Re: newsgroup for asking general SQL questions?

On Apr 27, 2:36 pm, Bob Badour <bbad..._at_pei.sympatico.ca> wrote:
> Ed Prochak wrote:
> > On Apr 27, 2:00 pm, Bob Badour <bbad..._at_pei.sympatico.ca> wrote:
>
[]
> >>>what's the difference between the two groups?
>
> >>I haven't paid any real attention to comp.databases for years so perhaps
> >>I am not qualified to enumerate differences. That said, I would expect
> >>the difference in the names probably reflects the difference in the content.
>
> >>I expect you will find greater support for hackish kludges in
> >>comp.databases than in comp.databases.theory Here, you will find greater
> >>support for discussing the underlying mathematics.
>
> >>You will also find considerable overlap between the two groups so if in
> >>doubt cross-post to both groups.
>
> > Well, there are some of us that try to counter the desire for hackish
> > kludges. In particular the desire for many new developers to think all
> > you need for a primary key is a numeric ID column gets my attention
> > regularly.
>
> > We generally discuss the practical matters of data models, joins and
> > such. I try to avoid DB theory here, so I have sent some discussion
> > back to you guys (e.g. neo with his XDB).
>
> Ahhh, so we have you to blame for Neo coming around... Okay everyone,
> grab the pitchforks and torches--we're heading over to Ed's house!

Honey, pull up the draw bridge, we got company coming!
>
> Getting a poster's
>
> > datamodel to First normal form is often a bigger issue here than
> > getting it into Fifth normal form. There is more focus on whether a
> > query fits the SQL standard than whether it fits a mathematical model
> > of databases. So there is some high level experience here but turned
> > more toward day to day DB issues. Here it is more about the practice
> > of DB development.
>
> In other words, hackish kludges. The theory is the practice of database
> development.

Maybe I need to visit .theory more often.
>
> > If in doubt about which group to post to (comp.databases or
> > comp.databases.theory) I would not suggest posting to both. Maybe
> > default to here (comp.databases) if you are unsure. If it really is a
> > theory question, we will let you know either the answer (if it's an
> > easy clear-cut one) or point you to the theory group if it involves an
> > issue settled.
>
> Ed, what makes you think anyone at comp.databases would recognize a
> question with a theory-based answer? Wouldn't that require at least a
> little knowledge of the theory?

Joe Celko would be one. I hope I know enough to know what I don'tt know. So I think there are a few of us over there.

>
> > Others may chime in here but I think that gives a better summary of
> > comp.databases than a place for "hackish kludges". 8^)
>
> > Ed
> > PS no offense taken, at least not by me.
>
> I stand by my earlier suggestion: if in doubt, cross-post. It's not like
> cross-posting comp.object, which only a troll would do.

Very good point!

 I thought I was saving you from answering some really elementary questions, but if you want them, we in comp.databases are very willing to share.

Have a good day,

   Ed Received on Wed May 02 2007 - 05:17:46 CEST