* Robert Haas (robertmhaas@gmail.com) wrote:
> On Mon, Oct 27, 2014 at 11:45 AM, Stephen Frost <sfrost@snowman.net> wrote:
> > I don't think Kevin was wrong to use a different relkind, but I don't
> > buy into the argument that a different relkind means it's not a view.
> > As for the other comments, I agree that a matview is *more* than a view,
> > but at its base, in my view (pun intended), it's still a view. Why not
> > call it a materialized query?
>
> Your view seems very odd to me. The access characteristics of a
> materialized view are completely unlike those of a view, and
> completely like those of a table. It has storage, and maybe indexes.
I agree with this, certainly, but these are not considerations that the
SQL spec takes into account. I've always found it odd of the spec to
avoid these considerations and concerns, but it is the spec and it's
viewpoint that we're discussing.
> Trying to say that it's the same kind of an object as something that
> has neither seems really odd. The overlap between the operations you
> can do on a materialized view and those you can do on a view is really
> pretty small.
... That overlap is exactly the set that you can do on *just* a view,
no? That's what I was driving towards anyway.
> It might have been better if the database industry had settled on some
> name for this kind of object that didn't reuse the word "view", but at
> this point I think we're stuck with the fact that adding the word
> "materialized" makes it into a completely different kind of object.
What would have been great is if the spec had decided to address these
issues, but I don't expect that to change.
> You wouldn't expect to find "butter" and "peanut butter" in the same
> aisle at the supermarket....
No, though they are both spreadable and tasty. :)
Thanks!
Stephen