Re: After 5

On 17/09/2014 19:54, Jukka K. Korpela wrote:

> Not disruptive at all, but do you really have modules?

(no worries, I remind you the unit of trouble-making at W3C
 is the "glazou" :-) )

Yes we do. Take CSS 2.1 and you'll see that 2.1 is divided in
existing more recent modules. Gather them and they represent
the current state of CSS.

> In traditional programming, a “module” is a unit with well-defined
> relations to other units. CSS “modules” are just parts or aspects of
> CSS. They may overlap, they may be ignorant of each other, and they may
> conflict with each other.

<pun>
  « CSS is not a programming language »
</pun>

> In consequence, if you ask what is CSS, e.g. as approved by the W3C CSS
> Validator, the answer is something like “oh well, CSS 2.1 plus those
> newer “CSS 3” or “CSS 4” specs that have REC or CR status, well I’m not
> sure about LC, and of course some parts of specs with lower maturity
> are, you see, CSS”.
> 
> Maybe this is the best we can have, and maybe it is the way to go. But
> speaking of “modules”, I think you should define how they are defined
> and how their conflicts are to be resolved.

That's also correct. The name "modules" comes from the original
modularization CSS 2 we did ages ago. The rationale behind the decision
is easy to explain: CSS level 1 had 12 pages, level 2 had 220 pages,
level 3 introduced so many new features that it's much more than that.
One single spec for everything was not feasible. For HTML, I still
remember when Sylvain Galineau attended a meeting with a print of the
html spec. 800 pages I think, ahem.

As I said in one article (interviewed by Molly Holzschlag IIRC), one
side-effect of modularization is specialization. During the CSS 2 era,
almost all WG members were contributing to all aspects of the document.
When you modularize and drastically increase the number of documents,
the creation of informal subgroups is almost inevitable. Some stuff
still gather everyone when we discuss them but some others don't. I
am not saying this is bad, please don't misunderstand me, I am only
saying this is how it goes.

I would like to add something about modularization we rarely think of,
being spec authors: a collection of lightweight specs is better for
Web authors than a single huge document. And I suppose implementors
don't really care as soon as the specs are well written.

</Daniel>

Received on Wednesday, 17 September 2014 18:56:31 UTC