W3C home > Mailing lists > Public > www-style@w3.org > April 2000

Re: Just a dream about inheritance

From: Jan Roland Eriksson <jrexon@newsguy.com>
Date: Wed, 26 Apr 2000 01:54:21 +0200
To: Daniel.Glazman@polytechnique.org
Cc: www-style@w3.org
Message-ID: <dmbcgsoi6883ne97kgklc3r51tdt5n0mso@4ax.com>
On Tue, 25 Apr 2000 11:54:52 +0200, Daniel Glazman
<glazou_2000@yahoo.fr> wrote:

[Ok, i'l bite, careful I hope :) ]

>Looking back to the P language (the style language used by Grif,
>Symposia and Amaya (?)),

Panorama recognizes its own brand of styling too I think?
(if SGML browsers can have a play here?)

>I still regret one of its wonderful features.
>It was possible, in the P language, to set the value of a property
>in function of the value of another property assigned to the parent
>element, the next element, the previous element, and so on.
>For instance :
>
>     width = previous.width + 0.5cm

If I may make a stop right here and argue for a point that must be
resolved before we can go on?

CSS as originally defined is a pure "descriptive" language, meaning that
whatever "property:value" that is sent along with a document, or linked
to the same, must be met in the client by a preprogrammed function that
can accept that value and match it to one of its already available
property handling routines.

No match found then naturally should mean "ignore that value"

At some point we need to find a consensus on how to proceed here, should
CSS stay as the descriptive language it is (mostly) today or should it
be opened up to be able to do subroutine - procedure - function calls to
other parts of what might be available in a ua?

I'm colored here, let that be no secret, so in that position I find it
revolting that MS has already gone ahead and implemented function calls
for property value assignments.

There was a time back in late 96 when Netscape found that they had to
"bite a stone" when their own JSSS proposal was turned down in favor of
CSS1. (every NS4 release after that time has suffered from it)

At some close time now, W3 should once again see to it that MS, at this
time, gets to learn the same lesson.

What needs to be settled is if CSS 1/2/3 eventually stays in a stable
and usable incarnation for the HTML docs that will be produced for years
to come.

Next, if XML could be developed as a sort of "SGML light" naturally XSL
should be developed to become "DSSSL light".

Mind you all, xml _needs_ a universal Turing complete transformation
language if it will ever be usable.

HTML only needs a presentational suggestion language to be usable.

There's a difference...

>Miam ! Just an old dream... Don't flame too strong :-)

No flames, just a serious question I hope.

-- 
Jan Roland Eriksson <jrexon@newsguy.com>
<URL:http://member.newsguy.com/%7Ejrexon/>
Received on Tuesday, 25 April 2000 19:47:57 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:54:04 GMT