First off, thanks for the reply, Ben.
On Jun 9, 2005, at 12:25 PM, Ben Curtis wrote:
Several times lately, while reviewing various parts of CSS, I have
found myself saying "why does it work like this?" There are
several things that seem overly complex, or things that seem
obvious to me but aren't there. I'm new at this, and the guys at
the W3C are very smart, so I'm sure there's a reason--I would just
like to know what it is. Anyone have any thoughts?
You can sign up for the W3C's style list and search the archives or
ask them there; they are busy, so be respectful and search the
archives first. Start here: http://www.w3.org/Style/CSS/
I should have said that I did look around the W3C site and try to
find where to sign up for their lists. The page that I got sent to
after searching their site consisted mostly of broken links
(including the "Contact the W3C" link, so I couldn't even submit the
issue), which is why I was asking here.
Since many people on this list are learning CSS, people often
express their wish-list. But this list's mission is practical
application of CSS, and so I advise you (and others) interested in
the theory and development of CSS as a language to go to the W3C
site. As the CSS3 modules solidify and get implemented, you will
surely hear about using them here first.
Can do. Thanks for the pointers.
1) For absolute positioning, wouldn't it make sense to allow the
author to specify the containing element/(positioning context)
directly in the CSS? Something like this:
/* Put the navbar near the top left corner, keep it fixed as
you scroll */
#navbar { position: 1em auto auto 1em viewport fixed; }
/* Put a dropshadow on the image (actually, it would be easier
to have an actual CSS keyword for dropshadows, but you get the
idea */
#dropshadow { position: 5px 5px auto auto #image; }
Say you created a page and laid it out with your first example.
This page gets archived, and 10 year from now re-displayed as a
portion of a larger XHTML2 page. Your viewport position would not
display the way you intended. Or someone makes an anonymizer that
nests pages. Or a cellphone rearranges your page by wrapping it in
a container to help people navigate. All of these are possible
because XML is nestable without changing your document's meaning.
But your style wouldn't accept that. Since CSS is being developed
to apply to XHTML, rules must work for all of XHTML's abilities.
A good point; it was a poor example. I still think the basic idea is
valid, though--that is, straightforward author control over the
positioning context. But, as you say, I'll take it off list.
Your second example is easier: where does it get positioned in
documents that do not have an element with an id of "image"?
*blink* This leaves me a bit flummoxed...if someone removes elements
from my page--that is, actually removes them from the content stream,
not just fails to display them--then of course the layout will not
look like I intended. Or am I not understanding you properly?
(For the particular use of a drop-shadow, see the CSS3 proposals
for drop shadows[1], or multiple-image backgrounds[2].) The
solution to this type of need is proposed in CSS3, such that you
can move content from one area of a page to another, essentially
changing the structure of the page. See the generated and replaced
content module.[3]
Thank you, I will.
2) A way to express units using mixed-unit expressions.
This occasionally gets discussed on the W3 style list. Join up.
Simple calculations of mixed units are being toyed with, but I do
not think there is a concrete proposal. More complex expressions
often get suggested, but as this strays toward scripting languages
it is often disregarded.
I figured I couldn't *possibly* have been the first one to think of
this.
3) Some way to do columns straightforwardly:
A current working draft.[4]
4. http://www.w3.org/TR/2001/WD-css3-multicol-20010118/
Ah, cool. Obviously I didn't look deeply enough. Thanks for the link.
Again, thank you...you would have been justified in saying "Off topic
for the list, N00B!", so I appreciate your informative answers.
--Dks
______________________________________________________________________
css-discuss [EMAIL PROTECTED]
http://www.css-discuss.org/mailman/listinfo/css-d
List wiki/FAQ -- http://css-discuss.incutio.com/
Supported by evolt.org -- http://www.evolt.org/help_support_evolt/