Re: RELAX NG and W3C XML Schema

["Eric Sedlar" <eric.sedlar@xxxxxxxxxx>]

> If that's what you were really after, why exactly did you call it "XML
> Schema" and thereby inflict the needs for this practice on the rest of us
> XML users?
>
> Gack.  If I wanted SQL, I'd use SQL.
>

First of all, I didn't author the Schema spec.  The Oracle representative
who did since retired, so I don't even know why he made some of the
decisions he did.

Schema wanted to support all of the document-centric stuff too.  So you are
willing to use an XML type description language that has a hard time
supporting existing search technologies, and are willing to cut off a huge
installed base?  The IETF has always favored getting a broader set of
technology platform implementations of its standards over ideological
purity--this doesn't sound like IETF-zen to me.

> >I agree that there is no need to privilege a particular XML structure
> >definition spec at this time.
>
> Good.  That will let this discussion proceed without further conflict.
>
> >These same battles will be fought out in the
> >individual IETF working groups anyway regardless of what this document
says
> >anyways.  I just felt that there were a number of accusations about XML
> >Schema on this that were poorly supported that needed to be answered.
>
> I think James Clark has already done an excellent job of deconstructing
W3C
> XML Schema. If you would like me to repeat the performance, I probably
can,
> though I don't have the time or his level of experience.  Your grouping of
> his complaints effectively watered them down, so I guess I still consider
> most of them unanswered.
>

James Clark's comments are mostly well-taken and my complaints about poorly
supported accusations do not apply to his message.  I didn't water down his
complaints--I merely said that they need to be taken in context of:

* stuff that is easy to fix in the next version of Schema
* stuff that can be fixed via supporting documentation (overly formalistic &
difficult to read spec)
* different but reasonable architecture choices

I just believe that his comments can be addressed by Schema without having
competing specs.  Whereas there is a reasonable complaint about getting
input into the W3C by open-source folks, that doesn't apply to James, who
clearly has had the chance to get his voice heard in that organization.

--Eric