iainshigeoka at yahoo.com
Mon Mar 4 17:18:24 UTC 2002
On 3/2/02 2:10 PM, "Justin Kirby" <justink at rochester.rr.com> wrote:
> I have updated the doc to incorporate some of the ideas...
> namely, the embedding sql
> XPath would drastically increase the complexity of the protocol without
> much benefit. esp now that SQL can be embedded.
Looks good. You may want to watch the discussion on standards-jig for
JEP-0020. It may be useful for version negotiation in your spec.
I think the JEP should also discuss the pros and cons of XPath and XQuery as
you see it and why you decided to create a new XML database query syntax
instead. This only needs to be a paragraph or two but should clearly
explain the reasoning behind the decision. I think the default should be to
use existing standards so this places the burden on you to justify not doing
It's shaping up though!
> On Fri, 2002-03-01 at 13:41, Iain Shigeoka wrote:
>> On 3/1/02 12:06 AM, "Fabrice DESRE" <fabrice.desre at francetelecom.com> wrote:
>>> Iain Shigeoka wrote:
>>>> I don't want to sound contrary but wouldn't it be easier to just use raw
>>>> inside of the <query> rather than XMLify the entire query?
>>> I agree. And it offers the possibility to use the full range of SQL
>>> expressiveness. Take a look at the XML format for XQuery to have an idea
>>> of how complex and "humanly" unusable an XML description of such a query
>>> language could be... (a simpler but equally meaningful example would be
>>> an XML syntax for XPath)
>> Yes, it is pretty messy (and verbose).
>>>> If we want to go XML for XMLs sake, we may be better served in using XML
>>>> Query since it is a standard. Why reinvent the wheel? It does lack
>>>> but we could use the most popular extension of it for updates...
>>> Because XQuery can only query XML databases, and not
>>> relationnal/object relationnal ones. It will take long before these
>>> databases disappear...
>> I'm not sure that is as much an issue. Most database vendors seem to be
>> doing XML bridges to their RDBMS and XQuery seems to be the way to go.
>> However it really seems like SQL for RDBMS is going to be superior for a
>> I'm not sure if I'm being consistent at all here... :)
>> Do You Yahoo!?
>> Get your free @yahoo.com address at http://mail.yahoo.com
>> Standards-JIG mailing list
>> Standards-JIG at jabber.org
> Standards-JIG mailing list
> Standards-JIG at jabber.org
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com
More information about the Standards