Link relation types and validity in <link>, was: ACTION-182 and Issue-27

On 30.07.2010 11:42, Julian Reschke wrote:
> On 29.07.2010 18:21, Maciej Stachowiak wrote:
>> ...
>> I looked at the list
>> archives:<http://www.ietf.org/mail-archive/web/link-relations/current/maillist.html>
>>
>>
>> I see at least one registration request made by an HTML WG member. Are
>> you giving your judgment that it is not a serious request?
>> ...
>
> As a matter of fact, yes; I would expect the person doing the request to
> actually supply the information required by the registration process,
> and also to reply to the requests for clarification.
>
> Furthermore, we do have a request for new link relation application data
> (<http://tools.ietf.org/html/draft-nottingham-http-link-header-10#section-6.3>),
> not an actual link relation
> (<http://tools.ietf.org/html/draft-nottingham-http-link-header-10#section-6.2>).
>
>
> The intent of my mail was to encourage people to finally send
> registrations, so that we *finally* make process with ISSUE-27. I think
> setting a deadline would help with this.
> ...

Following up to this: another week has gone with zero progress.

I'd like to dive into one specific question related to this: should the 
spec make the use of certain (registered) link relations invalid inside 
the <link> element? (that's part of the metadata Ian did attempt to add 
to the IANA registry).

Right now, HTML5 makes the following link relations invalid in <link>:

- bookmark
- external
- nofollow
- noreferrer

(see <http://dev.w3.org/html5/spec/links.html#linkTypes>).

1) The definition of "bookmark" is:

"Gives the permalink for the nearest ancestor section."

"bookmark" actually is defined in HTML4 slightly differently:

"Refers to a bookmark. A bookmark is a link to a key entry point within 
an extended document. The title attribute may be used, for example, to 
label the bookmark. Note that several bookmarks may be defined in each 
document." -- <http://www.w3.org/TR/html4/types.html#type-links>

Ignoring why this was changed for a second -- bookmark could easily 
defined for <link> as well, by changing the definition to:

"Gives the permalink for the nearest ancestor section, or the whole 
document otherwise."

So I really don't see why we would want to forbid it.

2) The definition of "external" is:

"Indicates that the referenced document is not part of the same site as 
the current document."

First of all, this doesn't sound terrible useful. Where does it come from?

But, assuming it *is* useful, why wouldn't it apply to <link>?

3) The definition of "nofollow" is:

"Indicates that the current document's original author or publisher does 
not endorse the referenced document."

Great. Why is this invalid on <link>? (see also 
<http://www.w3.org/Bugs/Public/show_bug.cgi?id=10172>)

4) The definition of "noreferrer" is:

"Requires that the user agent not send an HTTP Referer (sic) header if 
the user follows the hyperlink."

Great. Why is this invalid on <link>? (see also 
<http://www.w3.org/Bugs/Public/show_bug.cgi?id=10172>)


Summary: I see no evidence why disallowing *certain* link relation types 
on <link> make sense. Can we please discuss this, and then look at the 
related question about certain link relations being forbidden on a/@rel 
next?

Best regards, Julian

Received on Thursday, 5 August 2010 12:25:48 UTC