Re: [link-relations] NEW RELATION REQUEST: Relation name: "next"

[Leif Halvard Silli <[e-mail]>]

Mark Nottingham, Thu, 2 Sep 2010 21:34:41 +1000:
> Hi Mike,
> 
> Just to start discussion, I'll give initial thoughts in one e-mail, for now.
> 
> One general observation - relations that are potentially usable in 
> formats other than HTML should be defined in such a way that they're 
> not HTML specific. The only reason I mention this is that the 
> specification links go directly into HTML-specific text, without any 
> definition of semantics, etc.
> 
> E.g., with 'alternate', the link goes to:
> 
>> 4.12.3.1 Link type "alternate"
>> 
>> The alternate keyword may be used with link, a, and area elements.
>> 
>> The meaning of this keyword depends on the values of the other attributes.
>> 
>> If the element is a link element and the rel attribute also contains 
>> the keyword stylesheet
> [...]
> 
> It would be helpful (especially since alternate is already used by 
> Atom) if this were rewritten to something like:
> 
>> 4.12.3.1 Link type "alternate"
>> 
>> The 'alternate' link type gives alternate representations of the 
>> current document.
>> 
>> In HTML documents, the alternate keyword may be used with link, a, 
>> and area elements.
>> 
>> The meaning of this keyword depends on the values of the other attributes.
>> 
>> If the element is a link element and the rel attribute also contains 
>> the keyword stylesheet
> [...]
> 
> See what I mean?
> 
> 
> Beyond that, I see three groups of relations here;
> 
> 1) Updating HTML4 registrations, including:
>   next / alternate / bookmark / index / prev / stylesheet / help
> 
> There shouldn't be any issues with these. The only thing I'd note is 
> that 'alternate' is also used by Atom, but it's already registered by 
> HTML4, so it shouldn't be a problem. 
> 
> 2) Registering new relations, including:
>   archives / icon / noreferer / tag / pingback / prefetch / sidebar / 
> external / author / nofollow / search 
> 
> 'pingback' and 'search' refer to external specifications. We'll need 
> to talk through the implications of that.
> 
> 3) Updating registrations from other sources, including:
>   license / up / last / first
> 
> These might need a bit more discussion, especially license, which is 
> currently defined by RFC4946. Is there any reason that its definition 
> isn't adequate for HTML5?
> 
> The others need to be coordinated with the individuals who originally 
> registered the relations -- Noah Slater for up, and myself for first 
> and last. I don't think there will be any problem for mine, at least.

There is bug and tracker request w.r.t. to 'first' and some other 
relations in the HTMLwg. Thus I think that Mike was to early in sending 
registration requests for some of these link relations.

http://www.w3.org/Bugs/Public/show_bug.cgi?id=7475
-- 
leif halvard silli