Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!uwm.edu!bionet!agate!e260-1f.berkeley.edu!c60b-1eq
From: c60b-1eq@e260-1f.berkeley.edu (Noam Mendelson)
Newsgroups: comp.compression
Subject: Re: ARJ 2.00 file archiver
Message-ID: <1991Apr20.205915.6052@agate.berkeley.edu>
Date: 20 Apr 91 20:59:15 GMT
References: <1991Apr20.052556.3247@world.std.com>
Sender: root@agate.berkeley.edu (Charlie Root)
Organization: University of California, Berkeley
Lines: 23

In article <1991Apr20.052556.3247@world.std.com> robjung@world.std.com (Robert K Jung) writes:
>ARJ has been designed with a general purpose compression algorithm.
>Since PKZIP's compressor is pre-computed and tailored to IBM PC
>data, it will be very difficult to build a general compressor to
>match PKZIP's speed on compression and extraction.

ARJ 2.00 comes very close, although it lags behind in decompression.

>The main goals of ARJ are to provide as much practical compression as
>is possible on an IBM PC machine and as many features as practical.

How about UN*X compatibility?  There are ZOO, ARC, LHARC, and ZIP (although
not by Phil Katz, of course) programs available for UN*X systems, however
I haven't seen any source code for ARJ.

The question is: is ARJ designed for portability?
If not, then it should be optimized to deal with PC-specific data;
otherwise, it should be ported to other OS's.
-- 
+==========================================================================+
| Noam Mendelson   ..!ucbvax!web!c60b-1eq       | "I haven't lost my mind, |
| c60b-1eq@web.Berkeley.EDU                     |  it's backed up on tape  |
| University of California at Berkeley          |  somewhere."             |