Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!watmath!clyde!cbatt!cbosgd!ihnp4!houxm!hou2b!tpl
From: tpl@hou2b.UUCP (T.LEE)
Newsgroups: net.database
Subject: Re: Benchmark Claims by Various DBMS Vendors on USENET
Message-ID: <817@hou2b.UUCP>
Date: Fri, 8-Aug-86 09:17:02 EDT
Article-I.D.: hou2b.817
Posted: Fri Aug  8 09:17:02 1986
Date-Received: Sat, 9-Aug-86 11:21:00 EDT
References: <358@rtech.UUCP>
Organization: AT&T Bell Labs, Holmdel NJ
Lines: 18

I don't want commercial messages for sure; particularly, the kind
quoted by sales and marketing people.  They should avoid posting
anything simply to aviod conflict of interest.

But, USENET is a place to exchange real benchmarking experience and that
is one reason I read this group.  However, those who post their results
better design their experiments as rigorous/"scientific" as possible,
give all related context, and give a PROPERLY-QUALIFIED comparison.
Knowing benchmarking and comparing simply the processor itself is contraversial
as hell, dealing with DBMS is really a tough job.  It takes decent effort
to separate all non-DB factors if that can be done cleanly.

It IS us READERS's responisbility to ignore most of the junky quotes
that might be posted.
-- 
	ATT BL Holmdel
	....!hou2b!tpl
	T. Paul Lee