Path: utzoo!attcan!uunet!husc6!bloom-beacon!tut.cis.ohio-state.edu!mailrus!uwmcsd1!marque!gryphon!greg
From: greg@gryphon.CTS.COM (Greg Laskin)
Newsgroups: news.admin
Subject: Re: A counter-example for those who would eliminate PC binaries
Message-ID: <4683@gryphon.CTS.COM>
Date: 30 Jun 88 09:05:44 GMT
References: <264@octopus.UUCP>
Reply-To: greg@gryphon.CTS.COM (Greg Laskin)
Organization: Trailing Edge Technology, Redondo Beach, CA
Lines: 82

In article <264@octopus.UUCP> pete@octopus.UUCP (Pete Holzmann) writes:

>Consider: PC binaries are useful to many on the net. Probably a majority
>of net-users, if the readership stats are at all accurate.
>
>Consider: The source code for the Mahjongg tiles.
>
OK. Considered.  It's source code.  End of consideration.


>	1) This is pretty useless if you don't have a Sun [sure, it could
>	2) This posting is HUGE
>	3) This posting would be much more efficient if done as a binary
>		with accompanying format information. Since it isn't even
>		executable, there's no fear of viral infection!

Source code seems to survive transmission across the net better than
binary files.  One misplaced or inserted space kills a binary transmission.
You might get a error in a source file but they can frequently be easily
repaired.

>	4) This posting is occuring in a moderated group.

I don't believe the games group has ever produces 8 Megs of traffic in
a single month like comp.binaries.ibm.pc did recently.

>
>Here we had tremendous flames over the GIF files that got posted to the
>net, and now we have a single posting that far exceeds all the GIF files
>ever posted. And not a peep. 

It's source code.

>PC programmers are third-class citizens. 

binaries have little to do with programmers.  Programmers use source code.
End users use binaries.  The PROGRAMMERS that created this network are
chafing over the fact that it is becoming (has become) a common carrier
conduit for end users.  Another reason for the complaints is that the
binaries are frequently not the original work of the poster.  There's
a palpable difference between posting a binary for a program you wrote
and posting a binary that was picked up off a BBS somewhere.  The second
form is using the net as a common carrier distribution medium, the first form
is an exchange of technical information.

>I'm not complaining about the posting, mind you. But I think that those
>who have been vocal about PC's, and about binaries, ought to put their
>feet in a different pair of sandals and rethink the issues. Or at least
>the consistency of their statements.

I have PC/AT's.  Some of them run DOS.  Some run Unix.  gryphon can't handle
8 Megs a month of pc binaries so we expired them very, very fast (while
continuing to feed them downstream).  The problem here is that we don't seem
to be able to differentiate between reasonable and unreasonable volumes.

>Personally, I think that quality, useful postings of all kinds  should have
>a place on the net. If the existing distribution scheme can't handle the
>volume as the net grows, let's solve that problem. But not by disenfranchising
>
>But, if we're going to take the easy way out and simply ban X, we've got to
>be consistent and ban ALL 'X', even the brand that affects us personally.

The chances of the net either becoming fully moderated or being technologically
able to ban the posting of binaries within your lifetime is remote.
This having been said, you also need to examine the sitituation from the
perspective of those of us who have to move and store the traffic.  When
bandwidth becomes limited, the easiest fix is to kill the stuff with the
largest volume.  Every administrator on the net makes individual decisions
about this (and many are not vocal about their decisions).

>If rec.* goes, so do the sci-fi groups, Chuq. If programs for PC's go, so
>should those for Suns. If the net must be turned into a technical-information-
>only network, so be it. I don't think that's necessary, or desireable. But
>whatever we do, let's be consistent!
>
I suspect that if all of the major sites on the net agreed to drop rec.all,
an extremely remote possibility, the distribution would find alternate channels
through which it would move.  If it didn't, there would be good indication
that there was no interest in it in the first place.

-- 
Greg Laskin  greg@gryphon.CTS.COM    !gryphon!greg