Re: FW: @rel type 'describedby'

Dear Ms Cotton,

Thank you for this.

The short answer to your question is yes please. The longer answer is 
'yes for now.'

Our working assumption is that Mark Nottingham's internet draft on HTTP 
link [1] will move from version 3 (due any day I understand from Mark) 
to RFC in due course. In that event, the ATOM registry would be 
superseded and we would like describedby to be included in the new registry.

I should also note a comment I have received from Julian Reschke today 
suggesting that we should change the ATOM example included in our Rec 
Track doc [2] so that it uses link @rel rather than a different XML 
element as currently shown, i.e. the current text:

<wdrs:describedby>http://ecw.example.org/powder1.xml</wdrs:describedby>

will become

<link rel="describedBy" href="http://ecw.example.org/powder1.xml" />

in the version we use in our request for a transition to Proposed 
Recommendation.

Although this is a relatively small change to the doc, it clearly has 
positive implications in the present discussion.

So, yes, while the ATOM registry your cite is indeed the correct place 
to consider our request to add 'describedby', we recognise that it may 
change in the relatively near future and indeed would welcome such a 
change as set out in the HTTP Link draft.

Please let me know if you need anything further from me. I'm travelling 
at the moment with intermittent connectivity so may not be as responsive 
as I'd like.

Thanks for your time and help.

Phil.


[1] http://tools.ietf.org/html/draft-nottingham-http-link-header-02
[2] http://www.w3.org/TR/2008/WD-powder-dr-20081114/#eg4-4

Michelle Cotton wrote:
> Dear Mr. Archer,
> 
> My name is Michelle Cotton and your message has been passed to me by Barbara Roseman.
> I would like to verify what you are requesting to register so that I can provide the correct registration procedures.
> 
> Are you looking for a registration in the following registry?
> http://www.iana.org/assignments/link-relations/link-relations.xhtml
> 
> Thank you,
> 
> Michelle Cotton
> IANA
> 
> 
> 
> ------ Forwarded Message
> From: Phil Archer <parcher@fosi.org>
> Date: Wed, 19 Nov 2008 06:29:05 -0800
> To: <barbara.roseman@iana.org>
> Subject: @rel type 'describedby'
> 
> Dear Ms Roseman,
> 
> I was given your name following a conversation at a W3C meeting last
> week. I write on behalf of the W3C POWDER Working Group [1] to request
> the registration of a new link relationship type as follows:
> 
> Relationship type: describedby
> Purpose: to link a resource to a description that applies to
>           that resource
> Documentation: http://www.w3.org/TR/powder-dr/#assoc-linking
> 
> The Recommendations Track Document 'POWDER: Description Resources'
> (cited above) was published this week as a (second) Last Call and we
> noted that we do not expect to issue a separate call for implementations
> before seeking transition to Proposed Recommendation next month. I am
> also writing to IETF to register the MIME types documented in the same
> place.
> 
> Background
> ==========
> The Protocol for Web Description Resources (POWDER) defines a method by
> which descriptions may be applied to multiple resources, typically
> 'everything on a Web site.' The link relationship will be used to point
> from those resources to such a description, either in HTML link elements
> or through HTTP Link elements (currently under discussion through Mark
> Nottingham's Internet Draft [2] and, I understand, expected to be
> updated later this week and moved to RFC status subject to comments
> received).
> 
> The relationship A 'describedby' B does not imply that B is a POWDER
> file (the MIME type does that), simply that B provides a description of
> A. The representation returned from A and B is not constrained by the
> relationship.
> 
> Wider context
> =============
> I believe it is also appropriate to outline the broader context in which
> this request is made. There has been a good deal of discussion amongst
> various W3C Working Groups for more than a year on how @rel types should
> be managed. Various solutions have been proposed: the use of HTML 4's
> profile attribute being one, writing new types into a wiki being another
> and so on. Consensus has been hard to reach. At the recent W3C Tech
> Plenary, several groups, including POWDER, took part in a discussion
> with the HTML 5 WG on this issue. Although it would be wrong to suggest
> that there was unanimity on the way forward, there was general consensus
> that registering new relationship types should be a relatively
> lightweight process but clearly not so lightweight that it became
> unworkable.
> 
> Whether IANA decides to approve the POWDER WG's request to register
> 'describedby' or not, the process of registration is therefore something
> of significant interest beyond any one WG.
> 
> I have not copied this to POWDER's public mailing list as I am unsure
> whether that would be acceptable to you. If it is, I would be grateful
> if you would cc public-powderwg@w3.org in your reply.
> 
> Thank you.
> 
> Phil Archer
> POWDER WG Chair.
> 
> [1] http://www.w3.org/2007/powder/
> [2] http://tools.ietf.org/html/draft-nottingham-http-link-header-02
> 

-- 

Please note my new e-mail address. My ICRA/FOSI e-mail addresses will 
not function after the end of November.

Phil Archer
w. http://philarcher.org/

Received on Wednesday, 19 November 2008 23:22:17 UTC