[Gllug] [OT]Reforming wayward web designers

[Garry Heaton]

Tethys wrote:

> Garry Heaton writes:

> At a guess, because sometimes stupid kludges are needed to work around
> problems until the world learns to stop using buggy browsers.

Like, in 2020?

> Although
> I haven't read his exact words, I'd be very surprised to see him
> advocating using a hybrid approach in the long term.
> 
> Tet

It's the near-term that bothers me. CSS1 and CSS-P have been around since
the mid-90s and browser manufacturers, notably Opera, have had several
unsuccessful stabs at nailing the beast. The fact that they've failed so
often leads me to question how difficult it is to render CSS properly. Sure
the CSS alphabet itself isn't complicated in itself but its usage can be,
especially when heavy use is made of inheritance.

CSS-P was something of an afterthought and I get the impression that it
wasn't really well thought-out. If the W3C had been serious about devising a
layout spec we would have had more than a handful of position, top, left,
width, height and z-index to play with. Grid layouts, for example, are very
useful and one reason many HTML monkeys still use <tables>s. Simulating that
with CSS-P is laborious and inefficient.

CSS-P also fails to contain resized content as well as <td>s. I've even
heard some CSS commentators advising leaving space to the right of text
links on vertical columns - just to be sure the whole thing doesn't break
when the user increases the text size. This kind of thing isn't a solution
to anything and I don't see the problem being resolved anytime soon because
it isn't strictly an implemenation problem. CSS-P is just a poor layout model.

Garry


-- 
Gllug mailing list  -  Gllug at gllug.org.uk
http://lists.gllug.org.uk/mailman/listinfo/gllug