home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
ftp.cs.arizona.edu
/
ftp.cs.arizona.edu.tar
/
ftp.cs.arizona.edu
/
tsql
/
doc
/
tsql.mail
/
000043_rts _Thu Mar 18 11:27:59 1993.msg
< prev
next >
Wrap
Internet Message Format
|
1996-01-31
|
2KB
Received: from boojum.cs.arizona.edu by optima.cs.arizona.edu (5.65c/15) via SMTP
id AA06345; Thu, 18 Mar 1993 11:28:00 MST
Date: Thu, 18 Mar 1993 11:27:59 MST
From: "Rick Snodgrass" <rts>
Message-Id: <199303181827.AA14772@boojum.cs.arizona.edu>
Received: by boojum.cs.arizona.edu; Thu, 18 Mar 1993 11:27:59 MST
To: tsql@cs.arizona.edu
Subject: Re; Benchmark correspondence
Patrick and Edward make several excellent, well-articulated suggestions
concerning the positioning and schema of the benchmark.
1) Relax Restrictions
(a) allow each relation to be used more than once
(b) allow aggregation
2) The schema
2.1) Add "natural" as a criterion
(a) add gender as a time-invariant attribute
(b) add birthdate as a time-invariant, user-defined time attribute
(c) add a multi-valued attribute such as skills
It seems that the first version of the benchmark should be kept as
simple as possible. On the other hand, Patrick and Edward make cogent
arguments for including these aspects.
I propose that we add "natural" as a criterion to Version 1 of the
TSQL Benchmark, and agree now to add multiple relations, aggregation,
and gender, birthdate, and something like skills as attributes in
Version 2 of the benchmark. (Note: We need to be careful not to let
Version 2 expand too much.)
As soon as Version 1 stabilizes, let's begin work on Version 2. At
that point, it would be helpful if Patrick and Edward could make three
successive proposals, the first being the schema for version 2, the
second being the data base instance for version 2, and the third being
an extension of the taxonomy.
In the meantime, we should work on getting version 1 completed, so
that we can move on to the more relevant issues raised in the version
2 benchmark.
[As an aside, in this modern world of ours, gender isn't always
time-invariant!]