W3C home > Mailing lists > Public > public-html@w3.org > August 2008

Re: Deciding in public (Was: SVGWG SVG-in-HTML proposal)

From: Julian Reschke <julian.reschke@gmx.de>
Date: Fri, 01 Aug 2008 21:09:30 +0200
Message-ID: <48935F6A.3020404@gmx.de>
To: Edward O'Connor <hober0@gmail.com>
CC: HTML WG <public-html@w3.org>

Edward O'Connor wrote:
>>>> Could you clarify what extension points you're talking about? Class?
>>> Offhand, @class, @rel, @id, <div>, <span>, <meta>, @data-* all come
>>> to mind. I suppose @role counts, though why you'd use @role instead
>>> of @class eludes me.
>> To put in RDF statements?
> 
> No, to include a trackback link in an HTML document. All you need is
> @rel for this case.

Aha.

I agree that embedding a single link is a relatively simple use case.

> ...
>> div/span: don't see where there's extensibility in them.
> 
> Lacking semantics of their own, <div> and <span> are handy to when other
> elements' semantics would conflict with whichever semantics you're
> adding to things. For instance, prior to the introduction of <article>,
> authors could have used <div> (as in <div class="article">) for marking
> up articles.

Just clarifying: so these are not extensibility points on their own, 
they need @class.

>> That being said, it [@data-*] has the same problem as most of the
>> other things you mentioned, the lack of a disambiguation mechanism.
> 
> So far as I've seen, the lack of a disambiguation mechanism isn't a
> problem in practice. <div class="vcard"> means what it obviously means,
> regardless of the presence or contents of head@profile.

There is a practical problem. It doesn't scale (people have tried to 
register microformats and have been sent anyway). It also doesn't nest 
(remember the discussion about titles, or was it names?).

BR, Julian
Received on Friday, 1 August 2008 19:10:14 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 9 May 2012 00:16:21 GMT