W3C home > Mailing lists > Public > public-rdf-in-xhtml-tf@w3.org > October 2005

Re: Telecon Agenda - Tuesday, October 4th, 2005 - 1400 UTC

From: Mark Birbeck <mark.birbeck@x-port.net>
Date: Tue, 4 Oct 2005 12:26:07 +0100 (GMT Daylight Time)
Message-ID: <.192.168.1.104.1128425167.squirrel@localwebmail.x-port.net>
To: "Steven Pemberton" <steven.pemberton@cwi.nl>
Cc: "Ben Adida" <ben@mit.edu>, "public-rdf-in-xhtml task force" <public-rdf-in-xhtml-tf@w3.org>

Hi Steven,

On Tue, October 4, 2005 11:50 am, Steven Pemberton wrote:
>
> On Mon, 03 Oct 2005 05:18:04 +0200, Ben Adida <ben@mit.edu> wrote:
>
>> 2) CURI and backwards compatibility:
>>
>> This is an issue we haven't fully discussed. We should discuss it. CURIs
>> may present backwards-compatibility issue with respect to REL="next" and
>> the like from previous versions of HTML. We need to be conscious of
>> these issues and to work through them.
>
> Anything we do here will not work on old software.
>
> 	<link rel="next" href="http://www.w3.org/MarkUp/Group/2005/thingy"/>
>
> works just as well in the old and new versions, but the purely new stuff:
>
> 	<link rel="next" href=":htmlwg:thingy"/>
>
> will only work on new software.

This is true, but the issue Ben was raising concerns @rel. In the current
set-up we have QNames in @rel, and the meaning of rel="next" has been
clearly defined. But if we move to CURIEs, we will have one syntax in @rel
and another in @href. Ben is flagging up the fact that if we opt for
consistency across attributes, like this:

  <link rel="[next]" href="[htmlwg:thingy]"/>

then we have lost backwards compatibility in @rel.

Note blatant use of favoured syntax :)


> At the Hypertext coordination meeting last week, they were asking how we
> can go about making CURIs more generally usable across W3C. They agreed to
> talk about it more at the next call.
> Minuted at the first section of http://www.w3.org/2005/09/30-hcg-minutes

Great. We're using this in Sidewinder to provide for configurable links,
and it also makes for a novel way of doing Wikis, which we're still
working on.

By the way, I changed the name to CURIEs for the IPTC, since people were
searching on Google to find out more about CURIs, and then getting
confused when they came across "Canonical URIs". I'd always intended for
it to be pronounced 'Curie' as a little reminder of the Curie family! So
if no-one has an objection, then now it's even spelt right.


> By the way we still have to agree on a format for a CURI, either:
>
> 	href=":htmlwg:thingy"
> or
> 	href="[htmlwg:thingy]"
>
> and for bnodes:
>
> 	href=":_:a"
> or
> 	href="[_:a]"
>
> I like the colon version, since it suggests a missing scheme...

I have to say, that's why I *don't* like it ;) I think it could be
confusing. I think it's important to somehow or other stress that this is
a pre-processing layer, prior to the URI processing. Also, I suggested '['
and ']' because they are not valid URI characters; I know that ':' is not
a valid initial character for URIs, but it still might confuse.


>... but I can live with either.

I strongly favour '[' and ']'!

Regards,

Mark

Mark Birbeck
CEO
x-port.net Ltd.

e: mark.birbeck@x-port.net
w: http://www.formsPlayer.com/
b: http://internet-apps.blogspot.com/
Received on Tuesday, 4 October 2005 11:26:17 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 14:15:00 GMT