RE: [xri] Version Control Commit by blade

[Eran Hammer-Lahav <eran@hueniverse.com>]

There is a base URI but you should not actually use it in a Rel value. We are saying the same thing. Whatever we say must not create the impression that http://www.iana.org/assignments/relation/describedby is a valid value. It's not. However, clients can normalize the short strings internally if they want to store all relations as absolute URIs.

EHL

> -----Original Message-----
> From: Scott Cantor [mailto:cantor.2@osu.edu]
> Sent: Thursday, August 20, 2009 12:46 PM
> To: Eran Hammer-Lahav; 'Will Norris'; 'XRI TC'
> Subject: RE: [xri] Version Control Commit by blade
> 
> Eran Hammer-Lahav wrote on 2009-08-20:
> > Keep in mind that draft is only going to work if it accommodates
> reality.
> It
> > is too late to move to a URI only relation types world.
> 
> It's always too late because people keep making bad decisions and
> nobody
> stops them. But with respect to this spec, I don't understand your
> statement
> that there's no "base URI". Reading that draft implies there is one to
> me.
> If not, we should say "MUST be a URI or a string registered by IANA at
> <registry>".
> 
> -- Scott
>