ftp.cs.arizona.edu/ftp.cs.arizona.edu.tar/ftp.cs.arizona.edu/tsql/doc/tsql.mail/000108_rts _Wed May 5 09:28:38 1993.msg

Internet Message Format | 1996-01-31 | 2KB

Received: from boojum.CS.Arizona.EDU by optima.CS.Arizona.EDU (5.65c/15) via SMTP
    id AA12899; Wed, 5 May 1993 09:28:39 MST
Date: Wed, 5 May 1993 09:28:38 MST
From: "Rick Snodgrass" <rts>
Message-Id: <[e-mail]>
Received: by boojum.cs.arizona.edu; Wed, 5 May 1993 09:28:38 MST
To: [e-mail]
Subject: alternative taxonomy

Nandlal,
    Thank you for your clear answers to my questions. Thank
you also in advance for your efforts to compose an alternative
classification scheme.
    If your alternative classification scheme were finished
in the next 10 days, the research community would then need to discuss it,
which could take as long as a few weeks (Christian's taxonomy was
first proposed a month ago, and we are still discussing it!) That would
leave essentially no time for the most important task of the benchmark,
the queries themselves.
    I propose a two-pronged approach. The first prong would continue
using Christian's taxonomy, and would start immediately on the final
task of writing the queries, with the goal of completing an initial
draft by the workshop, June 15 (which is now less than 6 weeks away).
    The second prong would be you writing up an alternative
classification scheme, and making it available, preferably as a LaTeX file,
sometime before the workshop. I'll print out copies for all of the
workshop participants. Then after the workshop, when we start on the
next draft of the benchmark (recall that there are a host of important
issues that were delayed due to lack of time), we can take up the
question of whether the initial taxonomy should be replaced with the
alternative one, or whether some kind of integration would achieve the
strengths of both.