RE: [www-ql] <none>

At 06:05 PM 3/2/2001 +0100, Ingo Macherius wrote:
>Jonathan,
>
>I think you are twisting facts a bit. Algebra is preceding FLWR-XQuery in
>both time and completeness. Quilt preliminarily published many things which
>are results of Algebra. And just because Algebra voluntarily decided to drop
>early publication in favour of maturity, i don't think it is very fair to
>argue it should now adopt to XQuery.

Ingo, I can't comment on internal Working Group history in a public forum. 
We can discuss this in Germany in April when I visit next if you are 
interested in knowing the history.

> > the same period, and we were talking to each other, I think that the two
> > are broadly compatible. For any incompatibility we identify, there is no
>
>Broadly ? I'm a bit afraid until I can hear "fully".

The mapping in the current XQuery draft is basically complete. But it is 
the first published draft of this mapping. Take a look at it, if you want, 
and point out any bugs you find in the mapping.

> > general principle that tells us whether XQuery should change or
> > the Algebra
> > should change to support it. That's something the Working Group has to
> > decide on an issue-by-issue basis.
>
>There has been much more verification on the soundness of Algebra, using
>formal methods and use cases, then for FLWR-XQuery. There is a significant
>personal overlap between the authors of Quilt/FLWR-XQuery and the W3C-WG,
>and Quilt profited a lot from Algebra discussions. Given this, it is the
>"Bringschuld" (debt to be discharged at creditor's domicile) of XQuery to
>move first.

Well, I personally think the right thing to do is to look at each feature 
individually and figure out (1) what users want in the language, and (2) 
how this is realized in the Algebra. Since you are a W3C member, you could 
ask Peter to spell out for you what our internal discussion on this 
question. If you want to suggest that we adopt any particular first 
principles for this, you could send your comments to the comments list.


> > pleased to
> > see how easy it was to map XQuery onto the Query Algebra.
>
>Sounds promising. Do you have a URL for that mapping description ?

This is in the XQuery draft, Apppendix E.

>  And when
>do you expect the inverse mapping, Algebra->FLWR-XQuery, to be completed ?

Is there a reason to do such a mapping? I don't think this is possible, as 
I have said before.

Jonathan

Received on Friday, 2 March 2001 12:35:54 UTC