Re: Moving on (was Re: URIs quack like a duck)

-----Original Message-----
From: David Carlisle <david@dcarlisle.demon.co.uk>
To: timbl@w3.org <timbl@w3.org>
Cc: xml-uri@w3.org <xml-uri@w3.org>
Date: Thursday, June 01, 2000 5:45 AM
Subject: Re: Moving on (was Re: URIs quack like a duck)


>
>Tim Berners-Lee wrote:
>> But for now we need a solution to this so that the DOM can move forward.
>>
>> The fact that relative URIs are terated diferently by different groups
>> already clearly means that they should be warned against as a minimum.
>>
>> Then, when you don't use then, URI comparison and string comparison are
the
>> same - something we can take advantage of by allowing XML software at the
>> lower laters to be simple in its comparisons, but allowing the full
richness
>> to the upper layers.
>>
>> This seems to me the only way this can go.  We have made many attempts to
>> make complex compromises and wacky alternatives, but I think we come back
to
>> the basic options considered
>> by the xml-plenary as the options.  I don't think this is a time for
>> compromise. The NS spec in adopting compromise wording (bits to please
each
>> camp) left open the mess we are in now.
>> I think we have to be clear.
>
>This sounds to me like a good propsal, I interpret it as saying
>
>* confirm the current literal comparison semantics for namespace names
>* Re-issue namespace spec saying using relative URI references is
>  a bad idea.
>* Re-issue xpath to use literal (probably with a repeat of the warning
>  about using relative uris)


Breaking the first point out into two,

Re-issue namespace spec saying:-

* Confirm that the namespace attribute is a URI-reference
* Point out that that this implies that litteral comparison and URI
comparison are
  equivalent so long a  relative URIs are not used;
* using relative URI references is  a bad idea, because existing software
   does different things with them.

XPath does not need to be re-issued as it will interwork, as relative
URIs are excluded.  Software which absolutizes the URI-reference
and uses the URI will be legal. So will software which compares as
strings.  Yes, it is is a compromise.

This is largely the path David D suggested earlier.


>If the spec was going to be re-issued with descriptions of good
>practice I would also be in favour of it (or an additional spec)
>recommending some format of file to place at the namespace URI
>in the case that you do use a dereferencable URI scheme.

That would be a note.  I think that it might be a godo idea to point out
for example that
- such a document is not mandatory
- the document may include xml-schema

- a document can contain xml-schema and also other information

>As Tim Bray mentioned some kind of `packaging' document that could
>refer to schema, stylesheets, or anything else would be good.
>(placing a schema file there dorectly is usually bad)
>
>
>David
>
>
>

Received on Thursday, 1 June 2000 17:45:49 UTC