W3C home > Mailing lists > Public > www-tag@w3.org > March 2008

Re: I-D ACTION:draft-nottingham-http-link-header-01.txt

From: Mark Nottingham <mnot@mnot.net>
Date: Fri, 21 Mar 2008 08:38:40 +1100
Cc: www-tag@w3.org
Message-Id: <AF72A65C-2B43-4051-B7C6-F67A7722BDDB@mnot.net>
To: Jonathan Rees <jar@creativecommons.org>

BTW, I came at this from a very different angle -- trying to resolve  
the conflicts in how HTML and Atom use links in the least intrusive way.

The Profile mechanism failed that test -- it didn't have the range of  
flexibility that Atom links have, and would make some uses very  
difficult -- if not impossible -- to deploy.

When that's out of the picture, there are a few other decisions, but  
only a few. There are certainly other ways to a solution for a  
particular application, but my work pre-supposes that you'll be using  
HTTP, and that you want to work with the linking models already  
existent in HTML and Atom.

Cheers,




On 21/03/2008, at 1:46 AM, Jonathan Rees wrote:

> The alternatives are surveyed in the bibliographic wiki page http://esw.w3.org/topic/FindingResourceDescriptions 
>  that I referred you to earlier, and include alternative headers  
> (other than Link:), alternative HTTP methods (other than GET and  
> HEAD), and alternative protocols (other than HTTP). I have just  
> reorganized the page a bit to make these choices easier to see. I  
> have not prepared a summary of the whole area. I (agreeing with at  
> least one other TAG member) thought it would be a better investment  
> to spend time on use cases, so that we can figure out what problems  
> we're trying to solve and establish criteria for deciding which  
> proposal solves them best.
>
> Jonathan
>
> On Mar 19, 2008, at 8:12 PM, Mark Nottingham wrote:
>
>> What are the alternate solutions?
>>
>> On 20/03/2008, at 12:15 AM, Jonathan Rees wrote:
>>
>>> Thanks Mark. This looks quite good, and the use of URIrefs for  
>>> relationships, rather than tokens from an ad hoc vocabulary,  
>>> addresses one of the concerns that came up at a recent TAG meeting.
>>>
>>> What this needs (as far as my effort is concerned) is arguments  
>>> against alternative solutions. Use cases could go a long way in  
>>> providing these. I expect these have already been part of your  
>>> discussion. Pointers to the relevant discussion on ietf-http-wg  
>>> would be helpful, as I'm having difficulty locating it.
>>>
>>> Jonathan


--
Mark Nottingham     http://www.mnot.net/
Received on Thursday, 20 March 2008 21:39:26 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 26 April 2012 12:47:53 GMT