Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!wuarchive!emory!utkcs2!ornl.gov!de5
From: de5@ornl.gov (Dave Sill)
Newsgroups: comp.benchmarks
Subject: Re: c.b archives
Message-ID: <1991Apr11.192307.24064@cs.utk.edu>
Date: 11 Apr 91 19:23:07 GMT
References: <9042@mcshh.hanse.de> <1991Apr10.200334.19846@nas.nasa.gov> <1991Apr11.120601.12106@cs.utk.edu> <1991Apr11.164050.16935@nas.nasa.gov>
Sender: usenet@cs.utk.edu (USENET News Poster)
Reply-To: Dave Sill <de5@ornl.gov>
Organization: Oak Ridge National Laboratory
Lines: 21

In article <1991Apr11.164050.16935@nas.nasa.gov>, eugene@nas.nasa.gov (Eugene N. Miya) writes:
>
>No, the problem I have is what goes into an archive.  I think simple
>queries (Where can I get <obvious benchmark>) should be left out.
>"Me, too." posts as well.  No, the issue is time and quality of what
>goes in.  And I think a little bit of consensus.  I'm not for archiving
>all data.  I'd like to see some turn over so I can see progress
>[like people saying "I promise to summarize" and never do].

I was thinking of a comp.benchmarks archive, i.e., archiving messages
posted to comp.benchmarks.  Sure, it'd be nice to have a hierarchical
archive of source code, results, etc.  But that's a *lot* of work (I
know, I've done it), whereas a simple c.b archive could be trivially
automated.  If the fancy archive is too much work, and the trivial
archive isn't good enough, then we won't have *any* archive.  I'd much
rather have a trivial archive than no archive at all.

-- 
Dave Sill (de5@ornl.gov)	  It will be a great day when our schools have
Martin Marietta Energy Systems    all the money they need and the Air Force
Workstation Support               has to hold a bake sale to buy a new bomber.