Re: CSS aims

On 1/23/14 1:44 PM, Brian Kardell wrote:
> Having read what you said, I'm not sure which part of what I wrote in
> any of my posts you actually disagree with.

Perhaps I can make this clearer.

> There is a purist notion that there is a very important difference and
> that presentation should be the realm of CSS.  Layouts are presentation
> - but CSS has been trying to solve this problem (how do I express layout
> in CSS (meaning specifically NOT using markup, but the language of CSS)
> when I need boxes and relationships that aren't in the DOM) since at
> least 1996.  I think this is a laudable goal, but I have not seen any
> proposal for such a thing that doesn't involve dubious amounts of
> complexity or hasn't been ignored.

Part of my point is that you are asking for the wrong thing when you ask 
for "boxes and relationships."  There is no reason that those have to be 
baked into the rendering tools.

If that's "purist", I don't care much.

> The reason that this comes up is that there are currently a number of
> proposals underway in the CSS working group which involve some form of
> this box generation/management - IE - new magic.  We haven't explained
> the old magic :)
> Alan (one of our group members from Adobe) has a proposal for Regions in
> CSS which attempts to expose a cross-cutting primitive to explain these.
>   I think it's reasonable to debate whether it is the right primitive,
> etc - but there have been a number of efforts to stymie it simply
> because it recognizes a relationship between DOM and CSS which has
> always been there (and I am arguing, probably wont go away any time
> soon).  In other words, instead of requiring you to use some brand-new
> language for defining boxes or slots in a layout, Regions allows you to
> just use HTML.  This is considered to some a high crime because it goes
> against the "core aims of CSS".

I tend to agree that it's a bad idea.  I spent enough time with XSL-FO 
to think that describing boxes in markup is a recipe for doom.  Even 
with the ameliorating promise of "XSLT will make the transition" - and 
it kind of did - these kinds of proposals attempt to solve problems by 
flattening layers that are actually useful.

I don't see anything in the regions proposal that is more appealing or 
less likely to create headaches.

> My purpose in asking these questions might seem academic, but there are
> actually fairly concrete reasons that this sort of existential crisis
> blocks abilities to uncover primitives or make good proposals for CSS.
> I think it's important that we get past the purist argument if possible
> for all of the many reasons stated in this thread.

You seem to be labeling something "purist" because decades of experience 
happen to be standing in your way.  CSS doesn't provide quite what you 
(not just Brian, a group of developers) want, so it's obviously a 
problem with CSS.  Mixing up the layers seems like a quicker way to get 
there.

If you get there, I don't think you (or a lot of the rest of us) are 
going to enjoy what you've created.

Good luck,
-- 
Simon St.Laurent
http://simonstl.com/

Received on Thursday, 23 January 2014 18:55:59 UTC