Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!ncar!husc6!yale!mfci!rodman
From: rodman@mfci.UUCP (Paul Rodman)
Newsgroups: comp.arch
Subject: Re: Criteria ... [really: should comp.arch be split]
Summary: Porsches, not VWs
Message-ID: <839@m3.mfci.UUCP>
Date: 11 May 89 14:36:28 GMT
References: <2368@ogccse.ogc.edu> <1464@cfa.cfa.harvard.EDU> <141@dg.dg.com> <156@dg.dg.com> <658@pitstop.West.Sun.COM> <19088@winchester.mips.COM> <18154@cup.portal.com> <19182@winchester.mips.COM>
Sender: rodman@mfci.UUCP
Reply-To: rodman@mfci.UUCP (Paul Rodman)
Organization: Multiflow Computer Inc., Branford Ct. 06405
Lines: 26

In article <19182@winchester.mips.COM> mash@mips.COM (John Mashey) writes:
>
>Maybe it's time for another newsgroup.  On looking back on the last few
>years, I observe that there are at least 3 distinct kinds of discussion:
>	a) Pure architectural questions, opinions, information
>	b) Performance analyses
>	c) Economic and/or industry-structure discussions; competitive
>	arguments over non-architecture issues.

Personally, I wouldn't mind a comp.arch.highend vs lowend. I 
get bored with the small systems slant of this bboard. 
[Not that they aren't important!] 

Also I wouldn't mind a comp.implementation where the discussion lends
more toward design rules/tools/pfs and ps....rather than "endianness"
or other such stuff.

Considering that the lifetime of a computer is approximatly the same as the
time it takes to design it, it seems obvious that good tools => more
performance. (another facet to RISC vs CISC, too.)

<Sigh.> perhaps the real problem is that all the interesting topics are
also too important to blabber about on the net....:-) Oh well.

  Paul K. Rodman
  rodman@Multiflow.com