W3C home > Mailing lists > Public > xml-dist-app@w3.org > September 2000

RE: Schematron schema for SOAP 1.1 Envelopes

From: Henrik Frystyk Nielsen <frystyk@microsoft.com>
Date: Thu, 21 Sep 2000 12:03:49 -0700
Message-ID: <005601c023fe$b47fe5c0$fb4c1fac@redmond.corp.microsoft.com>
To: "S. Mike Dierken" <mike@knownow.com>, <xml-dist-app@w3.org>

> Relax people. Let's not put words in other people mouths.

Thank you for the wise words!

> Everybody is in violent agreement that the URI of a namespace is not
> required to be able to be used to access a resource of some kind. Many
> experienced people have written many times that doing so is a
> Bad Idea, and
> provide reasons and alternatives. Several orders of magnitude
> more people
> are going to try to do it anyway - I think that was the point
> Henrik was
> trying to make.

Regardless of whether one believes this is a good idea or not, what I am
trying to get at is that the properties of a URI including whether a URI
can be dereferenced is partially defined by the URI space (http:, ftp:,
mailto: etc) and if there is a known way to dereference a URI then as
David E. Cleary points out [1], then it is up to the application to decide
whether it wants to or not.

Absolutely as a side-comment explaining my earlier mail - I don't think
this is the right place to discuss this and so please don't start a long
thread on this on this list: I happen not to like schemaLocation as well
as targetNamespace because they are not consistent with the simple Web
model of the relationship between a URI and a document where a Web
document is what you get when you resolve a URI identifying a resource. In
other words, a document is a manifestation of the resource. The result is
that we now need two identifers (xml-ns and schemalocation) instead of one
(even though both can be ignored) and I think it is unfortunate to special
case schemas in this way. This doesn't mean that you are guaranteed a
schema - but you are not guaranteed that with schemaLocation either. That
is just a fact of life in a distributed environment.

> The links David sent in his message are very useful in
> understand more of
> the issue, but that isn't going to change the fact that
> avoiding the use of
> 'namespace URI == schema definition' requires user education.
> When I read Rick's message, it was very useful to have actual
> examples in
> XML of what he was talking about. I suggest that future
> discussions about
> this issue use real XML to show what the alternatives are and
> clearly show
> what problems will happen.


[1] http://lists.w3.org/Archives/Public/xml-dist-app/2000Sep/0026.html
Received on Thursday, 21 September 2000 15:04:41 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 22:01:10 UTC