Xref: utzoo comp.sys.ibm.pc:13199 comp.binaries.ibm.pc:569
Path: utzoo!mnetor!uunet!husc6!mit-eddie!ll-xn!ames!pasteur!cory.Berkeley.EDU!iverson
From: iverson@cory.Berkeley.EDU (Tim Iverson)
Newsgroups: comp.sys.ibm.pc,comp.binaries.ibm.pc
Subject: Re: Standard format for ARCed files
Message-ID: <1551@pasteur.Berkeley.Edu>
Date: 15 Mar 88 05:18:52 GMT
References: <5439@swan.ulowell.edu>
Sender: news@pasteur.Berkeley.Edu
Reply-To: iverson@cory.Berkeley.EDU.UUCP (Tim Iverson)
Distribution: na
Organization: University of California, Berkeley
Lines: 43

In article <5439@swan.ulowell.edu> boneill@hawk.ulowell.edu (SoftXc Coordinator) writes:
>I propose a standard format for ARCed files uploaded to comp.binaries.ibm.pc
>could be as follows:
>
>	1) The moderator would ARC the files using PKARC 3.5
>	2) The moderator would use the MAKESFX program to create a
>           self-extracting archive.
>
>This would eliminate the need for anyone to have a specific archive
>extractor, hopefully simplifying any problems.

This strikes me as one of the least intelligent ideas I have seen in a long
while.  First, I never run a program on my computer that someone I do not
know has compiled - if you don't haul the horse through your gate, you can't
be attacked.  Now you tell me that I have to *execute* a program that some
fool has posted to the net?  No way.  Moderation is no help as that can be
simply and easily circumvented.

Second, ARC is not the most widespread of archive programs.  PKARC works
only on PC's running MS-DOS, yet you wish to use it as a standard for
transmission between machines that for the most part are not PC's and are
running UNIX.  I would then have to download the file (which could be quite
large) before I could even get a listing of it.  ARC would be slightly
better than PKARC, as some could list and extract the contents before
downloading.  However, there does not exist an ARC program that will
function satifactorily on all or even most of the systems on the net.

There is only one program that I know of that can serve as a standard, if
that is something you feel is needed:  zoo.  It runs on just about any
machine I can think of and the resulting archive can be manipulated with
equal ease on any system zoo runs on.  Rahul Dhesi (sp?) has created a very
nice program that is eminently suitable as a standard for exchange of data
between disimilar machines.

If you do choose a transmission method as a standard, please do not choose
one that creates a vast security hole in the PC user community, nor one that
can only be used from the PC, as that is not the environment from which most
people are connected to the net.


- Tim Iverson
  iverson@cory.Berkeley.EDU
  ucbvax!cory!iverson