Re: Grinding to a halt on Issue 27.

raydwhitmer@aol.com wrote:

>
> tbray@textuality.com wrote:
> [...]
>
>> In any case, at the moment, we're paralyzed on this issue because of 
>> these unresolved differences.  This is on the face of it at one level 
>> ridiculous, because the first W in WWW stands for "World" and it's a 
>> no-brainer that identifiers ought to include non-ASCII characters.
>>
>> I think we do generally agree that the IRI work is in a good and 
>> useful direction, and that one thing that would be totally useful 
>> would be to get behind the work on the IRI draft:
>>
>>   http://www.w3.org/International/iri-edit/
>>
>> And get that nailed down and blessed.
>
>
> FWIW the DOM WG has been wrestling with use of IRI's and coming close 
> to being blocked by the issue, as well:
>
> http://lists.w3.org/Archives/Member/w3c-dom-ig/2003Apr/0036.html
>
> I cite the above note, because it contains my own latest proposal, 
> which appears dealing with some of the same sub-issues, although there 
> is at least one sub-issue I have not seen mentioned: how to start 
> allowing IRIs, when only a subset of IRIs are URIs.  URIs were 
> "Universal", and universal was not enough?  We already have normative 
> attributes and parameters calling them URI's.
>
> On the other hand, after coming close to this issue, we do not 
> consider it possible to avoid it.  Even if it were dropped from XML 
> Namespaces 1.1, which I suspect is not a good idea, the members of the 
> DOM WG are not likely to be around to fix DOM when URIs get replaced 
> by IRIs or something even more universal that comes along later (and 
> there are a number of improvements I could suggest to make them "more 
> universal").
>
> As such, lacking any stable specification, the latest proposal 
> declares what the DOM specification means when it says URI: something 
> that really is universal (hopefully a superset of URI spec, future 
> IRIs, and The Next Great Thing), represented by a string, and has a 
> few basic characteristics.  The characteristics were derived by 
> looking at the DOM spec, and what it needs to be able to do with these 
> universal identifiers: complete incomplete (relative) identifiers, 
> compare for equality, etc. but the details of completion and retrieval 
> are deferred to the spec, (i.e. HTML, XML1.0, XML 1.1) being implemented.
>
> I am not sure it is the best solution, but it is what we are 
> considering at our teleconference tomorrow.  Others have suggested we 
> should only refer to specific specifications for identifiers, but it 
> is not clear to me how that is compatible with the recent request from 
> the I18N WG "replace all URIs with IRIs" and the similar wisdom that 
> is sure to follow when The Next Great Thing appears.

I apologize for previously referring to something not publicly 
accessible.  In this case, the proposal was entirely mine and can be 
publicly displayed.  I quote it here so that response to my contribution 
to this thread is not inhibited, as long as everyone remembers this 
proposal is not presently approved as part of any working draft and is 
only my own effort:

Well, my attempt to get some answers on how to make the switch from URIs 
to IRIs in the DOM specification has been a bit less than successful:

http://lists.w3.org/Archives/Public/public-iri/2003Apr/

I see that there are somewhat-related issue before the TAG.

http://lists.w3.org/Archives/Public/www-tag/2003Apr/0090.html

I think we need to back up and specify less, doing it ourselves, 
something like this:

The DOM specification relies on DOMString values as resource 
identifiers, such that:

1.  A complete identifier absolutely identifies a resource on the web
2.  Simple string equality establishes equality of complete resource 
identifiers, and no other equivalency of resource identifiers is 
considered significant.
3.  An incomplete identifier is easily detected and completed relative 
to a complete identifier
4.  Retrieval of content of a resource may be accomplished where required

Within the DOM specification, these identifiers are called URIs, 
"Universal Resource Identifiers", but this is meant abstractly. 
 Absolute URIs refer to complete identifiers and relative URIs refer to 
incomplete identifiers.  The DOM specification is only bound by the 
rules for resource identifiers required by the markup language being 
represented.  For HTML 4.0 and XML 1.0, this would be URIs [link], and 
for XML 1.1 [link], this would be IRIs.  Generally the particular form 
of the identifier will be ignored.  But where incomplete identifiers 
must be completed or content must be retrieved the DOM implementation 
must at least support identifier types appropriate to the content being 
processed.

Thanks,

Ray

Received on Tuesday, 29 April 2003 10:47:50 UTC