Re: XQuery and GET

Hi Jonathan,

On Fri, Oct 17, 2003 at 09:39:17AM -0400, Jonathan Robie wrote:
> Hi Mark - I am not sure whether I understand this. Are you simply saying 
> you want to be able to write a query to create another query, then use that 
> query as an argument to GET?

No, just to make a URI out of an XQuery query.

(I really should have used "XQuery and URIs" as the subject of my initial
email)

> >Consider the recent TAG finding regarding the use of HTTP GET;
> >
> >"Use GET if [...] The interaction is more like a question (i.e., it is
> >a safe operation such as a query, read operation, or lookup)."
> >  -- http://www.w3.org/2001/tag/doc/whenToUseGet.html
> >
> >Being able to perform a query via GET requires that a URI serialization
> >of the query (i.e. the XQuery document) be provided.  Without this, the
> >only way that querying can be performed interoperably would be via POST.
> 
> True, but I think it's also important to keep in mind that:
> 
> >Web application design should be informed by the above principles, but 
> >also by the relevant limitations.
> >-- http://www.w3.org/2001/tag/doc/whenToUseGet.html#practical
> 
> There are limits to what fits easily on a URI. A 12 page query with 
> functions and schema imports probably doesn't naturally fit into this 
> framework.

Probably not, no.  But for smaller, perhaps simpler queries, I think it
would be enormously useful.  Where the WG decides to draw this line, I
don't know (FLWR??).  But I also wonder if drawing a line may be more
work than covering the general case, and then letting your customers
(XQuery processor designers and app-developers) decide what's too big to
go in a URI. 8-)  That's totally your call, of course, but anything you
can do in this direction would be progress, I'd say, and would address
my concern.

Thanks.

Mark.
-- 
Mark Baker.   Ottawa, Ontario, CANADA.        http://www.markbaker.ca

Received on Friday, 17 October 2003 12:30:54 UTC