Re: [link-relations] NEW RELATION - canonical

[Peter Saint-Andre <[e-mail]>]

On 6/3/11 3:27 PM, Frank Ellermann wrote:

> So far for the sponsored draft, I'm certainly not hot
> about it.  But it can take years until an RFC is ready.

It certainly shouldn't in this case!

> So why not register "canonical" as is *now*, and if an
> RFC later offers a better specification later it can
> simply update the registration?
>
> [RFC 5988 section 6.2.1]
> | relation types can be registered by third parties,
> | if the Designated Expert determines that an
> | unregistered relation type is widely deployed and
> | not likely to be registered in a timely manner.
> 
> It is "widely deployed", so what does "timely" mean?
> Clearly search engines will continue to do their job
> with or without registration of "canonical", with or
> without an RFC.
> 
> Only folks looking into the registry for "canonical"
> and finding nothing (at the moment) are a potential
> problem, if they also miss...
> 
> http://lmgtfy.com/?q=link+rel+canonical&l=1
> 
> If they invent a completely new successful relation
> with name "canonical" RFC it would be seriously bad.
> 
> I'm not really worried about this, but I don't like
> incomplete or outdated IANA registries.

The Designated Experts are:

Mark Nottingham, Eran Hammer-Lehav, Julian Reschke

Julian posted earlier on this list that it would be nice to have a spec,
but I leave it up to the DEs whether a link to a blog post is
acceptable, i.e.:

http://googlewebmastercentral.blogspot.com/2009/02/specify-your-canonical.html

Peter

-- 
Peter Saint-Andre
https://stpeter.im/