Re: [xri] URI Fragment in <Subject> or <Alias>

[John Bradley <jbradley@mac.com>]

Have I recently mentioned how I feel about httprange-14?

This is one possible interpretation.

A subject of https://test.example.com/jbradley#1234

Unless the fragment resolves to something internal to the XRD the XRD  
is about a non information resource (person in this case)

The XRD describes relationships to the person not the URI https://test.example.com/jbradley

This allows for differentiating between different non-information  
resources discoverd by the same URI.

In openID this could be used to provide identifier recycling detection.

So the subject is a non-information resource and the XRD describes  
relationships to that resource.

Being able to differentiate XRDs that describe things as opposed to  
URI may have advantages.
Though I doubt that anyone would understand it.   The whole abstract  
identifier debate is so convoluted in http as to not be useful.

The TAG has held for a long time that openID that represent persons in  
meat space should be abstract and that openID is broken as a result.

I don't see it bothering anyone though.

Good luck with that Eran:)

John B.
On 26-Jun-09, at 10:41 PM, Scott Cantor wrote:

> Scott Cantor wrote on 2009-06-26:
>> Eran Hammer-Lahav wrote on 2009-06-26:
>>> I am not sure we need to say anything about it, but I can't figure  
>>> out
>>> what it means if an XRD has a subject with a fragment as part of the
>>> URI.
>>
>> I don't think it means anything, in the context of a URI as an  
>> abstract
>> identifier for something that may or may not resolve. They're best  
>> avoided
>> simply because they create confusion, but don't really hurt anything
> either.
>>
>> The only meaning they can have would be relative to resolution for
>> particular MIME types that define the meaning, and I don't think that
>> applies to an identifier.
>
> I didn't say that well...presumably the subject of an XRD could be a
> resource or it could be an abstract thing identified by the URI. I was
> speaking more to the latter.
>
> If it's an actual resource, then I would assume the the fragment  
> would have
> meaning relative to the MIME type that results, as usual, and that  
> the XRD
> would be about that subset of content, and not the whole resource.
>
> If not, then I guess I'd suggest precluding fragments, because it  
> wouldn't
> make sense for two different Subjects differing by fragment to be  
> meant as
> the same thing.
>
> -- Scott
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe from this mail list, you must leave the OASIS TC that
> generates this mail.  Follow this link to all your TCs in OASIS at:
> https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php
>