.              unreasonable to expect site IDs to exceed 256 octets on
  1191.              occasion).
  1192.  
  1193.              1.2 Preservation of uniqueness
  1194.              ------------------------------
  1195.  
  1196.              A site that creates messages (by entering them into the
  1197.              distributed database) must also issue message IDs, and must
  1198.              ensure that the global uniqueness property of message IDs is
  1199.              preserved.
  1200.  
  1201.              A site MUST ensure that it issues unique local parts to
  1202.              individual messages.  Two or more sites may not have the same
  1203.              site identifier, unless they *all* co-operate to ensure that
  1204.              they do not issue duplicate local parts.
  1205.      FIDONEWS 14-24               Page 22                  16 Jun 1997
  1206.  
  1207.  
  1208.              The administrative procedures necessary to obtain a unique
  1209.              site identifier are beyond the scope of this document.
  1210.              Usually site identifiers will be FTN 5D addresses, or fully
  1211.              qualified DNS names, because administrative procedures for
  1212.              assigning such are already in place.  However, they are not
  1213.              restricted to be such.
  1214.  
  1215.              The means by which a site invents new local parts is beyond
  1216.              the scope of this document.  A discussion of some example
  1217.              options for implementors to consider is given in an appendix.
  1218.  
  1219.              1.3 Reference message IDs
  1220.              -------------------------
  1221.  
  1222.              Reference message IDs in a message denote messages to which it
  1223.              is related, comprising a "local subset" of the overall reply
  1224.              graph (i.e. the direct and indirect ancestors of the message),
  1225.              which each message carries around with it.
  1226.  
  1227.              Carrying around multiple reference message IDs provides
  1228.              overlap, allowing for the overall reply graph to be
  1229.              reconstructed even in the absence of intermediate messages (if
  1230.              they had expired, or had not yet arrived due to propagation
  1231.              lag, for example).
  1232.  
  1233.              UAs that conform to this standard MUST ensure that only
  1234.              messages that start new threads (i.e. messages entered into
  1235.              the network not in response to any existing message) have no
  1236.              reference message IDs.
  1237.  
  1238.              All other messages that they create MUST contain at least one
  1239.              reference message ID, being that of the message that is being
  1240.              responded to.
  1241.  
  1242.              [[ Luckily, schemes already in existence mean that in practice
  1243.                 non-conforming User Agents will generally preserve this
  1244.                 single back link, as well.  ]]
  1245.  
  1246.              When responding to a message, user agents must create the
  1247.              reference message ID list of the response by taking the list
  1248.              of reference message IDs from the original message, and
  1249.              appending the primary message ID of the original message to
  1250.              the tail.
  1251.  
  1252.              A reference message ID list should not be truncated, unless
  1253.              transport or storage limitations are in danger of being
  1254.              exceeded.  In which case, message IDs may only be removed from
  1255.              the head of the list.  Removing from the tail would eliminate
  1256.              links to immediate ancestor messages, and removing from the
  1257.              middle would alter the reply graph.
  1258.  
  1259.  
  1260.      ----------------------------------------------------------------------
  1261.      --
  1262.      2.0 Quoted printable encoding for storing 8-bit data in 7-bit
  1263.              transports
  1264.      FIDONEWS 14-24               Page 23                  16 Jun 1997
  1265.  
  1266.  
  1267.              --------------------------------------------------------------
  1268.              ----------
  1269.  
  1270.              To encode the 8-bit data in message IDs for transport by 7-bit
  1271.              transport layers, we use a variation on the widely used Quoted
  1272.              Printable form [RFC1521] [RFC1522].
  1273.  
  1274.              2.1 Grammar of Quoted Printable encoding
  1275.              ----------------------------------------
  1276.  
  1277.              The grammar of the 7-bit encoding of 8-bit data in a quoted
  1278.              printable word is as follows.
  1279.  
  1280.                      <q-p-word> ::=
  1281.                                      <word>
  1282.                                | <quoted-text>
  1283.                                | [ '=' ] 1*{ <q-p-character> } [ '=' ]
  1284.  
  1285.                      <q-p-character} ::=
  1286.                                      <any-ASCII-character-bar-ctls-wspace-
  1287.      quote-and-equals>
  1288.                                | <q-p-quoted-char>
  1289.  
  1290.                      <q-p-quoted-char> ::= '=' <upper-hex-char> <upper-hex-
  1291.      char>
  1292.  
  1293.              2.2 Conversion from 8-bit to 7-bit
  1294.              ----------------------------------
  1295.  
  1296.              Rule #1 (non-quoted transparent 7-bit): Where the 8-bit data
  1297.              consist of nothing but ASCII characters above SPACE and below
  1298.              DEL, they may be copied literally to the 7-bit representation.
  1299.  
  1300.              Rule #2 (quoted transparent 7-bit):  Where the 8-bit data
  1301.              consist of nothing but ASCII characters except CR and NUL,
  1302.              they may be converted to the 7-bit representation by enclosing
  1303.              them in quotes, and escaping every embedded quotation mark
  1304.              with a second quotation mark.
  1305.  
  1306.              Rule #3 (8-bit quoted): Where the 8-bit data contain CR or
  1307.              NUL, or any non-ASCII characters, they are converted to a 7-
  1308.              bit representation in two stages.
  1309.  
  1310.                      Firstly, all non-ASCII characters, all ASCII control
  1311.              characters, SPACE, DEL, '"', and '=', are converted to
  1312.              "quoted" form.  Quoted form is an '=' character followed by
  1313.              the hexadecimal value of the character represented as two
  1314.              uppercase hexadecimal digits.
  1315.  
  1316.                      Secondly, the entire string is then enclosed by one
  1317.              leading and one trailing '=' character.
  1318.  
  1319.              2.3 Conversion from 7-bit to 8-bit
  1320.              ----------------------------------
  1321.  
  1322.              Where the 7-bit field is delimited by equals signs, it is a
  1323.      FIDONEWS 14-24               Page 24                  16 Jun 1997
  1324.  
  1325.  
  1326.              fair bet that it comprises 8-bit data to which Rule 3 has been
  1327.              applied.  However, it is possible that sites in the 7-bit
  1328.              world may produce data with leading and trailing equals signs.
  1329.  
  1330.              Reverse of Rule #3 : If, after stripping the leading and
  1331.              trailing '=', the remaining text can be converted back using
  1332.              the reverse of Rule 3, then that 8-bit data is the actual
  1333.              message ID.  Otherwise the reverse of Rule 2 should be applied
  1334.              to the original 7-bit data.
  1335.  
  1336.              Reverse of Rule #2 : If the 7-bit data are enclosed by quotes
  1337.              the reverse of Rule 2 should be applied to remove the
  1338.              enclosing quotes and any embedded quotes (8-bit form does not
  1339.              have delimiter characters and so does not require quoting).
  1340.              Otherwise the reverse of Rule 1 should be applied.
  1341.  
  1342.              Reverse of Rule #1 : The 7-bit data are copied to the 8-bit
  1343.              data.
  1344.  
  1345.              2.4 Rationale
  1346.              -------------
  1347.  
  1348.              The intention is that <q-p-word> tokens will not be parsed as
  1349.              separate words by most 7-bit grammars.  The elimination of
  1350.              quotes, whitespace, and control characters by Rule 3 is part
  1351.              of achieving this.
  1352.  
  1353.              Rules 1 and 2 allow message IDs created by 7-bit standards to
  1354.              enter and travel within the 8-bit world, and be restored to
  1355.              their original form when they return to the 7-bit world.
  1356.              Returning 7-bit message IDs to their original form means that
  1357.              7-bit duplicate checking is not broken by 8-bit gateways.
  1358.  
  1359.              The unfortunate side-effect is that any 8-bit data generated
  1360.              in the 7-bit world will be returned to the 7-bit world as 7-
  1361.              bit data in Q-P encoded form.  However, the original 8-bit
  1362.              data are unlikely to work in the 7-bit world in the first
  1363.              place, so this is no great loss.
  1364.  
  1365.              Rule 3 is the most general rule of the three.  Rule 3 applies
  1366.              to true 8-bit message IDs generated in the 8-bit world that
  1367.              use 8-bit characters, allowing them to travel across the 7-bit
  1368.              world with a reasonable chance of remaining intact.
  1369.  
  1370.              The elimination of the equals sign by Rule 3, replacing it
  1371.              with its Q-P encoding, ensures that the decoding process can
  1372.              assume that an equals sign not followed by two uppercase hex
  1373.              characters is not a valid Rule 3 encoding, and so fall back to
  1374.              decoding Rule 2.
  1375.  
  1376.      ---------------------------------------------------------------------
  1377.      3.0 Storage of message IDs in type 2.0, 2.0+, and 2.2 message packets
  1378.              --------------------------------------------------------------
  1379.              -------
  1380.  
  1381.              Type 2.0 message packets [FTS0001], type 2.0+ message packets
  1382.      FIDONEWS 14-24               Page 25                  16 Jun 1997
  1383.  
  1384.  
  1385.              [FSC0039], and type 2.2 message packets [FSC0045] are used for
  1386.              message transport over much of FIDONET.  They do not have
  1387.              space in their message headers available for message IDs
  1388.              (along with a lot of other things), therefore message IDs must
  1389.              be transferred to the body of the message for transport in
  1390.              these forms, and retrieved from the body of the message
  1391.              afterwards.
  1392.  
  1393.              The existing "kludge line" mechanisms [FSC0068] are used to do
  1394.              this.
  1395.  
  1396.              There are two concerns here.
  1397.  
  1398.              Firstly, it is preferable that as much of the reply graph as
  1399.              possible is preserved, even in the face of tools that use
  1400.              existing MSGID/REPLY schemes [FTS0009].
  1401.  
  1402.              Secondly, message IDs are 8 bit data, and must be encoded into
  1403.              a 7-bit form that will be reliably transported in the bodies
  1404.              of type 2.0, 2.0+, and 2.2 message packets.
  1405.  
  1406.              3.1 Conversion to and from kludge lines
  1407.              ---------------------------------------
  1408.  
  1409.              The primary message ID of a message is stored to and retrieved
  1410.              from a "MSGID:" kludge line.
  1411.  
  1412.              All of the reference message IDs of a message are stored, in
  1413.              order from first to last, in a single "REFER:" kludge line.
  1414.              The last reference message ID of a message (its immediate
  1415.              ancestor, in other words) is stored in a "REPLY:" kludge line.
  1416.              Note that the information in the "REFER:" kludge line is a
  1417.              superset of the information in the "REPLY:" kludge line.
  1418.  
  1419.              If a message has zero reference message IDs (it is the start
  1420.              of a new thread), then the "REFER:" and "REPLY:" kludge lines
  1421.              are omitted.
  1422.  
  1423.              If, upon decoding from type 2.0, 2.0+, or 2.2 message
  1424.              transport format, a "REFER:" kludge line exists, then its
  1425.              contents are assumed to be the complete list of reference
  1426.              message IDs (in encoded form) for the message, and the
  1427.              "REPLY:" kludge line is ignored.  Otherwise, the content of
  1428.              the "REPLY:" kludge line (if any) is used for the single
  1429.              reference message ID of the message.
  1430.  
  1431.              3.2 Compatibility with existing MSGID/REPLY schemes
  1432.              ---------------------------------------------------
  1433.  
  1434.              There are two compatibility considerations.  It is important
  1435.              that encoded message IDs be correctly parsed by
  1436.              implementations using older less versatile standards.  It is
  1437.              also important that implementations expecting older
  1438.              MSGID/REPLY pairs will destroy as little linking information
  1439.              as possible.
  1440.  
  1441.      FIDONEWS 14-24               Page 26                  16 Jun 1997
  1442.  
  1443.  
  1444.              3.2.1 Grammar considerations
  1445.              ----------------------------
  1446.  
  1447.              There are two valid interpretations of FTS-0009, both of which
  1448.              (should) use the following grammar :
  1449.  
  1450.                      <msgid> ::= <soh> 'MSGID: ' <address-text>
  1451.      <whitespace> <hex-text>
  1452.                      <reply> ::=     <soh> 'REPLY: ' <address-text>
  1453.      <whitespace> <hex-text>
  1454.  
  1455.                      <soh> ::= ASCII SOH character
  1456.                      <address-text> ::= <quoted-text> | <word>
  1457.                      <hex-text> ::= 1*{ <hex-character> }
  1458.  
  1459.              The "VFIDO" interpretation assumes that MSGID/REPLY kludges
  1460.              are the textual representation of an (address, number) ordered
  1461.              pair.  Systems using this interpretation may change the case
  1462.              of <hex-text> or may renormalise <quoted-text> if they find it
  1463.              to be a FTN 5D address.
  1464.  
  1465.              Message IDs from this standard that are stored in MSGID/REPLY
  1466.              kludges will be mangled by software applying the VFIDO
  1467.              interpretation of FTS-0009.  Such software is not compatible
  1468.              with this standard.
  1469.  
  1470.              The "Mark Kimes" interpretation assumes that MSGID/REPLY
  1471.              kludges are text separated by whitespace, and preserves the
  1472.              contents of <quoted-text> and <hex-text> without change.
  1473.  
  1474.              The encoding scheme outlined in section 2.2 produces two
  1475.              whitespace separated text fields.  So software applying the
  1476.              "Mark Kimes" interpretation of FTS-0009 will not mangle the
  1477.              encoded message IDs.
  1478.  
  1479.              In many cases, softwares using the "Mark Kimes" interpretation
  1480.              will in fact parse <hex-text> as
  1481.  
  1482.                      <hex-text> ::= <word>
  1483.  
  1484.              As long as software applying the "Mark Kimes" interpretation
  1485.              of FTS-0009 is not written to truncate either field, or
  1486.              complain about a non-numeric <hex-text> portion, it is
  1487.              compatible with this standard.
  1488.  
  1489.              3.2.2 Reply linking
  1490.              -------------------
  1491.  
  1492.              FTS-0009 implementations will generate MSGID kludges, transfer
  1493.              the content (Mark Kimes interpretation) of the MSGID kludge
  1494.              data of an original message into the REPLY data of a response
  1495.              message, and will not generate a REFER kludge.
  1496.  
  1497.              So reply linking will be preserved, but reference information
  1498.              beyond the immediate ancestor of a message will be lost.
  1499.  
  1500.      FIDONEWS 14-24               Page 27                  16 Jun 1997
  1501.  
  1502.  
  1503.              3.3 Quoted printable encoding
  1504.              -----------------------------
  1505.  
  1506.              The 8-bit data in message IDs is encoded into 7-bit
  1507.              MSGID/REPLY data for transport in type 2.0, 2.0+, and 2.2
  1508.              message packets by using the quoted printable encoding
  1509.              outlined in chapter 2, along with the following grammar.
  1510.  
  1511.                      <msgid> ::= <soh> 'MSGID: ' <7-bit-encoding>
  1512.                      <reply> ::= <soh> 'REPLY: ' <7-bit-encoding>
  1513.                      <refer> ::= <soh> 'REFER: '
  1514.                                                      <7-bit-encoding> 0*{
  1515.      <whitespace> <7-bit-encoding> }
  1516.  
  1517.                      <7-bit-encoding> ::= <q-p-site-ID> <whitespace> <q-p-
  1518.      local-part>
  1519.  
  1520.                      <q-p-site-ID> ::= <q-p-word>
  1521.                      <q-p-local-part> ::= <q-p-word>
  1522.  
  1523.              Applying Rule 1 of Q-P encoding to local parts is safe as long
  1524.              as <hex-text> (from the FTS-0009 grammar) is in actuality
  1525.              treated as <word> by most implementations, as outlined in the
  1526.              compatibility notes.
  1527.  
  1528.              Rule 2 should not be applied to local parts, because the
  1529.              grammar of FTS-0009 does not allow for quoted text in the
  1530.              <hex-text> portion.
  1531.  
  1532.              The restrictions in Rule 3 have deliberate effect here.  FTS-
  1533.              0009 sites will rarely produce data with leading and trailing
  1534.              equals signs, so reversing Rule 3 will be unlikely to be
  1535.              subject to spurious data.  In theory, relaxing Rule 3 reversal
  1536.              to include decoding lowercase hexadecimal as well as uppercase
  1537.              hexadecimal would mean that sites that convert the case of
  1538.              MSGID/REPLY (as part of the "VFIDO" interpretation) would not
  1539.              break Q-P encoding.
  1540.  
  1541.              However, the "VFIDO" interpretation will usually do far more
  1542.              damage than simple case conversion, which will be impossible
  1543.              to restore.  Rather than attempt the reverse conversion (which
  1544.              could have the undesirable effect of causing different
  1545.              messages to end up with the same 8-bit message ID if the local
  1546.              part were truncated to eight characters in the 7-bit world),
  1547.              any "VFIDO" mangling that occurs will prevent Q-P decoding
  1548.              from succeeding.
  1549.  
  1550.              This means that 8-bit message IDs that look like incomplete or
  1551.              damaged Q-P encodings are not gateway problems, but are more
  1552.              likely to be the result of a site using the "VFIDO"
  1553.              interpretation in the 7-bit world.
  1554.  
  1555.              ------------------------------------------------------
  1556.       4.0 Storage of message IDs in type 2.3 message packets
  1557.              ------------------------------------------------------
  1558.  
  1559.      FIDONEWS 14-24               Page 28                  16 Jun 1997
  1560.  
  1561.  
  1562.              The storage format of type 2.3 messages (so-called "extensible
  1563.              type 2" [TYPE2EXT]) provides space in the message headers for
  1564.              both a primary message ID and an arbitrary list of reference
  1565.              message IDs.
  1566.  
  1567.              All message IDs are stored as 8-bit binary strings, using
  1568.              length counts rather than delimiters.  Therefore message IDs
  1569.              can be stored directly in type 2.3 messages.
  1570.  
  1571.              ------------------------------------------------------
  1572.       5.0 Storage of message IDs in type 3.x message packets
  1573.              ------------------------------------------------------
  1574.  
  1575.              There is such a wide variety of type 3 message formats that
  1576.              this standard doesn't hope to cover them all.
  1577.  
  1578.              For those with binary "chunks", chunk types 'PMID' (primary
  1579.              message ID) and 'RFER' (reference message IDs) are expected to
  1580.              have the following form :
  1581.  
  1582.               |-----------------------------------------------------|
  1583.               | Length of site identifier
  1584.               WORD32 |
  1585.               |-----------------------------------------------------|
  1586.               | Site identifider
  1587.               ...   |
  1588.               |-----------------------------------------------------|
  1589.               | Length of local part
  1590.               WORD32 |
  1591.               |-----------------------------------------------------|
  1592.               | Local part
  1593.               ...   |
  1594.               |-----------------------------------------------------|
  1595.  
  1596.              Those schemes that use text format headers and require field
  1597.              delimiters may care to use the Q-P encoding outlined in
  1598.              chapter 2.
  1599.  
  1600.              ---------------------------------------------------------
  1601.       6.0 Storage of message IDs in RFC822 and RFC1036 messages
  1602.              ---------------------------------------------------------
  1603.  
  1604.              The grammar of "Internet" messages is defined by the standards
  1605.              for ARPA text messages [RFC0822] and for Usenet news messages
  1606.              [RFC1036].
  1607.  
  1608.              6.1 Restrictions on interconversion
  1609.              -----------------------------------
  1610.  
  1611.              Interconversion between a FIDO message ID and an RFC822
  1612.              Message-ID is restricted by several factors.  The major factor
  1613.              is that RFC0822 actually places greater restrictions upon
  1614.              Message-IDs than this standard does upon FIDO message IDs (in
  1615.              part because this standard is designed to also be able to
  1616.              handle X.400 message identifiers and others transparently as
  1617.              well).  It mandates that the <address> portion of a Message-ID
  1618.      FIDONEWS 14-24               Page 29                  16 Jun 1997
  1619.  
  1620.  
  1621.              be a valid DNS name.
  1622.  
  1623.              A secondary factor is reversibility, in that many gateways
  1624.              exist between FTN and RFC822, and so message IDs that cross
  1625.              the boundary more than once will retain as much of their
  1626.              original ID information as possible.  There is more
  1627.              information contained within a FIDO message ID than in an
  1628.              RFC822 Message-ID.  In particular, the <address> portions of
  1629.              RFC822 Message-IDs are not case sensitive, whereas the site ID
  1630.              of a FIDO message ID is treated as 8-bit data for the purposes
  1631.              of comparison.
  1632.  
  1633.              These are handled by restricting the allowable conversions
  1634.              that a conformant gateway may use on a message ID, by ensuring
  1635.              that all of the FIDO information is not lost when converted to
  1636.              the (narrower bandwidth) RFC822 Message-ID format, and by
  1637.              allowing gateway softwares to infer a meaning from the site
  1638.              identifier portion of a message ID.
  1639.  
  1640.              This is the *only* part of this standard where it is allowed
  1641.              for softwares to place a meaning on the site identifier of a
  1642.              message ID.
  1643.  
  1644.              6.1 Converting to RFC822 form
  1645.              -----------------------------
  1646.  
  1647.              6.1.1 Site identifier recognition
  1648.              ---------------------------------
  1649.  
  1650.              Gateway softwares are allowed to examine a site identifier of
  1651.              a message ID and determine whether it is in a format that they
  1652.              recognise or not.  This standard specifies what gateway
  1653.              softwares should do when they encounter a site identifier that
  1654.              is a recognisable DNS name or one that is recognisable FIDO 5D
  1655.              address, and what form the DNS name for RFC822 must take.
  1656.  
  1657.              Site identifiers that are not FIDO 5D addresses are really
  1658.              beyond the scope of FIDONET documentation.  If an
  1659.              implementation recognises another form of site identifier
  1660.              (such as X.400 O/R addresses) then it is free to translate
  1661.              that site identifier to and from DNS form, as long as it knows
  1662.              how (there are RFCs on how to perform X.400 conversion).
  1663.  
  1664.              This message ID standard imposes no restrictions on site
  1665.              identifiers, allowing any scheme to be administered on
  1666.              FIDONET.  It is therefore up to the site identification
  1667.              schemes themselves to provide their own mappings to and from
  1668.              DNS names.
  1669.  
  1670.              Gateways are free to drop messages with message IDs that they
  1671.              do not understand how to convert.  Both the FIDONET and RFC
  1672.              worlds depend heavily upon message IDs for detecting messages
  1673.              duplicates, and so it is better that a gateway should NOT
  1674.              distribute messages with message ID formats that it doesn't
  1675.              understand how to convert to RFC822 form, rather than that it
  1676.              does so incorrectly.
  1677.      FIDONEWS 14-24               Page 30                  16 Jun 1997
  1678.  
  1679.  
  1680.              6.1.1.1 Site identifiers that are DNS names
  1681.              -------------------------------------------
  1682.  
  1683.              If the site identifier of a message ID can be parsed as a
  1684.              legal DNS name according to the grammar of RFC822 then, even
  1685.              if it cannot be resolved to an IP address or MX record, it
  1686.              must be used as the domain name of the RFC message ID, and the
  1687.              local part must be passed through unchanged.
  1688.  
  1689.              This allows for RFC message IDs to enter and leave 8-bit
  1690.              FIDONET without change, even via gateways that have no
  1691.              knowledge of or connectivity to the originating RFC host.
  1692.  
  1693.              6.1.1.2 Site identifiers that are FIDO 5D addresses
  1694.              ---------------------------------------------------
  1695.  
  1696.              The conversion process for message IDs where the site
  1697.              identifier can be parsed as a FIDO 5D address in the forms
  1698.              DOMAIN#Z:N/N.P or Z:N/N.P@DOMAIN depends from the "domain" (in
  1699.              the FIDO sense of the word) of the address.
  1700.  
  1701.              6.1.1.2.1 Site identifiers that are 5D addresses in FIDONET
  1702.              -----------------------------------------------------------
  1703.  
  1704.              If the site identifier of a message ID is parseable as a FIDO
  1705.              5D address of the form Z:N/N.P@FIDONET or FIDONET#Z:N/N.P
  1706.              (i.e.  in the FIDONET domain itself), then the DNS name used
  1707.              for the RFC message ID must be the DNS equivalent of that
  1708.              address.
  1709.  
  1710.              This is because MX records exist in the DNS for all of the
  1711.              zone:net pairs for 5D addresses in the FIDONET "domain", in
  1712.              the form
  1713.  
  1714.                      p#.f#.n#.z#.fidonet.org
  1715.  
  1716.              where # is a number without leading zeroes giving the
  1717.              appropriate portion of the 5D address.  Therefore this is the
  1718.              conversion that must be used.
  1719.  
  1720.              6.1.1.2.2 Site identifiers that are 5D addresses outside of
  1721.              FIDONET
  1722.              --------------------------------------------------------------
  1723.              -----
  1724.  
  1725.              Most other "domains" (in the FIDO sense of the word), are free
  1726.              to choose their own DNS domain name, but have not yet done so.
  1727.  
  1728.              Therefore, constructs such as p3.f0.n444.z81.os2net.ftn (which
  1729.              several people have INCORRECTLY inferred from other FTS
  1730.              documentation) are NOT ALLOWED as the DNS name in an RFC
  1731.              Message-ID.  .ftn is NOT a valid top-level DNS domain, for a
  1732.              start, and there is no guarantee that OS2NET would adopt that
  1733.              DNS name, either.
  1734.  
  1735.              (p#.f#.n#.z#.os2net.fidonet.org anyone ?)
  1736.      FIDONEWS 14-24               Page 31                  16 Jun 1997
  1737.  
  1738.  
  1739.              6.1.1.2.3 Conversion of local parts
  1740.              -----------------------------------
  1741.  
  1742.              Where a gateway has recognised a site identifier to represent
  1743.              a FIDO 5D address that it knows the DNS name for, the local
  1744.              part must then be encoded.
  1745.  
  1746.              According to the grammar in RFC822, any ASCII character (from
  1747.              NUL to DEL) is legal in the local part of an RFC822 Message-
  1748.              ID, because <quoted-pair> (q.v.) allows any special characters
  1749.              to be escaped.
  1750.  
  1751.              Since RFC822 transport is merely 7-bit just like type 2.0,
  1752.              2.0, and 2.2 message packets are, we use the quoted-printable
  1753.              scheme given in chapter 2.
  1754.  
  1755.              However,
  1756.  
  1757.              6.1.1.3 Site identifiers that are not recognisable 5D
  1758.              addresses
  1759.              --------------------------------------------------------------
  1760.              -
  1761.  
  1762.              No implementation may extend the FIDO 5D address to DNS name
  1763.              conversions for site IDs that are given above.  If the message
  1764.              ID is "almost, but not quite" a FIDO 5D address, then the
  1765.              message should for preference be discarded at the gateway
  1766.              rather than being passed through.
  1767.  
  1768.              Message IDs with abitrary site identifiers are perfectly
  1769.              acceptable to this standard, since it ascribes no meaning to
  1770.              site identifiers within FIDONET.  However, RFC822 and the
  1771.              existing RFC domain name system can only handle a restricted
  1772.              subset of the whole range of FIDO 5D addresses.
  1773.  
  1774.              6.1.1.4 Other site identifiers
  1775.              ------------------------------
  1776.  
  1777.              As mentioned before, gateways are allowed to support other
  1778.              site identification schemes that are not FIDO 5D addresses,
  1779.              and convert site identifiers in those forms to DNS names as
  1780.              they please.
  1781.  
  1782.              It should be borne in mind when designing such conversion
  1783.              schemes that the domain part of an RFC 822 message ID can only
  1784.              contain ASCII characters that are not control characters,
  1785.              whitespace, or special delimiter characters, because of the
  1786.              definition of <atom> in that standard (q.v.).  The quoted
  1787.              printable encoding outlined in chapter 2 of this document is
  1788.              probably not sufficient for handling full 8-bit site
  1789.              identifier schemes, in which case the scheme in RFC1522 should
  1790.              be investigated.
  1791.  
  1792.              6.1.2 Preserving information
  1793.              ----------------------------
  1794.  
  1795.      FIDONEWS 14-24               Page 32                  16 Jun 1997
  1796.  
  1797.  
  1798.              Although this standard recognises two forms for a FIDO 5D
  1799.              address, there is only one valid form for that address in the
  1800.              DNS.  For reverse conversions to succeed, when an RFC message
  1801.              re-enters 8-bit FIDONET (possibly via another gateway), the
  1802.              *exact form* of the original site identifier must be
  1803.              reconstructed, otherwise FIDO softwares will treat the two
  1804.              message IDs as different.
  1805.  
  1806.              Although other schemes exist, which encode the 5D address in
  1807.              the local part, and use a "generic" domain name of
  1808.              "fidonet.org" (which is not a valid host name), it is
  1809.              preferred that the semantics of a message ID ("WXYZ local part
  1810.              generated at ABCDE site") be preserved, especially as FIDONET
  1811.              sites are visible to the RFC world via the DNS anyway.
  1812.  
  1813.              It is therefore suggested that the original FIDONET site
  1814.              identifier (since it will be 7-bit text) be encoded as a
  1815.              <comment> token immediately following the relevant message ID,
  1816.              using quoting to escape any embedded punctuation (q.v. the
  1817.              grammar in RFC 822).
  1818.  
  1819.              6.2 Converting from RFC822 form
  1820.              -------------------------------
  1821.  
  1822.              When converting from RFC822 form back to 8-bit FIDONET message
  1823.              IDs, gateways should determine whether the address portion of
  1824.              the Message-ID is a hostname under the fidonet.org domain.
  1825.  
  1826.              If it is, a comment token should be scanned for to find the
  1827.              original form of the 5D address, and the site identifier
  1828.              should be reconstructed from it if found, or from the given
  1829.              DNS name in the form DOMAIN#Z:N/N.P if no comment token were
  1830.              present.  The inverse of the quoted printable encoding
  1831.              outlined in chapter 2 should then be applied to the local
  1832.              part.
  1833.  
  1834.              Otherwise, the 7-bit RFC822 Message-ID should be stored in the
  1835.              8-bit FIDONET message ID without change.
  1836.  
  1837.              6.3 Reply linking
  1838.              -----------------
  1839.  
  1840.              According to RFC1036, message IDs occur in the Message-ID:
  1841.              and in the References:  header for news (echomail).  Although
  1842.              RFC822 specifies an In-Reply-To:  header for mail (netmail),
  1843.              it makes it difficult to use, because it need not contain a
  1844.              message ID.
  1845.  
  1846.              The model for message identification used by RFC1036 closely
  1847.              matches the model outlined in this standard (it is probable
  1848.              that there is only one way to skin this particular cat).
  1849.              There is thus a direct mapping between the primary message ID
  1850.              defined by this standard and the RFC1036 Message-ID:  header,
  1851.              and also between the reference message IDs defined by this
  1852.              standard and the RFC1036 References: header.
  1853.  
  1854.      FIDONEWS 14-24               Page 33                  16 Jun 1997
  1855.  
  1856.  
  1857.              This means that in normal use the reference message ID list
  1858.              will be properly maintained by Usenet softwares.
  1859.  
  1860.              -----------------------------------------------
  1861.       A.0 Discussion on generating unique local parts
  1862.              -----------------------------------------------
  1863.  
  1864.              How any given site generates unique local parts is up to it.
  1865.              So this appendix should only be taken as a guideline.
  1866.  
  1867.              On sites where there is only one piece of software assigning
  1868.              message IDs (e.g.  there is only one UA, or the MTA itself
  1869.              assigns message IDs), then a simple "take a ticket" scheme
  1870.              could work.  Multiple instances of that piece of software
  1871.              running simultaneously would need to arbitrate access to that
  1872.              "ticket dispenser" amongst themselves.
  1873.  
  1874.              A discussion of `sequencers' (which is the proper name for
  1875.              this idea) and how atomic operations on them can be
  1876.              implemented, can be found in any good computer science
  1877.              textbook on concurrent systems.
  1878.  
  1879.              Unfortunately, in today's heterogeneous world, it is difficult
  1880.              to the point of impossibility to get every piece of software
  1881.              to agree to use one single central sequencer.
  1882.  
  1883.              It is obvious that using just the date/time for a message ID
  1884.              is insufficient on multitasking systems, or even on single
  1885.              tasking systems that can generate multiple messages per clock
  1886.              tick.
  1887.  
  1888.              What is less obvious is that it is not a good idea to use the
  1889.              name of the software generating the message ID and a sequencer
  1890.              maintained by that software as the unique local part.  The
  1891.              problem here is that it is not guaranteed that different
  1892.              softwares will use different names (especially if they are
  1893.              called "Message Editor" (-:), so it is possible that different
  1894.              softwares could generate duplicate local parts.
  1895.  
  1896.              Some form of "product ID code" would of course rectify this,
  1897.              but given the amount of software in use and under development
  1898.              these days, a centrally administered product ID database
  1899.              hasn't been a viable option for decades now.
  1900.  
  1901.              There are, of course, simpler schemes, that can guarantee to
  1902.              produce unique local parts, because they rely on features that
  1903.              are guaranteed unique to every individual application running,
  1904.              and do not rely on different applications co-operating to use
  1905.              the same central facilities, such as a site-wide sequencer.
  1906.  
  1907.              One commonly used scheme is to use a combination of the
  1908.              current date and time and the process and thread IDs of the
  1909.              software creating the message ID.
  1910.  
  1911.              e.g.  1995Jan31.123426.26.1
  1912.                or  1995013112343600260001
  1913.      FIDONEWS 14-24               Page 34                  16 Jun 1997
  1914.  
  1915.  
  1916.              This doesn't have to be human-readable calendar time, of
  1917.              course.  It could equally well be the POSIX 1003.1 time
  1918.              (seconds since The Epoch), or the Julian date plus the time of
  1919.              day.
  1920.  
  1921.              If the time isn't granular enough, a sequence number (which
  1922.              can be maintained individually by each process) can be added
  1923.              to increase its granularity.
  1924.  
  1925.              On just about every operating system in the world, including
  1926.              multi-user ones, the <time,process,thread,seq> 4-tuple will be
  1927.              unique on one machine *forever* (or until the clock wraps
  1928.              around, at least).
  1929.  
  1930.              e.g.  1995Jan31.123426.26.1.2
  1931.                or  19950131123436002600010003
  1932.  
  1933.              On multiple machine sites, where all machines share the one
  1934.              site identifier, the above scheme can be extended to include
  1935.              the "hidden" local machine name, which will be assumed to be
  1936.              made available (in some fashion) to the softwares generating
  1937.              the message IDs.
  1938.  
  1939.              This yields a unique <machine,time,process,thread,seq> 5-
  1940.              tuple.
  1941.  
  1942.              e.g.  utopium.1995Jan31.123848.26.1.4
  1943.                or  utopium.19950131123907002600010005
  1944.  
  1945.              Again, the "intra-site" machine name can be anything, from the
  1946.              local uname() (for UNIX people) to the NETBIOS machine name
  1947.              (for PC based LAN systems).
  1948.  
  1949.         -------------------------
  1950.       Bibliography and Author
  1951.         -------------------------
  1952.  
  1953.              [FTS0001] A Basic FIDONET Technical Standard, version 15.
  1954.              Randy Bush, Pacific Systems Group.  FIDONET#1:105/6.0.
  1955.              30th August 1990.
  1956.  
  1957.                                ( Defines the type 2.0 packet message
  1958.      transport format.  )
  1959.  
  1960.              [FTS0009] A standard for message identifiers and reply chain
  1961.              linkage, version 1. Jim Nutt.  FIDONET#1:114/30.0.  17th
  1962.              December 1991.
  1963.  
  1964.                                ( Defines the MSGID/REPLY kludges.  )
  1965.  
  1966.              [FSC0034] Gateways to and from FIDONET.  Technical,
  1967.              administrative, and policy considerations.  Randy Bush,
  1968.              Pacific Systems Group.  FIDONET 1:105/6.0.  30th August 1990.
  1969.  
  1970.                                ( Discussion on features that should
  1971.      be preserved across gateways, and on good gateway behaviour in
  1972.      FIDONEWS 14-24               Page 35                  16 Jun 1997
  1973.  
  1974.  
  1975.      general.  )
  1976.  
  1977.              [FSC0039] A type 2 packet extension proposal, version 4. Mark
  1978.              A. Howard.  FIDONET#1:260/340.  29th September 1990.
  1979.  
  1980.                                ( Defines the type 2.0+ packet message
  1981.      transport format. )
  1982.  
  1983.              [FSC0045] A proposal for a new packet format, version 1. Thom
  1984.              Henderson.  FIDONET#1:107/542.1.      17th April 1990.
  1985.  
  1986.                                ( Defines the type 2.2 packet message
  1987.      transport format.  )
  1988.  
  1989.              [FSC0068] A proposed replacement for FTS-0004, version 1. Mark
  1990.              Kimes.  FIDONET#1:380/16.0.  13th December 1992.
  1991.  
  1992.                                ( Defines kludge lines.  )
  1993.  
  1994.              [RFC0822] Standard for the format of ARPA Internet text
  1995.              messages.  David Crocker, University of Delaware.  13th August
  1996.              1982.
  1997.  
  1998.                                ( Defines the grammar and semantics of RFC
  1999.      messages.  )
  2000.  
  2001.              [RFC1036] Standard for the interchange of USENET messages.
  2002.              M Horton, AT&T bell labs; and R. Adams, Centre for seismic
  2003.              studies.  December 1987.
  2004.  
  2005.                                ( Defines changes to the grammar and
  2006.      semantics of RFC822 that are required for news instead of mail,
  2007.      including reply linking.  )
  2008.  
  2009.              [RFC1521] MIME (Multipurpose Internet Mail Extensions) Part
  2010.              One: Mechanisms for specifying and describing the format of
  2011.              Internet message bodies.  N. Borenstien, Bellcore; and N.
  2012.              Freed, Innosoft.  September 1993.
  2013.  
  2014.                                ( Defines Quoted Printable encoding of text.
  2015.      )
  2016.  
  2017.              [RFC1522] MIME (Multipurpose Internet Mail Extensions) Part
  2018.              One: Message header extensions for non ASCII text.  K. Moore,
  2019.              University of Tennesee.  September 1993.
  2020.  
  2021.                                ( Defines how to use Q-P encoding in message
  2022.      headers.  )
  2023.  
  2024.              [TYPE2EXT] An extension to type 2.0, 2.0+, and 2.2 message
  2025.              transport formats to eliminate most kludge lines from the
  2026.              message body.  Jonathan de Boyne Pollard.  FIDONET#2:440/4.0.
  2027.                                 [ Not yet released.  ]
  2028.  
  2029.              -----------
  2030.              Jonathan de Boyne Pollard
  2031.      FIDONEWS 14-24               Page 36                  16 Jun 1997
  2032.  
  2033.  
  2034.              FIDONET#2:440/4.0
  2035.  
  2036.      -----------------------------------------------------------------
  2037.  
  2038.      FIDONEWS 14-24               Page 37                  16 Jun 1997
  2039.  
  2040.  
  2041.      =================================================================
  2042.                             COORDINATORS CORNER
  2043.      =================================================================
  2044.  
  2045.  
  2046.      Nodelist-statistics as seen from Zone-2 for day 164
  2047.      By Ward Dossche, 2:292/854
  2048.         ZC/2
  2049.  
  2050.       +----+------+------------+------------+------------+------------+--+
  2051.       |Zone|Nl-136|Nodelist-143|Nodelist-150|Nodelist-157|Nodelist-164|%%|
  2052.       +----+------+------------+------------+------------+------------+--+
  2053.       |  1 |  8367| 8277   -90 | 8277     0 | 8182   -95 | 8182     0 |31|
  2054.       |  2 | 15879|15855   -24 |15835   -20 |15774   -61 |15703   -71 |60|
  2055.       |  3 |   800|  761   -39 |  765     4 |  758    -7 |  758     0 | 3|
  2056.       |  4 |   543|  543     0 |  543     0 |  519   -24 |  514    -5 | 2|
  2057.       |  5 |    87|   87     0 |   87     0 |   87     0 |   87     0 | 0|
  2058.       |  6 |  1083| 1077    -6 | 1078     1 | 1078     0 | 1078     0 | 4|
  2059.       +----+------+------------+------------+------------+------------+--+
  2060.            | 26759|26600  -159 |26585   -15 |26398  -187 |26322   -76 |
  2061.            +------+------------+------------+------------+------------+
  2062.  
  2063.      -----------------------------------------------------------------
  2064.  
  2065.      FIDONEWS 14-24               Page 38                  16 Jun 1997
  2066.  
  2067.  
  2068.      =================================================================
  2069.                                  NET HUMOR
  2070.      =================================================================
  2071.  
  2072.  
  2073.      From: "Mike Riddle" <mriddle@monarch.papillion.ne.us>
  2074.      To: "Baker, Christopher" <cbaker84@digital.net (Christopher Baker)
  2075.      Date: Fri, 16 May 97 08:12:01 -0600
  2076.      Reply-To: "Mike Riddle" <mriddle@monarch.papillion.ne.us>
  2077.      Subject: Fwd: Java's Year 292 Million Bug
  2078.  
  2079.      ==================BEGIN FORWARDED MESSAGE==================
  2080.      Received: from austin.onu.edu (austin.onu.edu [140.228.10.1]) by
  2081.      monarch.papillion.ne.us (8.7.4/8.6.9) with ESMTP id BAA23416 for
  2082.      mriddle@monarch.papillion.ne.us>; Fri, 16 May 1997 01:19:06 -0500
  2083.      (CDT) Received: from austin.onu.edu (localhost.onu.edu [127.0.0.1])
  2084.             by austin.onu.edu (8.8.5/8.8.5) with SMTP id CAA85786
  2085.             for <mriddle@monarch.papillion.ne.us>; Fri, 16 May 1997
  2086.      02:11:31 -0400
  2087.      Date: Fri, 16 May 1997 02:11:31 -0400
  2088.      Errors-To: david@drw.onu.edu
  2089.      Reply-To: network2d-l@austin.onu.edu
  2090.      Originator: network2d-l@austin.onu.edu
  2091.      Sender: network2d-l@austin.onu.edu
  2092.      From: jenniferrose <jjrose@redoak.heartland.net>
  2093.      To: mriddle@monarch.papillion.ne.us
  2094.      Subject: Java's Year 292 Million Bug
  2095.  
  2096.  
  2097.      YEAR 2000 WOES DON'T AFFECT JAVA UNTIL A.D. 292271023
  2098.  
  2099.      [5/13/97]
  2100.  
  2101.      Sun Microsystems Inc. today acknowledged the Year 292 Million Bug in
  2102.      the Java computer language, which could cause problems for Social
  2103.      Security recipients and millions of other computer-dependent users in
  2104.      292271023 A.D.
  2105.  
  2106.      Dr. James Gosling, the inventor of Java, divulged the problem and
  2107.      hastened to add that a team of specialists is now at work attempting
  2108.      to solve the problem sometime within the next 292,271 millennia.
  2109.  
  2110.      "We can't be certain Java will be around that long," said Gosling,
  2111.      inventor of Java. "But then again, we can't take any chances. Two
  2112.      hundred and ninety two million-plus years may seem like a long time
  2113.      for a species. But relatively speaking, in astronomical terms, it's
  2114.      nothing." Added Gosling, "I don't mean to brag, but Java is taking on
  2115.      a life of its own. We do see it as the computing platform of the 21st
  2116.      century and well beyond."
  2117.  
  2118.      For more information contact Lisa Poulson at lisa.poulson@eng.sun.com
  2119.      or 408-343-1630.
  2120.  
  2121.      http://java.sun.com/pr/1997/may/spotnews/sn970513.html
  2122.  
  2123.      _____________________________________________________
  2124.      FIDONEWS 14-24               Page 39                  16 Jun 1997
  2125.  
  2126.  
  2127.      jennifer j. rose        jjrose@redoak.heartland.net     jrose@giga.com
  2128.      Lawyer, Attorney, Counselor, All-Purpose Bitch
  2129.      P.O. Box 616    Shenandoah, IA 51601-0616
  2130.      voice   712-246-5531   fax  712-246-5533   home  unlisted
  2131.      VISA/Mastercard Accepted
  2132.  
  2133.      ===================END FORWARDED MESSAGE===================
  2134.  
  2135.      -----------------------------------------------------------------
  2136.  
  2137.      FIDONEWS 14-24               Page 40                  16 Jun 1997
  2138.  
  2139.  
  2140.      =================================================================
  2141.                                   NOTICES
  2142.      =================================================================
  2143.  
  2144.  
  2145.      Christopher Baker
  2146.      Rights On! 1:18/14
  2147.  
  2148.                            A_THEIST Echo Available
  2149.  
  2150.  
  2151.      A_theism means free of religion in the way a_political means free of
  2152.      politics or a_sexual means free of sex characteristics or drives.
  2153.  
  2154.      With that in mind and ever cognizant of the continued pressure of
  2155.      religion to intrude itself into our government and its operations, the
  2156.      A_THEIST Echo is provided to inform and alarm and hopefully wake up
  2157.      the sleeping and too long silent majority to the peril on our
  2158.      doorstep.
  2159.  
  2160.      It is a Zone 1 Backbone Echo Hosted and Moderated by Rights On!
  2161.      [1:18/14] and Christopher Baker [card carrying member of American
  2162.      Atheists, Inc.]. Initial links may be obtained from your local
  2163.      Backbone source connection.
  2164.  
  2165.      The Echo is open to anyone who can discuss, without proselytizing, the
  2166.      extreme desirability of maintaining the absolute separation of State
  2167.      and church in this country as provided for in the U.S. Constitution
  2168.      and other Constitutions around the world.
  2169.  
  2170.      A sample of the first few messages and the statement of purpose of the
  2171.      Echo is available as A_THEIST.ZIP from this system anytime except
  2172.      0100-0130 ET and Zone 1 ZMH [USR HST ds online] if you wish to get an
  2173.      idea of whether to commit disk space to the Echo. An archive of the
  2174.      past traffic from the Echo is also available as A_ECHO1.ZIP,
  2175.      A_ECHO2.ZIP, and A_ECHO3.ZIP.
  2176.  
  2177.      Ask your Backbone connection to get it for you! The complete info is
  2178.      available in the current ELISTnnn.XXX file available from your NEC or
  2179.      REC or here. [Request ELIST.]
  2180.  
  2181.      I hope you will join us or ask your Sysop to request a link via their
  2182.      regular Backbone connections!
  2183.  
  2184.      QOFM.
  2185.      Chris
  2186.  
  2187.      -----------------------------------------------------------------
  2188.  
  2189.                                 Future History
  2190.  
  2191.       1 Jul 1997
  2192.         Canada Day - Happy Birthday Canada.
  2193.  
  2194.       9 Jul 1997
  2195.         Independence Day, Argentina.
  2196.      FIDONEWS 14-24               Page 41                  16 Jun 1997
  2197.  
  2198.  
  2199.       1 Aug 1997
  2200.         International FidoNet PENPAL [Echo] meeting in Dijon, France
  2201.  
  2202.      13 Oct 1997
  2203.         Thanksgiving Day, Canada.
  2204.  
  2205.       1 Dec 1997
  2206.         World AIDS Day.
  2207.  
  2208.      10 Dec 1997
  2209.         Nobel Day, Sweden.
  2210.  
  2211.      12 Jan 1998
  2212.         HAL 9000 is one year old today.
  2213.  
  2214.      22 May 1998
  2215.         Expo '98 World Exposition in Lisbon (Portugal) opens.
  2216.  
  2217.       1 Dec 1998
  2218.         Fifteenth Anniversary of release of Fido version 1 by
  2219.         Tom Jennings.
  2220.  
  2221.      31 Dec 1999
  2222.         Hogmanay, Scotland. The New Year that can't be missed.
  2223.  
  2224.       1 Jan 2000
  2225.         The 20th Century, C.E., is still taking place thru 31 Dec.
  2226.  
  2227.      15 Sep 2000
  2228.         Sydney (Australia) Summer Olympiad opens.
  2229.  
  2230.       1 Jan 2001
  2231.         This is the actual start of the new millennium, C.E.
  2232.  
  2233.      -- If YOU have something which you would like to see in this
  2234.         Future History, please send a note to the FidoNews Editor.
  2235.  
  2236.      -----------------------------------------------------------------
  2237.  
  2238.      FIDONEWS 14-24               Page 42                  16 Jun 1997
  2239.  
  2240.  
  2241.      =================================================================
  2242.                             FIDONEWS PUBLIC-KEY
  2243.      =================================================================
  2244.  
  2245.  
  2246.      [this must be copied out to a file starting at column 1 or
  2247.       it won't process under PGP as a valid public-key]
  2248.  
  2249.  
  2250.      -----BEGIN PGP PUBLIC KEY BLOCK-----
  2251.      Version: 2.6.2
  2252.      Comment: Clear-signing is Electronic Digital Authenticity!
  2253.  
  2254.      mQCNAzINVLcAAAEEAM5dZN6t6j5Yc0kl7qegVFfiBeVoteuhDg4ay8h43u38Q4kO
  2255.      eJ9Mm7J89wXFb9vgouBVb4biIN6bTWCwcXTbGhBe5OIceLvluuxuEKsaIs/UwXNe
  2256.      Ogx5azIPhRfC7MJDe41Z8tMEBuHY/NE88cuxQ8yXWO126IRttavu6L/U5BwRAAUR
  2257.      tCRGaWRvTmV3cyBFZGl0b3IgPDE6MS8yM0BmaWRvbmV0Lm9yZz6JAJUDBRAyGwFS
  2258.      JZMgw7eCKz0BAZl0A/9xrfhpsEOqGiPfjy2qd9dv6tvSVPPVFu+Wy1lGTHYtuTtg
  2259.      FIN3fQ47AM3XzqHxWRWvp/xZYgR6sRICL7UFx94ShYBQc7CyqBBZKA0IvIWqXP/g
  2260.      c4Br+gQJR6CLiQK7TUyjUbqNbs6QAxuNUi4xFQM+O2Gene5/iTjHFmmSDj2C9YkB
  2261.      FQMFEDIOmHDTQ6/52IG1SQEBQ78H/Rz/mleIrtZwFIOhzy3JH4Z6FUTfZuM9nPcs
  2262.      1ZLjZCPptHvY7wEYJWGr03lPPJ6tj1VBXwTrWJTf/hOLsoi00GKV8t1thjqGDo23
  2263.      O91/bSQ+Vn0vBQ2vOEJys8ftxdoLJAyI5YLzHVT+RsMTQLIXVuPyrNcKs1vC2ql+
  2264.      UDHpU1R+9cG9JUEHpGI6z0DPnQ74SKbQH3fiVBpHhYx4BmvcBC4gWQzKMkDWFiq3
  2265.      8AssIZ7b9lWl3OBgQ4UM1OIDKoJyjRewIdKyl7zboKSt6Qu8LrcsXO3kb81YshOW
  2266.      ZpSS3QDIqfZC4+EElnB15l4RcVwnPHBaQY0FxUr4Vl4UWM36jbuJAJUDBRAyDpgY
  2267.      q+7ov9TkHBEBAQGoA/sFfN07IFQcir456tJfBfB9R5Z6e6UKmexaFhWOsLHqbCq6
  2268.      3FGXDLeivNn6NTz81QeqLIHglTuM3NP1mu8sw215klAG8G3M1NA2xLw7Eqhspze2
  2269.      raGvNeEwxl8e+PY9aZwBj4UWU+CmIm6QNiP0MtvR7QYDIKn5mZCDc3CLmr942IkB
  2270.      FQMFEDIOh0O8AhTPqRipPQEB4EYH/1gkDmdHL6lbEkFuQLrylF+weBl0XQ+kv7ER
  2271.      vWXYrvIrkppxtc4VAge6CXXEbOGJnvkFHgyNZzO9Q9O64QsmZvjip+4lhDLeNrdH
  2272.      X9DizS4YKXxkSKr9Yltmn2/AlBCx6jwcDIfkqy/P1tNWcikxZZMd6KryK0Wsres9
  2273.      Ik12OmVmJjQSxb5bS6Q8aYUbV3qwosGXTqy+BzYh/UYAX/XJIWa5kxFVSPKFSZ+5
  2274.      toiSzANd9SpHPEogGvQDHJlJ23lmsMx/6uHsR1LTsQ8su8zIk92XyqePJTjlMx2j
  2275.      D7KJWNR7Zzu4QHCXBkga5W8l2FfPk7D3+o7bXTLRuR1yTYGdNoiJAJUCBRAyDhwt
  2276.      SlKLwP4OFW0BAdaMA/9rcWQlSq44K9JuJ7fZUgt9fwxGreTud9fC8DvlbUW79+CA
  2277.      AHLTLLagcEF1OKsWzVBWcA2JEAp+TUTqktRN0oD8vnaw3uNJd1G5KK59hw0WR8x1
  2278.      v4ivypbSjiq95Y3gBunb7WjpyiFRWDlm0PrKrWHtbWzjnpPIpetln1UuqsSfbokB
  2279.      FQIFEDIOG9C3N61ZQ4Dr/QEBIzMH/1VxxztmBPBszbjZLDO8Svcax9Ng8IcWpcDy
  2280.      WqHCAA2Hoe5VtMD0v6w31ZgVqTPIvCark2Y/aTR1GofiuN9NUqbVV534AgAYLzYk
  2281.      DMT1swsPvqDTpOYgQl6PCGh6A5JGAbWJfKkX9XCUHJAAmiTsEVRNnjOgL+p6qjoh
  2282.      EfIG8CGehghWSRKl5eGeDAtbXupZKNjFI1t2XV+ks0RFQ/RPuTH7pF7pk7WO6Cyg
  2283.      +Dk2ZMgua0HRL1fXvHKb5Xzr3MVgsbAl5gP8ooIiD9MI/x5Irh3oo58VyoEZNBs/
  2284.      Kz+drGFDPljcS6fdiVCFtYIzMrshY6YsfLi0aB8fwOvFtxgBqli0J0NocmlzdG9w
  2285.      aGVyIEJha2VyIDwxOjE4LzE0QGZpZG9uZXQub3JnPrQoQ2hyaXN0b3BoZXIgQmFr
  2286.      ZXIgPGNiYWtlcjg0QGRpZ2l0YWwubmV0Pg==
  2287.      =61OQ
  2288.      -----END PGP PUBLIC KEY BLOCK-----
  2289.  
  2290.  
  2291.      File-request FNEWSKEY from 1:1/23 [1:18/14] or download it from the
  2292.      Rights On! BBS at 1-904-409-7040 anytime except 0100-0130 ET and Zone
  2293.      1 ZMH at 1200-9600+ HST/V32B. The FidoNews key is also available on
  2294.      the FidoNews homepage listed in the Masthead information.
  2295.  
  2296.      -----------------------------------------------------------------
  2297.      FIDONEWS 14-24               Page 43                  16 Jun 1997
  2298.  
  2299.  
  2300.      =================================================================
  2301.                             FIDONET BY INTERNET
  2302.      =================================================================
  2303.  
  2304.      This is a list of all FidoNet-related sites reported to the Editor as
  2305.      of this appearance.
  2306.  
  2307.      ============
  2308.  
  2309.      FidoNet:
  2310.  
  2311.        Homepage     http://www.fidonet.org
  2312.        FidoNews     http://ddi.digital.net/~cbaker84/fidonews.html
  2313.        HTML FNews   http://www.geocities.com/Athens/6894/
  2314.        WWW sources  http://www.scms.rgu.ac.uk/students/cs_yr94/lk/fido.html
  2315.        FTSC page    http://www2.blaze.net.au/ftsc.html
  2316.        Echomail     http://www.portal.ca/~awalker/index.html
  2317.        WebRing      http://ddi.digital.net/~cbaker84/fnetring.html
  2318.  
  2319.      ============
  2320.  
  2321.      Zone 1:       http://www.z1.fidonet.org
  2322.  
  2323.        Region 10:  http://www.psnw.com/~net205/region10.html
  2324.  
  2325.        Region 11:  http://oeonline.com/~garyg/region11/
  2326.  
  2327.        Region 13:  http://www.smalltalkband.com/st01000.htm
  2328.  
  2329.        Region 14:  http://www.netins.net/showcase/fidonet/
  2330.  
  2331.        Region 15:  http://www.smrtsys.com/region15/ [disappeared?]
  2332.  
  2333.        Region 16:  http://www.tiac.net/users/satins/region16.htm
  2334.  
  2335.        Region 17:  http://www.portal.ca/~awalker/region17.htm
  2336.            REC17:  http://www.westsound.com/ptmudge/
  2337.  
  2338.        Region 18:  http://www.citicom.com/fido.html
  2339.  
  2340.        Region 19:  http://www.compconn.net
  2341.  
  2342.      ============
  2343.  
  2344.      Zone 2:       http://www.z2.fidonet.org
  2345.  
  2346.      ZEC2:         http://fidoftp.paralex.co.uk/zec.htm [shut down?]
  2347.      Zone 2 Elist: http://www.fidonet.ch/z2_elist/z2_elist.htm
  2348.  
  2349.        Region 20:  http://www.fidonet.pp.se (in Swedish)
  2350.  
  2351.        Region 24:  http://www.swb.de/personal/flop/gatebau.html (in German)
  2352.  
  2353.        Region 25:
  2354.                    http://members.aol.com/Net254/
  2355.  
  2356.      FIDONEWS 14-24               Page 44                  16 Jun 1997
  2357.  
  2358.  
  2359.        Region 27:  http://telematique.org/ft/r27.htm
  2360.  
  2361.        Region 29:  http://www.rtfm.be/fidonet/  (in French)
  2362.  
  2363.        Region 30:  http://www.fidonet.ch  (in Swiss)
  2364.  
  2365.        Region 34:  http://www.pobox.com/cnb/r34.htm  (in Spanish)
  2366.            REC34:  http://pobox.com/~chr
  2367.  
  2368.        Region 36:  http://www.geocities.com/SiliconValley/7207/
  2369.  
  2370.        Region 41:  http://www.fidonet.gr (in Greek and English)
  2371.  
  2372.        Region 48:  http://www.fidonet.org.pl
  2373.  
  2374.      ============
  2375.  
  2376.      Zone 3:       http://www.z3.fidonet.org
  2377.  
  2378.      ============
  2379.  
  2380.      Zone 4:       (not yet listed)
  2381.  
  2382.        Region 90:
  2383.          Net 904:  http://members.tripod.com/~net904 (in Spanish)
  2384.  
  2385.      ============
  2386.  
  2387.      Zone 5:       (not yet listed)
  2388.  
  2389.      ============
  2390.  
  2391.      Zone 6:       http://www.z6.fidonet.org
  2392.  
  2393.      ============
  2394.  
  2395.      -----------------------------------------------------------------
  2396.  
  2397.      FIDONEWS 14-24               Page 45                  16 Jun 1997
  2398.  
  2399.  
  2400.      =================================================================
  2401.                            FIDONEWS INFORMATION
  2402.      =================================================================
  2403.  
  2404.      ------- FIDONEWS MASTHEAD AND CONTACT INFORMATION -------
  2405.  
  2406.      Editor: Christopher Baker
  2407.  
  2408.      Editors Emeritii: Tom Jennings, Thom Henderson, Dale Lovell,
  2409.                        Vince Perriello, Tim Pozar, Sylvia Maxwell,
  2410.                        Donald Tees
  2411.  
  2412.      "FidoNews Editor"
  2413.          FidoNet  1:1/23
  2414.          BBS  1-904-409-7040,  300/1200/2400/14400/V.32bis/HST(ds)
  2415.  
  2416.       more addresses:
  2417.          Christopher Baker -- 1:18/14, cbaker84@digital.net
  2418.                                        cbaker84@aol.com
  2419.                                        cbaker84@msn.com
  2420.  
  2421.      (Postal Service mailing address)
  2422.          FidoNews Editor
  2423.          P.O. Box 471
  2424.          Edgewater, FL 32132-0471
  2425.          U.S.A.
  2426.  
  2427.  
  2428.      voice:  1-904-409-3040 [1400-2100 ET only, please]
  2429.                             [1800-0100 UTC/GMT]
  2430.  
  2431.      ------------------------------------------------------
  2432.  
  2433.      FidoNews is published weekly by and for the members of the FIDONET
  2434.      INTERNATIONAL AMATEUR ELECTRONIC MAIL system.  It is a compilation
  2435.      of individual articles contributed by their authors or their
  2436.      authorized agents.  The contribution of articles to this compilation
  2437.      does not diminish the rights of the authors.  OPINIONS EXPRESSED in
  2438.      these articles ARE THOSE OF THE AUTHORS and not necessarily those of
  2439.      FidoNews.
  2440.  
  2441.      Authors retain copyright on individual works; otherwise FidoNews is
  2442.      Copyright 1997 Christopher Baker.  All rights reserved.  Duplication
  2443.      and/or distribution permitted for noncommercial purposes only.  For
  2444.      use in other circumstances, please contact the original authors, or
  2445.      the Editor.
  2446.  
  2447.                             =*=*=*=*=*=*=*=*=
  2448.  
  2449.      OBTAINING COPIES: The most recent issue of FidoNews in electronic
  2450.      form may be obtained from the FidoNews Editor via manual download or
  2451.      file-request, or from various sites in the FidoNet and Internet.
  2452.      PRINTED COPIES may be obtained by sending SASE to the above postal
  2453.      address.  File-request FIDONEWS for the current Issue.  File-request
  2454.      FNEWS for the current month in one archive.  Or file-request specific
  2455.      back Issue filenames in distribution format [FNEWSEnn.ZIP] for a
  2456.      FIDONEWS 14-24               Page 46                  16 Jun 1997
  2457.  
  2458.  
  2459.      particular Issue.  Monthly Volumes are available as FNWSmmmy.ZIP
  2460.      where mmm = three letter month [JAN - DEC] and y = last digit of the
  2461.      current year [7], i.e., FNWSFEB7.ZIP for all the Issues from Feb 97.
  2462.  
  2463.      Annual volumes are available as FNEWSn.ZIP where n = the Volume number
  2464.      1 - 14 for 1984 - 1997, respectively. Annual Volume archives range in
  2465.      size from 48K to 1.4M.
  2466.  
  2467.  
  2468.      INTERNET USERS: FidoNews is available via:
  2469.  
  2470.                           http://www.fidonet.org/fidonews.htm
  2471.                           ftp://ftp.fidonet.org/pub/fidonet/fidonews/
  2472.                           ftp://ftp.aminet.org/pub/aminet/comm/fido/
  2473.  
  2474.                                       *=*=*
  2475.  
  2476.      You may obtain an email subscription to FidoNews by sending email to:
  2477.  
  2478.                           jbarchuk@worldnet.att.net
  2479.  
  2480.      with a Subject line of: subscribe fnews-edist
  2481.  
  2482.      and no message in the message body. To remove your name from the email
  2483.      distribution use a Subject line of: unsubscribe fnews-edist with no
  2484.      message to the same address above.
  2485.  
  2486.                                       *=*=*
  2487.  
  2488.      You can read the current FidoNews Issue in HTML format at:
  2489.  
  2490.                           http://www.geocities.com/Athens/6894/
  2491.  
  2492.      STAR SOURCE for ALL Past Issues via FTP and file-request -
  2493.      Available for FReq from 1:396/1 or by anonymous FTP from:
  2494.  
  2495.                           ftp://ftp.sstar.com/fidonet/fnews/
  2496.  
  2497.      Each yearly archive also contains a listing of the Table-of-Contents
  2498.      for that year's issues.  The total set is currently about 11 Megs.
  2499.  
  2500.                                  =*=*=*=
  2501.  
  2502.      The current week's FidoNews and the FidoNews public-key are now also
  2503.      available almost immediately after publication on the Editor's new
  2504.      homepage on the World Wide Web at:
  2505.  
  2506.                   http://ddi.digital.net/~cbaker84/fidonews.html
  2507.  
  2508.      There are also links there to jim barchuk's HTML FidoNews source and
  2509.      to John Souvestre's FTP site for the archives. There is also an email
  2510.      link for sending in an article as message text. Drop on over.
  2511.  
  2512.                             =*=*=*=*=*=*=*=*=
  2513.  
  2514.      A PGP generated public-key is available for the FidoNews Editor from
  2515.      FIDONEWS 14-24               Page 47                  16 Jun 1997
  2516.  
  2517.  
  2518.      1:1/23 [1:18/14] by file-request for FNEWSKEY or by download from
  2519.      Rights On! BBS at 1-904-409-7040 as FIDONEWS.ASC in File Area 18.  It
  2520.      is also posted twice a month into the PKEY_DROP Echo available on the
  2521.      Zone 1 Echomail Backbone.
  2522.  
  2523.                                 *=*=*=*=*
  2524.  
  2525.      SUBMISSIONS: You are encouraged to submit articles for publication in
  2526.      FidoNews. Article submission requirements are contained in the file
  2527.      ARTSPEC.DOC, available from the FidoNews Editor, or file-requestable
  2528.      from 1:1/23 [1:18/14] as file "ARTSPEC.DOC".  ALL Zone Coordinators
  2529.      also have copies of ARTSPEC.DOC. Please read it.
  2530.  
  2531.      "Fido", "FidoNet" and the dog-with-diskette are U.S. registered
  2532.      trademarks of Tom Jennings, P.O. Box 410923, San Francisco, CA 94141,
  2533.      and are used with permission.
  2534.  
  2535.              "Disagreement is actually necessary,
  2536.               or we'd all have to get in fights
  2537.               or something to amuse ourselves
  2538.               and create the requisite chaos."
  2539.                                 -Tom Jennings
  2540.  
  2541.       -30-
  2542.  
  2543.      -----------------------------------------------------------------
  2544.  
  2545.