Date:    Wed, 07 Jul 93 13:01:30 -0700
From:    Rick Snodgrass <rts@cs.arizona.edu>
To:      ahn@cbnvma.att.com, ariav@taunivm.tau.ac.il, csj@iesd.auc.dk,
         elmasri@cse.uta.edu, fabio@deis64.cineca.it,
         kaefer@informatik.uni-kl.de, rts@cs.arizona.edu
Subject: Welcome!

Welcome to the TSQL2 language design working group!!!!

First of all, thank you for agreeing to serve on the TSQL2 working
group. I appreciate your willingness to devote a portion of your
summer (and fall!) to this effort, which I believe will benefit the
entire temporal database community.

In this message, I propose some procedures, and elicit your reactions.
A subsequent message will get the technical discussion going.

This working group currently has 7 members.

Ilsoo Ahn
ahn@cbnvma.att.com

Gadi Ariav
ariav@taunivm.tau.ac.il

Christian S. Jensen
csj@iesd.auc.dk

Ramez Elmasri
elmasri@cse.uta.edu

Fabio Grandi
fabio@deis64.cineca.it

Wolfgang Kaefer
kaefer@informatik.uni-kl.de

Richard Snodgrass
rts@cs.arizona.edu

I imagine that the group will expand somewhat over the next week (the
mail to the tsql mailing list just went out yesterday). If you know of
others who agree with the charter, you may want to encourage them to
volunteer to participate.

Interactions on our design should be via email to the entire group.
The tsql mailing list should not be used for our discussions.  I
encourage you to set up a mail alias, say in your .mailrc file. Mine
is

alias wgtsql2 ahn@cbnvma.att.com ariav@taunivm.tau.ac.il csj@iesd.auc.dk elmasr
i@cse.uta.edu fabio@deis64.cineca.it kaefer@informatik.uni-kl.de rts@cs.arizona
.edu

A few ground rules are in order. First, the group's charter listed the
following constraints.

* The initial design must be completed by August 23, 1993.

        I'll send more on this in a subsequent email message.

* The design will be based on SQL2. In particular, none of the
        significant extensions envisioned in SQL3 will be assumed
        to be present.

* Initially, the language design will support user-defined time and
        valid time. Support for transaction time may be added later.

        We need to discuss this further, per Ilsoo's email message.

* The design will be implementable in the context of conventional
        relational database technology. In particular, implementations
        of the the data model and query language constructs will be
        possible without necessitating major changes to conventional
        database management systems.

* Discrete time will be assumed.

        We need to discuss this further, per Ilsoo's email message.

* Past and future valid time will be supported.

The initial goal of this working group will be to develop a language
design for a temporal extension of SQL.  This is clearly a very
ambitious goal. To encourage significant progress, I propose the
following procedural rules.

1. Complete unanimity will in general not be possible, and so should
not be required. On points of contention, the majority rules. It will
of course be best if we achieve near unanimity on the important
design aspects.

2. Rather than discussing abstract and often contentious issues, which
generally leads to stalemate, the discussion will focus on concrete
strawman proposals. In addition, the discussion will focus on semantic
issues, initially eschewing syntactic concerns.

3. Constructive criticism is preferable. In particular, criticism
should be accompanied by a concrete proposal that addresses the
concern.

4. Design tasks will be given time limits in order to ensure adequate
progress. Please respond quickly. Late responses, especially those
that attempt to reopen discussion of aspects for which consensus was
previously reached, are generally not productive. Those too busy or
otherwise unable to respond should be prepared to accept the group's
collective decision on the aspect under discussion.

5. A response is not necessary if one doesn't feel strongly about the
issue under discussion or agrees with the statement(s) put forth for
discussion.

Finally, let's resolve the authorship question early.  Any reports
from the working group will be authored by all participants who
contribute. I, as coordinator, will be the first author; the rest will
be listed alphabetically. Anyone having substantial disagreement with
any report can request that their name be removed.

If you have any concerns about anything mentioned above, please raise
it now, either with the entire group or with me. It is important that
everyone be comfortable with these rules. The above is at this point
only a proposal, and is not cast in stone.

On a related matter, this first mailing to the working group is by
necessity quite legalistic. In that way, I hope it is not
representative of future interactions. Let me emphasize that I think
that the participants of this group are ideally suited to the
challenging task ahead, and I anticipate that we'll all find the
discussions enlightening and ultimately very productive. I really do
think that this group can have a significant and lasting impact on
future temporal database research, standards, and commercial products.
I am looking forward to working with this group.

Sincerely,
        Rick