W8SDZ@SIMTEL20.ARMY.MIL (Keith Petersen) (09/05/88)
The time has come to stand up and say to SEA - ENOUGH!!! They are aledgedly
contacting numerous shareware authors who have written programs that read
and/or write ARC files, demanding that the authors turn over their source
code and pay 6 percent of their gross sales of such products to SEA.
The whole thing centers around SEA's claim to a valid trademark on the word
"ARC". Below you will find evidence that "ARC" was in COMMON USAGE prior
to the date that SEA filed their trademark.
------------------------------------------------
Exhibit #1 - ARCHIV11.HLP (excerpts - notice the copyright date)
Notice
Changes in the system
The problem
What is an archive?
File lists
ARCHBLD -- build an archive
ARCHGET -- retrieve files from an archive
ARCHDIR -- display archive directory
ARCHTYPE -- display or print archive members
Field modification
:Notice
Copyright 1982 Michael M Rubenstein
This documentation and the accompanying programs may
be freely distributed for noncommercial purposes
provided that the copyright notice is retained.
The archive programs were written in pl/i and com-
piled using the Digital Research pl/i-80 compiler.
This help file is for the following versions
archbld 1.1
archdir 1.1
archget 1.1
archtype 1.0
A large number of people have taken an interest in
these programs and offered suggestions. I am grate-
ful for all such assistance.
The archive system is a collection of three programs
which take a large number of small files and combine
them into a larger file. In actual practice, an-
other small file is built to keep track of the large
data file. For example, one of my disks contains 27
files using 84k. 21 of these files are less than 2k
of data. Most of these actually contain less than
1k of data. The largest file is 10k (that's quite a
large source file for me.) Even this file is wast-
ing about 1.5k.
I ran this disk through the archbld program. That
reduced the space to 48k for the data and 2k for the
directory. That's a lot better.
We can do better still. I now ran the data file
through Richard Greenlaw's excellant squeeze pro-
gram. That reduced it to 30k of data (and still 2k
of directory), about 38% of it's original size.
Michael M Rubenstein
:What is an archive?
An archive consists of two files, a data file and a
directory file.
The data file contains all the members of the
archive.
The directory file contains information on what
files are in the archive and where they are.
-------------------------------------------------------
Exhibit #2 - A message from the SysOp of a local Remote CP/M system.
Notice the date.
Date: 06/10/82 13:10 EST
From: Ron Fowler (sysop, RCPM Westland, Mich)
To: All RCPM Sysops
Sub: Use of ARCH programs
As promised, I've uploaded Michael Rubensteins ARCHIVE programs
to SYSOP CBBS. There are four files (not including this message,
which is not part of the package):
ARCHBLD.COM - builds new archive files
ARCHGET.COM - retrieves files from archive
ARCHDIR.COM - lists directory of an archive
ARCHIVE.HLP - help file for ARCH system (to be
used with Rick Conns HELP pgm)
I believe this set of programs will prove extremely useful in RCPM
applications, especially considering the distribution problems we've
had with public-domain packages made up of many files (YAM, SQ/USQ,
and others). There are a couple of things to consider, however:
1) What should we consider as a maximum size for an archive file? I
believe most users have in excess of 250K of disk storage per
drive (with the notable exception of Osborne users); since archives
can be de-archived to a different drive, I would think 200K would
be a reasonable maximum (perhaps the Osborne segment would prefer
80 or 90K? -- do we have *that* many Osborne users?)
2) The ARCHIVE files work quite well in conjunction with Greenlaw's
SQ/USQ file compression system. This brings up an interesting ques-
tion: Should we squeeze a fileset BEFORE archiving, or after? There
are advantages and disadvantages to each:
a. Squeezing *after* archiving hastens the compression/decompression
process: the relatively slow SQ and USQ programs have to operate on
only one file. This also implies a lot less keyboard work.
b. Squeezing *before* archiving yields a better compression rate, especially
when binary and ascii files are intermixed. This is because the SQ pro-
gram analyzes the file(s) before actually compressing, and selects an
optimal encoding tree based on code distribution frequency.
c. It is possible that a user with relatively small-capacity disk storage
may not have enough room to unsqueeze a large .AQC (squeezed .ARC) file.
My own preference is to squeeze AFTER archiving, but I have plenty of disk
storage, and prefer the simplicity of dealing with only one file.
*****
I picked up these programs from CIS (Compuserve Information Service...nee
Micronet). They are written in PL/I, and the author tells me that source is
available from the Laurel, MD RCPM system.
Some improvements I'd like to see (any PL/I hackers among us?):
1) 'Twould be nice to have the .DIR and .ARC files combined into one file...
should not be too difficult to write a utility to do this, although that
would add another level of complexity to the package.
2) I'd like to see an ARCHTYP program to type out individual files with the
archive file.
3) ARCHADD and ARCHDEL to add and delete files from the archive; presently
it seems that the only way to add a file to an archive is to de-archive
the entire set, then build a whole new archive.
4) Integration of all these function into one program, preferably written
in assembly language.
******
That's all I have for now...comments about the package should be left on SYSOP,
and I'll forward them to the author via CIS. I'd like to see some discussion
regarding RCPM use (especially as concerns SQ/USQ) here on SYSOP.
Enjoy!
--Ron Fowler
6/10/82
-------------------------------------------------------
Exhibit #3 - Proof of common usage by publication:
SIG/M, the (ACNJ) CP/M user group which distributes world-wide, released
disk #112 with ARC files on it. Although not the archive program itself,
this does establish a date of world-wide distribution of ARC files.
----------
[-CATALOG.112 - The catalog of the diskette]
%
Sig/M volume 112 SIGNON/RBBS programs
- Toronto RCP/M System
-CATALOG.111 contents of Sig/M volume 111
released March 15, 1983
%
index name size crc description
112.01 RBBENT27.BQS 15K BA 04
112.02 RBBEXIT1.BQS 4K 50 7B
112.03 RBBMIN27.BQS 19K 8B FB
112.04 RBBMIN27.DQC 10K 4A E9
112.05 RBBS-I/O.NQW 3K 46 CE
112.06 RBBS-PC .BQS 18K 1B B8
112.07 RBBS31 .AQC 19K C5 32
112.08 RBBS31 .DQC 3K 09 01
112.09 RBBSTIME.FQX 2K 1F A8
112.10 RBSCLOCK.BAS 3K B7 6D
112.11 RBSPURGE.BQS 3K D6 D0
112.12 RBSPURGE.DQC 2K 24 4B
112.13 RBSUTL31.AQC 8K FB 12
112.14 SIGNON .SEE 2K BF 98 Explains disk.
112.15 SIGNON-1.ARC 64K C0 B2
112.16 SIGNON-1.DIR 1K 9F F6
112.17 SIGNON-2.ARC 62K 3B 28
112.18 SIGNON-2.DIR 1K A2 3A
Copyright (c) 1983 by Sig/M-Amateur Computer Group
of New Jersey Inc., Box 97, Iselin NJ, 08830-0097
-------------------------------------------------------
Exhibit #4 - More proof of common usage:
[SIGNON.SEE - A "readme" file for the SIGNON program distribution,
clearly stating that the archive programs are needed to extract the
files and view the directory.]
What to take:
You'll need SIGNON-1.ARC, SIGNON-1.DIR, SIGNON-2.ARC and
SIGNON-2.DIR, plus ARCGET11.OBJ (and probably ARCDIR11.OBJ
as well). To see what's in these files, type "ARCDIR
SIGNON-1", etc., which will give you a directory of the
contents. To break the packages apart, the proper command
is "ARCGET *.* SIGNON-1", etc. The result will be a large
number of squeezed files which you'll then have to unsqueeze.
Read the SIGNON.DOC file -- that'll explain how the program
is expected to work and the order in which you should go about
testing and assembling its components. The result will be a
reasonably fancy user logging program which can be run as
a front end to MINIRBBS (or maybe, with a bit of work, MINI-
CBBS), especially suitable when those programs are used in an
RCPM system. SIGNON's main attraction is for sysops rather
than users -- it maintains user statistics and control over
user status much more effectively than any other program thus
far placed in the public domain.
Because initially most sysops using SIGNON will be current users
of ENTERBBS/RBBENTXX, a utility for converting ENTERBBS's 'USERS'
file to a format readable by SIGNON will be available within a
few days. That way, you won't have to ask your regular users to
begin all over again.
SIGNON is being distributed in two packages to make it possible
for owners of slow modems to take one package at a time from
systems that limit connect time to one hour/day. SIGNON-1
contains the stuff you'll need to get started. There should
be enough in it to keep most sysops busy for a day or two.
The program is slated for inclusion in a CPMUG disk, but that
will undoubtedly take a few months to come out. Any sysop
who's not in a hurry (if any such exist) might consider waiting
until then.
-- 7/25/82
--------------------------------------------------
Additional exhibits in the form of working executables which produce ARC
programs for CP/M are available. In addition, I am prepared to sign a
deposition to the effect that my Remote CP/M system did use, and make
available for download, these programs on or about the dates shown above.
--Keith Petersen, W8SDZ
SysOp, RCP/M Royal Oak - 313-759-7569 (300, 1200, 2400, 9600 bps HST)
August 4, 1988