XQuery as a general data processing language WAS: [xquery-talk] XQuery and Web 2.0

[Peter Coppens]

Well...ok....what about something like a non proprietary LINQ done  
properly?

It would still suffer 'the impossible to optimize'  issue but at least  
it would open up the libraries of the programming environment you are  
working with pretty seamlessly, would integrate nicely with the  
programming environments a lot of people care about today, and I am  
guessing that the actual query part that goes out to the indexed data  
store can be optimized. I understand it might not be easy to  
'identify' that part of the query ...perhaps that is what you mean  
with 'impossible'?

Personally I am really trying hard to use XQuery in my day to day  
development activities (I do like the language a lot) and the only  
thing I can conclude from my daily experience is that for most things  
at hand it introduces as much issues as it solves.

That means
(1) I am not using it properly and/or
(2) the language is still missing features and/or
(3) it is not the right tool for my job.

Show us normal mortals writing the less fancy apps of the web (and I  
tend to think that is were your eventual main stream adoption of the  
language will have to come from, as there are a lot of us) how to use  
XQuery.

I think I am entitled to claim I know the language good enough...but I  
just can't figure out how to put it to use.

Peter



On 26 Apr 2008, at 03:40, Daniela Florescu wrote:

>>
>> What do you think about e.g. LINQ which from what I understand of  
>> it seem to try and deal with this integration/mismatch problem?  
>> Would that not better fit better under the "general data  
>> processing" language umbrella?
>
>
> No, personally I don't think LINQ is the solution. For the following  
> reasons:
>
> (a) the XML they support is a bastardization of the normal XML
> so it will satisfy only some users (I never understood their choice  
> on that...POX is very
> powerful)
>
>
> (b) they only support a subset of the processing facilities that  
> people need
> (e.g. groupby, full text, windowing, dynamic namespace generation,  
> etc, etc).
> For the rest you go back to low level programming.
>
> (c) There is no way low level programs in C#+ XML/LINQ  can  be  
> optimized automatically
> as well as XQuery can be  optimized automatically on large volumes  
> of data, and/or indexed
> data (I am willing to take a beer bet with Erik Meijer any time on  
> that :-)))
>
> and finally (and probably the most important, or the fundamental  
> factor):
>
> (c) the world needs a good non proprietary language for data  
> processing.
>
> This is a way too important problem for people to accept to be  
> locked in by a
> specific vendor, with no open source implementations.
>
> Best regards
> Dana
>
>
>
> _______________________________________________
> talk at x-query.com
> http://x-query.com/mailman/listinfo/talk