W3C home > Mailing lists > Public > xmlschema-dev@w3.org > October 2004

Re: 3rd try on versioning question

From: <noah_mendelsohn@us.ibm.com>
Date: Thu, 21 Oct 2004 12:53:21 -0400
To: "Dean Hiller" <dean@xsoftware.biz>
Cc: xmlschema-dev@w3.org, David_Orchard/CanWest/IBM%IBMCA@lotus.com
Message-ID: <OFE3C5BE79.24687DA3-ON85256F34.005B26E2@lotus.com>

Dean Hiller writes:

> wow, alot of great responses this time. 
> I really like the counter on the html example.
> That is really good. 

I'm glad it was helpful, thank you.

> I will have to think more about that.  It really
> depends on the way html authoring is done these
> days.  I guess we are not to the point where it
> is primarily done with tools yet especially when
> it comes to jsp's.

Perhaps there is still a bit of confusion.  HTML is only an example.  Many 
users of XML have vocabularies that would look unnatural or inconvenient 
if they sprouted explicit version control on individual instance elements 
after the initial release.  Whatever we do needs to anticipate the needs 
of such users, not just those who author HTML.

You might be interested in an analysis that I did for the schema WG and 
later posted in a publicly accessible archive [1].  This analysis is not 
consensus of the Schema WG;  there are other members of the WG who have 
somewhat different view of these issues and who especially would differ 
with some of the mechanisms discussed in the second part of the note.  You 
may also want to keep an eye on the work that David Orchard and Norm Walsh 
have been doing toward a TAG finding [2] on XML Versioning (draft at 
[3]--I wouldn't be surprised to see new drafts soon). 

At the very least, I hope that you will get a feeling that we are all 
trying hard to understand the requirements and use cases, and that taken 
together those use cases embody a broader range of concerns and 
constraints than many casual observers might notice.  Whether we can in 
fact do something useful in this space, either by providing explicit 
mechanisms or best-practices advice remains to be seen.  Versioning is 
known to be a very, very hard problem.


[1] http://lists.w3.org/Archives/Public/www-tag/2004Aug/0010.html
[2] http://www.w3.org/2001/tag/findings
[3] http://www.w3.org/2001/tag/doc/versioning-20031003

Noah Mendelsohn 
IBM Corporation
One Rogers Street
Cambridge, MA 02142
Received on Thursday, 21 October 2004 16:56:16 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 14:56:06 UTC