Re: Schema 1.1: xs:anyEnumeration considered?

Pete Cordell writes:

> One of the primary use-cases I'm interested is in using schema in 
emerging 
> protocols having extensibilitys equivalent to HTTP, SMTP and (VoIP's) 
SIP. 
> These are highly extensible protocols, developed by people (the IETF) 
who 
> are intimately familiar with extensibility issues.

One of the things that became clear early to those of us who are trying to 
make progress on versioning and extensibility is that there are many, many 
use cases.  They differ in many dimensions, including the ones you point 
to.  While I can't say that we necessarily got a representative set of 
even the most important ones, we did try.  If you're interested, there are 
a couple of Web pages that are publicly accessible.  These include:

* XML Schema Working Group Versioning Resources (see [1]) -  This is 
basically a page with pointers to some other interesting pages, including:

* XML Schema Versioning Use Cases (see [2])

* We then tried to integrate a look at quite a range of mechanisms that 
had been proposed and how they would or wouldn't address at least some of 
the use cases (see [3])

* Also referenced is a very early "thought piece" that I wrote for the 
group in 2004.  It certainly did not at the time and does not now 
represent any sort of consensus of the workgroup, but it was among the 
analyses that was considered (see [4]).  BTW: it's labeled as a "rough 
draft", but as I recall no less rough draft was ever circulated. 
Interestingly, I think the mechanisms proposed for XML Schema 1.1 do fit 
moderately well with what I suggested as desiderata in this note.  That's 
not particularly because my note was adopted as the basis for our work (it 
wasn't), but I do note that where we landed is in some sense not too far 
from where I was hoping we would when we got serious about this 3 years 
ago.  FWIW: I would prefer to change the terminology a bit.  The paper 
uses the terms language and vocabulary more or less interchangeably, which 
I now think is unfortunate.  I'd prefer to think of a language as 
(roughly) a class of documents conforming to some particular 
specification.  I'd prefer to use the word vocabulary for things like tag 
names, and perhaps enumeration values, which are used in building those 
documents.

Noah


[1] http://www.w3.org/2005/05/xsd-versioning-resources.html
[2] http://www.w3.org/XML/2005/xsd-versioning-use-cases/ 
[3] http://www.w3.org/XML/2004/02/xsdv.html
[4] 
http://lists.w3.org/Archives/Public/www-tag/2004Aug/att-0010/NRMVersioningProposal.html

--------------------------------------
Noah Mendelsohn 
IBM Corporation
One Rogers Street
Cambridge, MA 02142
1-617-693-4036
--------------------------------------

Received on Tuesday, 20 March 2007 21:33:00 UTC