Re: [XRI] XRI-as-Relative-URI proposal

Hi Stuart,

I did a blog post on how XRI relate to cool URI at http://thread-safe.livejournal.com/14165.html 
..

I am traveling at the moment and will get to some direct comments on  
the weekend.

We are exploring Link headers as part of our work on XRDS-Simple.  We  
intend to move in that direction for our http: and https: bindings.    
We are in contact with Mark Nottingham on the topic.

Given the current non-standard status of link headers we want to  
coordinate with other efforts before we make specific proposals on how  
XRI  will use them in the http: sub scheme.

I will post a copy of my blog post to the email list for the record.

Regards
John Bradley
http://xri.net/=jbradley


On 23-Oct-08, at 11:18 AM, Williams, Stuart (HP Labs, Bristol) wrote:

>
> Hello Drummond et. al.,
>
> A couple of comments:
>
> --
>
> I think this chimes in a bit with Jonathan's comment. I think it  
> might be helpful to say a little more about how the proposed http(s)  
> bindings might use 303 and/or link-headers to provide separate  
> identifiers for 'thing' and 'description of thing' (or indeed if  
> that's the plan, how as Jonathan suggests, you might generate  
> distinct for each from a given 'pure-XRI'). I'd also say that *if*  
> the thing that the identifier denotes is a document (or more  
> generally and information resource) it's quite in order to give a  
> 200 response are return a webarch:represention of the resource -  
> it's equally in order to provide indirections to descriptions that  
> refer to other places where equivalent resources are available as  
> well.
>
> --
>
> I think that it would be helpful to spell out that, from an XRI pov,  
> the following are necessarily synonym pairs
>
> Public community roots
> http://boeing.com.xri.net/@boeing*example/+documents and http://oasis-open.org.xri.net/@boeing*example/+documents
> http://oasis-open.org.xri.net/@oasis!2047!22/!7 and http://boeing.com.xri.net/@oasis!2047!22/!7
>
> Private community roots
> http://boeing.com.xri.net/@(http://boeing.com)*example/+documents  
> and http://oasis-open.org.xri.net/@(http://boeing.com)*example/+documents
> http://oasis-open.org.xri.net/@(https://oasis-open.org)!2047!22/!7  
> and http://boeing.com.org.xri.net/@(https://oasis-open.org)!2047!22/!7
>
> and IIUC the following are not (necessarily synonym pairs):
>
> Proxy delegation
> http://boeing.com.xri.net/*example/+documents and  http://oasis-open.org.xri.net/*example/+documents
>
> ie. from the examples folks may be lead to think that the oasis or  
> boeing bits in the URI authority have to match with the oasis or  
> boeing bits of the 'pure-XRI' authority, when IIUC they don't.
> That is said more generically in the proposed bindings section, but  
> the examples might invite a wrong conclusion.
>
> --
>
> BR
>
> Stuart
> --
> Hewlett-Packard Limited registered Office: Cain Road, Bracknell,  
> Berks RG12 1HN
> Registered No: 690597 England
>
>> -----Original Message-----
>> From: www-tag-request@w3.org [mailto:www-tag-request@w3.org]
>> On Behalf Of Drummond Reed
>> Sent: 23 October 2008 08:05
>> To: www-tag@w3.org
>> Cc: 'Peter Davis'; jbradley@mac.com
>> Subject: [XRI] XRI-as-Relative-URI proposal
>>
>>
>> First, the XRI TC would like to thank the TAG and other members of  
>> this list
>> for the extensive feedback they have provided on XRI 2.0 since  
>> early July.
>> Proof that it has been productive has been the emergence of a new  
>> proposal
>> for how XRIs can better fit with AWWW architecture. This proposal has
>> received extensive discussion within the XRI TC and on the XRI TC  
>> mailing
>> list over the last month, and has had an initial preview with the  
>> TAG. They
>> found the proposal encouraging and suggested public discussion on  
>> this list
>> as a next step.
>>
>> The proposal is written up on an XRI TC wiki page at:
>>
>>        http://wiki.oasis-open.org/xri/XriAsRelativeUri
>>
>> I won't even attempt a one-sentence summary here as it would only  
>> duplicate
>> the summary there, and the page already serves as a mini-FAQ for  
>> most of the
>> questions it has generated.
>>
>> To those who have contributed to the XRI discussions here, and to  
>> anyone
>> else interested in the topic of abstract identifiers (in AWWW lingo,
>> identifiers of non-information resources that do not have direct
>> representations, only descriptions), we are very interested in your  
>> feedback
>> about this proposal.
>>
>> I'm sure other XRI TC members will chime in with their perspectives  
>> and
>> questions too (you'd need a team of a thousand wild horses to
>> stop them ;-)
>>
>> Thanks in advance,
>>
>> =Drummond
>>
>>
>>
>>
>