Re: View updating in practice?

"Paul Vernon" <paul.vernon_at_ukk.ibmm.comm> writes:

> Ohoo, this could be a long running thread :-)
>
> Were you following the discussions eailer about Date & McGoveran's
> view updating scheme? If not, start from about message 169 on this
> thread
> http://groups.google.co.uk/groups?hl=en&lr=&ie=UTF-8&threadm=fedf3d42.02100108
> 58.76f2f1e5%40posting.google.com&rnum=1

I checked google before posting, but the thread did not contain what I was looking for :(

>
> (P.S. anyone know the best way to capture concise google groups
> URLs ?)
>
> "Jens Lechtenbörger" <lechtej_at_uni-muenster.de> wrote in message
> news:m2of9087s0.fsf_at_pcwi1068.uni-muenster.de...
> > Dear reader,
> >
> > I wonder if anybody uses view updates in practice and if so what
> > kind of updates on what kind of views.
> >
> > Personally, I believe that view updates are dangerous in most
> > situations as users cannot understand what is going on.
>
> This is similar to my general feelings to. It is my position that
> unless views are 'information equivalent' to their underlying base
> relvars (which is, I believe is equivalent to taking a constant
> complement), then view updates should simply be seen as a
> shorthand - a way for users to get the DMBS to generate base
> update statements from the users desire of what the value of some
> relational expression looks like.

The constant complement approach is less restrictive. E.g., under constant complement you may issue UPDATE statements against projection views, or INSERT into selection views.

> >To give two
> > simple examples: In case of deletions from projections, users
> > delete tuples they do not know completely. In case of
> > insertions into joins, users do not know whether they are
> > actually inserting into both operand relations or whether there
> > was a dangling tuple in any of the operand relations that gets
> > "absorbed" by the newly inserted tuple. It is easy to see that
> > users will not be able to undo their view updates (such that the
> > old _database_ state is reached) using further view updates in
> > the above situations. If a was a user of such a database I
> > would be very irritated.
> >
> > I wrote a paper on this topic for ICDT'03. (Take a look at
> > http://dbms.uni-muenster.de/publications/downloads/view-updates.pdf
> > if you are interested in the details.) Unfortunately, the paper
> > got rejected, but I'm not convinced by the reviewers' critique
> > ;) Now, I would like to know what people do in practice. My
> > questions are as follows:
>
> I'll read it over the weekend. I would be interested if you could
> briefly outline what the reviewers' critique was.

Here is a short summary:

One reviewer really liked the paper and voted to accept it.

The other three reviewers all voted to reject: The first did not have specific criticism but would rather see the paper in SIGMOD Record than in ICDT.
The second wanted to see more concrete results on schema design as proposed briefly on p. 9.
The third reviewer found the main theorem "not at all surprising", rating it a marginal contribution and questioning the usefulness of constant complement translators; next, s/he wanted to know what "principles are violated" if users cannot undo view updates using further view updates; then, s/he found my schema design argument (p. 9 again) not convincing and the issue of inheritance especially confusing; finally, s/he wondered why database users were not complaining if there was a problem with view updates.

Well, one reason to start this thread were the comments by the third reviewer. I really want to know who is using view updates for what application scenarios.

> > 1. Do you use view updates? What is the scenario?
> >
> > 2. If you do _not_ use view updates, did you at least think
> > about
> > using them? If so, what was the application scenario and why
> > did you not implement it (e.g., no support by DBMS or too
> > complicated for users)?
> >
> > 3. Does anybody use SQL:1999 INSTEAD triggers to implement view
> > updates?
>
> I very probably will do when I get the chance to use DB2 V8 on a
> project (it's still in beta at present). In fact I could do with
> them on my current project, I'll write up the scenario here if I
> get the time.

I hope you get the time :-)

Best wishes,

Jens Received on Sat Nov 09 2002 - 23:41:46 CET