W3C home > Mailing lists > Public > public-html@w3.org > February 2009

Re: google supports rel="canonical" (ISSUE-27 rel-ownership)

From: Leif Halvard Silli <lhs@malform.no>
Date: Wed, 18 Feb 2009 06:21:36 +0100
Message-ID: <499B9AE0.5030702@malform.no>
To: Sam Ruby <rubys@intertwingly.net>
CC: Dan Connolly <connolly@w3.org>, "public-html@w3.org WG" <public-html@w3.org>
Sam Ruby 2009-02-18 01.54:
> Dan Connolly wrote:

>> <link rel="canonical"
>> href="http://www.example.com/product.php?item=swedish-fish" />
>>  --
>> http://googlewebmastercentral.blogspot.com/2009/02/specify-your-canonical.html 

I was discussing off-list with Rob 2 weeks ago whether the 
opposite of rel="alternate", namely rel="original" could be 
useful. Perhaps they snitched our talk ...

Or perhaps not. rel="original" is hereby proposed. (Something can 
be original in relation to the alternate, even if something else 
is the canonical version.)

>> seems relevant to one of our issues...
>> ISSUE-27 rel-ownership @rel value ownership, registry consideration
>> http://www.w3.org/html/wg/tracker/issues/27
>> I see Anne found this a few days ago...
>> http://annevankesteren.nl/2009/02/rel-canonical

Anne is there recommending adding it to the WHATwg wiki. Such a 
tracker is a great idea. But where is the official W3 rel value 
tracker? Or more accurately: where is the profile tracker?

> Unfortunately, the conclusions I draw from this aren't very encouraging. 
>  We've discussed at length how making certain attributes mandatory often 
> does not have the desired effect on authoring behavior, at least in the 
> domain of HTML.  Alas, it appears that the situation is pretty much the 
> same for enumerated lists of attribute values, and on attribute values 
> in general.  Google, Microsoft, and Palm at least are going to do what 
> they are going to do:
> http://www.w3.org/QA/2009/02/palm_webos_approach_to_html_ex.html

We can at least conclude that HTML from the start lacked a 
specified way for registering extensions.

For example, HTML 4 did not offer a wiki to track rel values, 
simply because it did not consider (?) that single vendors would 
propose one rel value here, and another value there.

But HTML 4 *could* have offered a "profile tracker", where authors 
and vendors could register their profiles. And W3 could have had a 
working group that considered extensions to the default HTML 4 

In other words: even if @profile is kept, the reason for doing so 
should not be in order to not track (and incorporate) distributed 
leif halvard silli
Received on Wednesday, 18 February 2009 05:22:18 UTC

This archive was generated by hypermail 2.3.1 : Thursday, 29 October 2015 10:15:42 UTC