charter and votes to publish documents

My objection to publishing the microdata document was
contained in my message of 1/8/2010 2:17 PM:

> I'm opposed to publishing microdata as a FPWD of W3C HTML
> WG because it is not in the charter, or reasonably linked
> to any deliverable the working group is committed to, and
> there are many other much higher priority tasks and
> documents that the working group should put its attention
> to.  I support publication of HTML 5 without microdata
> because it actually brings the working group closer to its
> goals.

That is, my objection was based on the conjunction of three
assertions:

1) It is not in the charter
2) it is not reasonably linked to any deliverable the
   working group is committed to
3) There are many other much higher priority tasks and
   documents that the working group should put its attention
   to.

I will note that working groups in W3C do often publish
work outside their charter. "Not in the charter" by itself
does not rule out publication. (On the other hand, it should 
be a factor, because it may subject a working group's product
to legitimate criticism from other chartered groups or from
non-working group participants. The charter normally defines 
the scope for determining whether people join a working
group, and charter overlap is of concern to the W3C AC
when doing charter review.)

The three factors combined, though, are sufficient grounds
for me to believe that the document should not, at this
time, be a working group priority. I don't rule out it
later being published, or later asking for a charter
extension, or asking that another working group take
the work on, or asking that another working group be formed,
or submitting it as a member contribution, or publishing 
it outside the W3C entirely.

I don't find convincing an argument interpreting "a
mechanism to do P things such as X, Y, or Z" as covering "a
method of doing a P thing W, where W does something similar
to Z" [[see postscript]] -- and so I have not my opinion (1) 
that microdata is not in the charter. 

I don't find the argument "microdata does something similar
to RDFa" a convincing reason to change my opinion (2) that
microdata is not reasonably linked to any deliverable
the working group is committed to.

I have also raised a concern about the focus of the 
current HTML+RDFa document, as it does not, in my opinion,
meet the condition of actually being called for "in the
charter", although perhaps it is arguably closer.  But it
also seems like the HTML+RDFa could, with some amount of
work, be generalized so that he mechanism it proposes for
including RDFa could be generalized to apply to other
vocabularies. That is what I was asking about.
	
About the charter itself, you write:

> It is acceptable for us to use any number of fundamental
> extension mechanisms to provide for mixing in additional
> vocabularies, if that seems like the technically sound
> approach. We should not be struggling to shoehorn every
> kind of extension vocabulary into a single mechanism.

I have no problem with that; but the documents in question
are not focused on the extensibility mechanisms, they focus
on particular extensions. I believe the charter was written
in a way specifically to allow distributed extensibility,
because of the word "independently" in the phrase
"independently developed vocabularies". And we seem to be
going down the road of adding new mechanisms for each new
extension vocabulary, and adding the vocabularies themselves,
not just the extensibility mechanisms.

> We should be interested in allowing for all sorts of
> vocabularies, not just the three examples.

I agree completely.

> We should not thrown by the fact that an extension can add
> further extension mechanisms.

I didn't think this was a factor I raised.

> Julian, Larry, and others who have charter concerns, if
> you are not persuaded by the above alone, we can take
> steps to get a more official interpretation.

I agree with the statements you've made about the charter,
so I'm not sure how a "more official interpretation" would
help.

What I don't see is how you are applying them to the reason
I've given for opposing the publication of the current
microdata document, and my concerns about the HTML+RDFa
document.  I understand how the working group might need to
resort to multiple mechanisms to adding different kinds of
independently developed vocabularies. I think *that*
discussion would be worth having, and a document
specifically addressing extensibility mechanisms would be
worth publishing -- that would definitely be in the charter,
highly relevant to our deliverables, and help reach closure
on the focus of work in the W3C HTML WG.

A focusing on getting consensus on the RDFa vs microdata
debate itself, though is, in my opinion, a distraction which
is out of scope.

*Independently* developed vocabularies should be developed
independently (i.e., not here), rather than the HTML WG 
adding them one at a time, much less with a different
extensibility mechanism for each.

Perhaps they might be "part of" the Web Hypertext
Application Technology platform, but I think if we're
going to prioritize work in W3C HTML WG, focusing on
HTML itself  and its general extensibility mechanisms 
must be the focus.

> Ruby is a set of elements that are intended to be used
> directly in the HTML namespace - they do not have a
> namespace of their own. They also need to plug into the
> content model in a very specific way.

That has been the traditional way in which HTML has been
extended to add the Ruby vocabulary.

> ITS is a vocabulary in custom XML namespace.

That has been the traditional way in which HTML has been
extended to add the ITS vocabulary.

> RDFa is a set of non-namespaced attributes that can go on
> any element in any namespace, plus reliance on at least
> the surface syntax of xml: namespace prefix declarations

That is the proposed way of adding RDFa.

> It's worth noting especially that we have some
> pre-existing extensibility mechanisms, including XML
> namespaces in the XML serialization only, class, rel and
> <meta>.

How many extensibility mechanisms does HTML need? We are
encouraged to develop *a* mechanism, but it seems like we
are going down a path of adding new mechanisms for almost
every new vocabulary. Along the way, some other traditional
extensibility mechanisms have been rejected (specifically,
DOCTYPE-based version-specific behavior).

I think this path is counterproductive to reaching the
chartered goals of the HTML working group.
One-extension-mechanism-per-extension does not, to me,
satisfy what we were asked to develop.

You issued a "call for consensus" asking for opinions about
publishing a document. I replied with an opinion.

I don't think it should require any extraordinary process
for someone to disagree with that publication and to give
reasons for the disagreement.

Larry
--
http://larry.masinter.net


[*] I reviewed several source books on English
grammar teaching English as a Second or Foreign
Language. 

For example, the use of the determiner "a" to 
indicate singular (as opposed to "some" or
"the") was covered on page 13 in the lengthy
"Systems in English Grammar, an Introduction
for Language Teachers" isbn:013156837.

However, the meaning is subtle. "The Grammar
Book, an ESL/EFL Teacher's Course" isbn:0838447252
chapter 15 on "Articles" takes several pages
to explain this.

The second phrase with some difficulty is
the use of "such as" used to introduce examples.
"Practical English Usage" isbn:019431197,
page 544 notes that "such as" with  a noun,
is used to introduce examples, and itself
gives two examples:
	
a) "My doctor told me to avoid fatty foods
   such as bacon or hamburgers."
b) "In such areas as North Wales or the Lake
   District, there are now too many walkers 
   or climbers."

Now, (a) could reasonably be understood that
"bacon" and "hamburgers" are only examples
of fatty foods, and that your doctor would
like you also to avoid other fatty foods such
as cheese and lard. And (b) should certainly
not be interpreted that you might not also think
there are too many walkers in Mallorca.

But the use of examples indicates that the
examples are meaningful. I don't think you 
would be following your doctor's advice if
you avoided lard but continued to eat 
hamburger and bacon, or that anyone should
think that because Mallorca is "similar"
to "North Wales" by some measure of similarity,
that the speaker doesn't actually have any
opinion at all about walkers or climbers
in North Wales.

And similarly, I don't think HTML WG is
meeting its charter if we address microdata
but don't address Ruby, ITS, RDFa, SVG,
MathML with a coherent extensibility policy.

-- lmm

Received on Wednesday, 13 January 2010 17:57:38 UTC