W3C home > Mailing lists > Public > public-html@w3.org > May 2010

Re: HTML5 Profiles - First Editors Draft

From: Manu Sporny <msporny@digitalbazaar.com>
Date: Mon, 24 May 2010 22:11:30 -0400
Message-ID: <4BFB31D2.9030501@digitalbazaar.com>
To: HTMLWG WG <public-html@w3.org>
On 05/23/2010 02:33 PM, Toby Inkster wrote:
> A few thoughts on the current draft:
> * Currently @profile is only allowed on <head>, so there is a
>   convention that profiles found on that element are allowed to
>   apply to the entire document. Microformats currently use this,
>   with profiles found on <head> applying to microformats which are
>   pretty consistently used in <body>. GRDDL does too.

Keep in mind that the spec is very careful not to state that there MUST
be scope semantics associated with @profile. In fact, I tried to make
sure that the spec didn't say anything about how wide or narrow a scope
was intended to be - that decision is left up to the processor.

We /could/ state that a structured data processor like Microformats or
GRDDL MAY extend the scope of a profile by storing it in a processing
context. Thus, any @profile specified on the <head> element MAY also
apply to the body of the document if the structured data processor
enables this feature.

> * The relationship to GRDDL needs to be clarified; should GRDDL
>   processors continue to only look for profiles on <head> and ignore
>   others? If profiles are found on other elements, should only the
>   subtree be passed through the GRDDL transformation?

I don't think we should update how GRDDL operates using this
specification. GRDDL continues to operate as-is. Is there any reason why
we need to change GRDDL's processing rules based on this spec. I hope
the answer is no, because I tried to make sure that we wouldn't have to
do that.

> * "When processing attribute values, each URI must be processed from
>   left to right" - firstly this is ambiguous, does it merely define
>   a chronological order that processors must use (in which case, why
>   rule out parallel processing?) or does it mean that profiles to the
>   right over-rule profiles to the left in the case of conflicts? If the
>   latter, this is consistent with the current RDFa Core 1.1 draft, but
>   at odds with the XMDP specification which says that profiles to the
>   left win in the case of conflicts. There is IIRC an open issue on the
>   RDFa WG tracker about this.

It's the latter. That's a good summary of the situation. I don't quite
understand the reasoning behind XMDP where the first profile overrules
all other profiles in the attribute. Anyone on here know the reasoning
behind that decision? It seems backwards.

-- manu

Manu Sporny (skype: msporny, twitter: manusporny)
President/CEO - Digital Bazaar, Inc.
blog: Bitmunk 3.2.2 - Good Relations and Ditching Apache+PHP
Received on Tuesday, 25 May 2010 02:11:59 UTC

This archive was generated by hypermail 2.3.1 : Thursday, 29 October 2015 10:16:02 UTC