Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!watmath!clyde!cbatt!ihnp4!qantel!intelca!oliveb!glacier!decwrl!amdcad!amdimage!prls!philabs!linus!raybed2!applicon!hdsvx1!hoffman
From: hoffman@hdsvx1.UUCP (Richard Hoffman)
Newsgroups: net.database
Subject: Re: Problem with relational theory
Message-ID: <504@hdsvx1.UUCP>
Date: Tue, 9-Sep-86 08:32:31 EDT
Article-I.D.: hdsvx1.504
Posted: Tue Sep  9 08:32:31 1986
Date-Received: Sun, 14-Sep-86 07:09:24 EDT
References: <595@ur-tut.UUCP> <452@hdsvx1.UUCP> <819@navajo.STANFORD.EDU>
Reply-To: hoffman@hdsvx1.UUCP (Richard hoffman)
Organization: Schlumberger HDS, Houston
Lines: 38

In article <819@navajo.STANFORD.EDU> avg@navajo.UUCP (Allen Van Gelder) writes:
>I disagree with Hoffman's attempted justification of the way
>Eric Carleen's query is handled.  Having a calculus-based language
>loses its point if the user has to know how the system is going
>to translate it to relational algebra, possibly incorrectly.

I wasn't attempting to justify it, merely to explain it.  I agree with you
that QUEL should be kinder to the non-RC-aware user; nevertheless, the
story I told is the reason that Eric's query was not handled the way he
expected it to be.

>[Eric's] query, essentially:
>	retrieve t1  where t1.name = t2.name  or t1.name = t3.name
>has an obvious meaning in relational calculus, and he has a right
>to expect the answer to be based on this meaning.

>The correct way to translate an "or" in a selection is as a union.

RTI implements "or" across multiple tables as a join (or, as I explained,
cartesian product followed by selection).  They do not implement unions
at all (although I imagine that they will have to do so before they finish
their fully distributed data base system).  There you have part of the
problem with Ingres -- it does not fully support the relational model, in
that important operators such as "union" and "difference" and "divide" are
missing (note that most of these can be obtained through user work-around,
or multiple queries, but that's not the point).  In fact, after 15 years
of work and talk and usage, I believe there is still not a single commercially
available RDB which successfully implements the entire relational model.

After listening to a steady stream of user complaints about and problems
with Ingres for almost three years, I am well aware of its many faults.
However, I have yet to find a better overall alternative in terms of
cost, performance, ease-of-use and maturity of the human interface.
-- 
 Richard Hoffman                | "Oh life is a wonderful cycle of song,
 Schlumberger Well Services     |     A medley of extemporanea.
 hoffman%hdsvx1@slb-doll.csnet  | And Love is a thing that can never go wrong
 PO Box 2175, Houston, TX 77252 | ... And I am Marie of Roumania." --D. PARKER