Path: utzoo!utgpu!news-server.csri.toronto.edu!rutgers!ucsd!pacbell.com!decwrl!ucbvax!STC06.CTD.ORNL.GOV!de5
From: de5@STC06.CTD.ORNL.GOV (SILL D E)
Newsgroups: comp.arch
Subject: Call for Discussion: comp.benchmarks
Message-ID: <9008061617.AA08817@stc06.CTD.ORNL.GOV>
Date: 6 Aug 90 16:17:01 GMT
Sender: daemon@ucbvax.BERKELEY.EDU
Reply-To: Dave Sill <de5@ornl.gov>
Followup-To: news.groups
Organization: Oak Ridge National Lab
Lines: 34

[I didn't include comp.arch in the original posting of this article.
I've seen been told that benchmarks are frequently discussed here.]

"There are lies, damned lies, and benchmarks," to paraphrase someone
whose name escapes me at the moment [Mark Twain, as several folks have
reminded me].  Like it or not, though, benchmarks are a fact of life
in the computing world.  So in the spirit of `knowing thy enemy', I'd
like to propose the formation of an unmoderated newsgroup for:

    -  Discussions about benchmarks including:
        . hardware benchmarks
        . numeric benchmarks
        . operating system benchmarks
        . compiler benchmarks
        . graphics benchmarks
        . window system benchmarks
        . multiple interactive user load simulations
        . validity of benchmarks/benchmarking
        . benchmarking as a procurement tool
        . benchmarking anecdotes

    -  Publishing benchmark results
        . especially user-run rather than vendor-run

    -  Publishing benchmark source code and availability info
        . including bug fixes/reports

If it's warranted, I'll issue a Call for Votes at least 14 days after
this posting.  [That would be around Aug. 18]

--
Dave Sill (de5@ornl.gov)
Martin Marietta Energy Systems
Workstation Support