Re: HTTP Link Use cases

As I believe Tuesday is CT TF day, I'll just give a little update on 
this. There is a wiki page being maintained by Jonathan Rees on this 
issue at [1]. He repeated his call for use cases over the weekend [2] to 
which I responded [3]. There's a load more on this issue over on the TAG 
list (and, confusingly, on the HTTP-WG list too) but the consensus seems 
to be emerging around Mark Nottingham's (updated) draft at [4].

The key discussion has been around the value of the rel attribute. The 
way it's heading is as Mark N sets out. The value is a URI. If a 
relative URI is given, such as 'stylesheet', then the assumption must be 
that it is relative to the IANA registry's namespace which is 
http://www.iana.org/assignments/link-relations.html#".

If you don't want to use an IANA rel type (and don't want to register 
it) then you give a complete URI, such as 
http://www.w3.org/2008/03/noTransform.

Such an HTTP Link Header would be semantically equivalent to an HTML 
Link tag or an RDF triple of the form

<resource> describedBy <description>

(Indeed, describedBy may well be the term used, in which case, the 
POWDER WG may well choose to use this and not rel="powder" - time will 
tell).

HTH.

Phil.

[1] http://esw.w3.org/topic/FindingResourceDescriptions
[2] http://lists.w3.org/Archives/Public/www-tag/2008Mar/0105.html
[3] http://lists.w3.org/Archives/Public/www-tag/2008Mar/0114.html
[4] http://tools.ietf.org/html/draft-nottingham-http-link-header-01

Jo Rabin wrote:
> Hi Phil
> 
> Thanks for the gentle reminder of the need for action. ISSUE-238 and
> ACTION-703 both have a direct relevance to this, and both are
> essentially down to me to chase. I hadn't forgotten our call, or either
> of these actions, it's just that I have been running around like a
> headless chicken since the Seoul F2F and have only just managed to get
> my head back above water (if that is not too grotesque a mixed
> metaphor).
> 
> There will definitely be input from BP on this, but there are definitely
> only 24 hours in each day.
> 
> Jo
> 
> ---
> Jo Rabin
> mTLD (http://dotmobi.mobi)
> 
> mTLD Top Level Domain Limited is a private limited company incorporated
> and registered in the Republic of Ireland with registered number 398040
> and registered office at Arthur Cox Building, Earlsfort Terrace, Dublin
> 2.
> 
> 
>> -----Original Message-----
>> From: public-bpwg-request@w3.org [mailto:public-bpwg-request@w3.org]
> On
>> Behalf Of Phil Archer
>> Sent: 17 March 2008 09:53
>> To: Public MWBP
>> Subject: HTTP Link Use cases
>>
>>
>> Jo and I discussed this on the phone the other day so this note is
>>
>> a) to remind him of that conversation;
>> b) alert other members of the group to the issue.
>>
>> We have discussed the potential usefulness of the HTTP Link Header in
>> the mobile space in past meetings (I recall doing so most recently at
>> TPAC last year). The issue continues to surface and resurface on the
>> IETF/W3C HTTP group and has lead to some very recent and extensive
>> discussion. Happy Halpin kicked things off this time [1] and this lead
>> to mark Nottingham breathing new life into his draft [2]. I chimed in
>> with the POWDER use case [3]. In between these are messages from the
>> likes of Roy Fielding and Julian Reschke.
>>
>> The bulk of the discussion centred on the need for/best approach to
>> providing an HTTP Profile header, i.e. an extensible and unambiguous
> way
>> to extend relationship types. It's not as easy as it sounds...
>>
>> If the MWBP in general, and the CTTF in particular, wishes to support
>> the reinstatement of HTTP Link and comment on the wider discussion.
> NOW
>> is the time when such an input can have most effect.
>>
>> Cheers
>>
>> Phil.
>>
>>
>> [1]
> http://lists.w3.org/Archives/Public/ietf-http-wg/2008JanMar/0444.html
>> [2]
> http://www.mnot.net/drafts/draft-nottingham-http-link-header-01.txt
>> [3]
> http://lists.w3.org/Archives/Public/ietf-http-wg/2008JanMar/0499.html
>

Received on Tuesday, 25 March 2008 15:00:45 UTC