RE: Liaison with CSS WG to provide a mechanism for expressing the style of document semantics

Andrew -

That is an excellent question, and please forgive me if I'm not too familiar
with the appropriate channels for these things. Here is why, off hand, I
think that this might not be ready for them *quite* yet:

* I am still trying to understand the "legend" proposal (no one else seems
to care one way or the other); are we even sure we want to do it at all? If
not, why tie them up with it?
* Until we clarify fully what we are even considering, we should not bring
it to the CSS WG.

No matter what happens, this needs to be a full, joint venture between the
CSS group and this one. I saw what happened when ARIA was developed
separately from HTML (2 months of pointless & fruitless argument over the
"colon or hyphen" issue), and I saw what happened regarding @alt. Whatever
we think is best, we can't keep working in total isolation.

And yes, that seems to violate what I said above, but not entirely. What we
need to do is understand what we are talking about, but bring it to the CSS
WG before we make up our minds on anything. That is what is killing this
whole process. Different working groups put together a full spec outside of
purview (but interlocked with) another group's efforts, and then dump it on
some other group and say, "here's our decisions". Let's not do that here. So
let's figure out what exactly we are talking about, and then let's get the
CSS folks involved, since the very nature of this proposal definitely needs
both groups to have equal input.

J.Ja

-----Original Message-----
From: Andrew Sidwell [mailto:w3c@andrewsidwell.co.uk] 
Sent: Tuesday, June 03, 2008 8:52 PM
To: Justin James
Cc: 'Robert J Burns'; 'HTML Issue Tracking WG'
Subject: Re: Liaison with CSS WG to provide a mechanism for expressing the
style of document semantics

Justin James wrote:
> Robert -
> 
> I think I "get" it now. There are two items which are throwing me for a
> loop; the word "legend", and the way I *thought* you were using the
concept
> of semantics.
> 
> OK, here is my revised understanding:
> 
> "Legend" tries to provide a mechanism so that the semantic value of a tag
> can be derived from the CSS definitions applied to the tag (with the usual
> hierarchies, inheritance, etc.). It is similar in *concept* to
"aria-role",
> but instead of defining a function role at the element level, it specifies
a
> semantic purpose at the CSS definition level.

I hate to interrupt, but isn't this more productively taken to the CSS 
WG?  Bringing up a change to CSS as an issue here when there is no 
actual action this WG can take on it seems counterproductive.

A

Received on Wednesday, 4 June 2008 02:01:27 UTC