W3C home > Mailing lists > Public > www-style@w3.org > January 2003

Re: XBL is (mostly) W3C redundant, and CSS is wrong W3C layer for semantic behavior *markup*

From: Ian Hickson <ian@hixie.ch>
Date: Sun, 5 Jan 2003 03:08:08 +0000 (GMT)
To: Shelby Moore <shelby@coolpage.com>
Cc: "www-style@w3.org" <www-style@w3.org>
Message-ID: <Pine.LNX.4.21.0301050227450.4908-100000@dhalsim.dreamhost.com>

On Sat, 4 Jan 2003, Shelby Moore wrote:
> 
> They are always explicit.  So if specifications wanted to
> "_completely_ control semantics" then they would say that.

Here is a specification that changes (well, extends, really) the semantics
of the HTML spec:

   http://www.hixie.ch/specs/pingback/pingback

Even if the HTML spec said "no other spec may change this spec's
rules", the pingback spec above would still be extending HTML (albeit
maybe with slightly more controversy).

As I said:

>> Even if HTML stated that it wasn't the final word on the semantics,
>> that still wouldn't explain how XBL could change them.
>> 
>> Just because _something_ can change the semantics, doesn't mean
>> _everything_ can.
> 
> If _one_ thing can change the semantics of a specification, then it means
> "specification does NOT _completely_ control semantics".

That is not the point I am contesting. The point I am contesting is that
XBL can change the semantics.

Analogy: Just because human beings are able to understand Quantum
Mechanics, does not mean that a particular human understands QM.


>> There are multiple fundamental disagreements, those that I mentioned in
>> that post are unrelated to semantics.
> 
> The issue of whether specification _completely_ controls semantics is the
> most important fundamental question to resolve first.

Well let us resolve that straight away then: No, it cannot.

The real issue (per the subject line) is whether XBL is "W3C redundant"
and why CSS is the wrong layer for "semantic behaviours".

My argument, which has stayed constant throughout this discussion, is that
XBL is not redundant because no other spec can do the same thing, and that
CSS is indeed the wrong layer for semantic behaviours, but that XBL is not
changing the semantics.


>> But the implementation has nothing to do with its semantics; it's just
>> a presentational aspect.
> 
> If the implementation violates the HTML 4.01 spec, then per my Axiomatic
> Proof [1], it has changed the semantics.
>
> [1] http://lists.w3.org/Archives/Public/www-style/2003Jan/0073.html

Your proof has several glaring errors, including:

| Specification defines the meaning of HTML, which within the constraints
| of that meaning, defines what implementation _should_ do. 

Untrue: Meaning is unrelated to what implementations should do.

| Thus any deviations which are not allowed by the specification, would be
| changing the semantics as defined by the specification.

Untrue: Example being an implementation that failed to correctly parse
the SHORTTAG NET feature. Like, for example, _every_ single Web browser
currently being used. Note how the semantics of HTML documents have not
changed.

Another example: Even if there were _no_ implementations, HTML documents
would still have exactly the same semantics.

-- 
Ian Hickson                                      )\._.,--....,'``.    fL
"meow"                                          /,   _.. \   _\  ;`._ ,.
http://index.hixie.ch/                         `._.-(,_..'--(,_..'`-.;.'
Received on Saturday, 4 January 2003 22:08:10 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:54:19 GMT