...

View Full Version : W3C - thinking TOO far ahead?



whammy
03-02-2003, 04:37 AM
I would just point to this:

http://www.w3.org/TR/xhtml-forms-req

And see what you guys think... looks to me like they are getting way ahead of themselves, considering the fact that there isn't even any consistent cross-browser CSS support. Yeah, they have some great ideas for the future, but...

I think before they go this far, they should concentrate their efforts on fixing the problems with current implementations of XHTML, CSS, etc. so that developers can REALLY "use what works" without resorting to HTML 4.01 or whatever.

Have any of you tried to center a nested table using only CSS, and make it work cross-browser? (Not to mention avoiding using tables, and just using CSS!) It's a nightmare, and I think, impossible ... an example: I have been coding in pure XHTML for I don't know how long, but I found it impossible to center a nested table (and believe me, I know how to google!), and after a few hours of research, I gave up, took the XML tags, etc. out of my page (since I knew it would no longer validate), and just used table align="center". It worked in every browser! ;)

These guys seem really aware of their aims, but not the aims of the regular person that wants to just create a website, or the developer that actually works for a living.

And we help these people (the day-to-day web developers, that is) - all W3C seems to be doing (along with some really good ideas) is ignoring a basic rule:

K.I.S.S. (Keep It Simple, Stupid!)

;-)

I totally agree with their aim - but I think it's important to make sure that the NEW technology is up to speed with what's currently being implemented, or at least backwards compatible to a certain point. And in the case of CSS, as many of you Mozilla users know, this is lacking in many, many areas.

WA
03-02-2003, 08:51 PM
I'll move this to the General Building category, since a lot of people may be interested in adding their input to this topic...

jkd
03-02-2003, 09:44 PM
You are wrong. W3C has nothing to do with implementation (it may suggest algorithms and whatnot), the implementation is what the vendor's are concerned with. And vendors are keeping up, except for Microsoft.

And there is consistent crossbrowser support of CSS. Not in IE, but in every single other significant browser.

New technology has to be described before it can be implemented, and this is what W3C does.

Just because you can't center a nested table doesn't mean it can't be done? Just set margin: 0 auto; onto that table.

whammy
03-02-2003, 10:10 PM
Hmm, that was one of the many things I tried... didn't work for me consistently though. Perhaps I'll give it another shot tomorrow (or maybe not - enough time wasted on it already, when align="center" works).

I guess you're right - the W3C isn't to blame, the vendors are.

We definitely need to see the vendors follow Mozilla's lead and start making things work the same way in all browsers!

brothercake
03-02-2003, 10:26 PM
Centering a nested table may not be easy with CSS - but CSS and table-based layouts are, to some extent, anathema. I'm sure that, whatever end-layout you're aiming for, it *is* possible if you do it without tables at all.

I think XHTML 1.0 Strict is well enough supoorted that valid, table-less layouts can be built to create anything that table-layouts could create. At least, I've yet to find an exception. Mozilla has played a big part in this - even if it hasn't been taken up by so many end-users, its existence has forced us all to rethink about standards compliance.

Having said that, I also have some frustration with the w3c. XHTML 2 and XForms is way out there, IMO; I can't see technology as non-backwardly compatible as that being useful for a good 4-5 years. To be honest, I think XHTML 2 has already lost the plot by dropping type-speficic includes like <img> and <applet>

Which is the rub; there are certainly aspects of the current specifications - some are either too vague and some far too specific - that make our everyday jobs much harder. There are times when a browser vendor has to make choices - how you decide whether to implement a stupid standard, or implement a broken interpretation? Case by case, obviously, and that's what happens. For the most part mozilla does the best job, but its puritanical adherence is, at times, as frustrating as IE's wilful disregard.

I don't believe there is a "business case for standards compliance" - it's a matter of professional pride - but if the standards-makers are not sensitive to people's actual needs, you can end up with a mess of skewed-implementations to support the mess of proprietary coding that we're "forced" to use.

Don't get me wrong - I'm as much a standards and accessibility freak as many others, but I also have to be practical - if feature X either works with a hack, or doesn't work at all, then damm right I'll apply the hack.

jkd
03-02-2003, 10:31 PM
Originally posted by brothercake
To be honest, I think XHTML 2 has already lost the plot by dropping type-speficic includes like <img> and <applet>

But <object> effectively replaces all of that.

brothercake
03-02-2003, 10:34 PM
Originally posted by jkd
But <object> effectively replaces all of that.

Yeah ... but what about the legacy browsers? And the specialist browser market, some of whom are only up to HTML 3.2 and basic forms/links javascript?

I spent so much time learning how to make accessible, well-formed and semantically meaningful documents, and XHTML 2 just wrecks all that. Suddenly legacy compatibility is impossible without server-side client sniffing.

Sure that's possible with XML-centric data management and a choice of stylesheets, but then there's caching problems associated with server-side browser detection, forcing the notion of "offline browsing" to a thing of the past (if you take the paradigm to its full conclusion)

What I want is for *one* webpage to be useable by all browsers. With XHTML 1 and CSS I have that now. XHTML 2 will make that harder.

And to what benefit? What does XHTML 2 offer that we need and can't already do? None that I've been able to figure out; but I'm open to having my mind changed on that.

krycek
03-02-2003, 11:18 PM
hmmmmm... some very good arguements all round :)

jkd - you are right about implementation being the responsibility of the vendors, but I think whammy's point about racing ahead was that because the browser makers have not got everything done just yet, it may be better to make sure everything is up-to-date first.

However there is the valid arguement that if the future is not carefully defined, it will be shoddily created. After all, we have enough problems with browser makers doing their own thing, with the existence of W3C standards. If the W3C took more time to define these standards, then we may well find the situation developing where the browser makers are once more introducing their own concepts - because of a lack of definition of those concepts in the standards.

Then again, the problem of XHTML vs HTML is a very real one. I like to use XHTML in everything nowadays, but jkd there are some things that you JUST CAN'T DO with CSS only. For instance, can you tell me how to make two side-by-side columns in CSS, the left one fixed width and the right one varying, and have both of their heights the same BUT varying with content...? This is unbelievably easy with tables, but absolutely impossible with current CSS.

Because of these limitations, us designers are faced with two options: change the design to suit the medium, or change the medium to suit the design. In many examples I find ways of cleverly altering the design to acheive effectively similar displays, but then there are some occasions when I have to reluctantly concede that HTML is the only option open to me.

So... jkd, YOU are wrong because NOT everything can be acheived through CSS. I will glady withdraw that statement if you can prove to me that I am wrong - however I am not talking so much about the standards as I am about getting my page to look right in the browser. After all, standards are great but my clients are not going to care about them if their website does not look right! :)

Mozilla is a great browser and it does a lot to close the gap between idealism and reality, but even if Mozilla was perfect, IE is DEFINITELY not, and whatever us designers do, we can't afford to ignore IE. Which is why I use a range of hacks for IE and also why, when hacks are not available, I have to use another method entirely even if Mozilla can do the right thing.

So, my opinion is that yes maybe the W3C is rushing, but they do need to set future standards before the browser makers do it first - and also I agree that the current standards have to be obeyed before vendors move on to new ones. :)

::] krycek [::

jkd
03-03-2003, 12:57 AM
Originally posted by krycek
[B]So... jkd, YOU are wrong because NOT everything can be acheived through CSS. I will glady withdraw that statement if you can prove to me that I am wrong - however I am not talking so much about the standards as I am about getting my page to look right in the browser.

http://www.w3.org/TR/2001/WD-css3-multicol-20010118/

WA
03-03-2003, 01:23 AM
I don't believe there's one correct stance on this issue; on the contrary, it is usually through the butting heads of people on both sides of the fence that the right balance emerges, and in this case, standards are being devised and implemented. So it's a good thing.

However, I think it's important to refrain from using direct vocabulary like "You are wrong" (uh hmmm, Jason?) when discussing issues like this, since it immediately puts people on the other end on the defensive, and suggests there is no room for compromise. I know Jason quite well, and know he means well, and is just being passionate, but it's important to be receptive to other people's arguments. Usually that's a much more effective way to make them receptive to yours as well. :thumbsup:

Carry on, but please people, lets try to argue while maintaining that laid back atmosphere this topic should be immersed in.

krycek
03-03-2003, 01:32 AM
Originally posted by jkd
http://www.w3.org/TR/2001/WD-css3-multicol-20010118/

That's NOT proof, jkd :D

...show me some working examples... :)

Oh yeah and WA - I'm in complete agreement with what you said (claps :)) but I didn't think anyone took what jkd said the wrong way - those of us who know him, know his style ;) He's always up for a bit of banter :p

So jkd... prove it! The challenge rest with you! :)

::] krycek [::

jkd
03-03-2003, 02:22 AM
Originally posted by krycek
[B]...show me some working examples... :)

http://www.jasonkarldavis.com/krycek.html

krycek
03-03-2003, 05:11 AM
Originally posted by jkd
http://www.jasonkarldavis.com/krycek.html

--- baffled! ---

That worked fine! And pretty well in IE, too! Not sure why I have never been able to get that particular layout to work before... I'll have to go back and take another look at my earlier method I guess.

However I did find one problem: overflow. How do you propose to deal with that? What I mean is, take the following example (based on yours):


<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN"
"http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en">
<head>
<title></title>
<style type="text/css">
html, body {
height: 100%;
margin: 0;
background: black;
}
.container {
height: 100px;
width: auto;
margin: 10px;
padding: 10px;
background: yellow;
}
.left {
float: left;
height: 100%;
width: 200px;
background: red;
overflow: hidden;
}
.middle {
float: left;
height: 100%;
width: 100px;
background: blue;
overflow: hidden;
}
.right {
height: 100%;
width: auto;
background: green;
}
</style>
</head>
<body>
<div class="container">
<div class="left">
LEFT | lots of text
lots of text lots of text
lots of text lots of text
lots of text lots of text
lots of text lots of text
lots of text
</div>
<div class="middle">
MIDDLE | lots of text
</div>
<div class="right">
RIGHT | lots of text
lots of text lots of text
lots of text lots of text
lots of text lots of text
lots of text lots of text
lots of text
</div>
</div>
</body>
</html>

This works fine unless the right-hand column is squashed too much, whereupon the text wraps back, inside the padding of the container. I'm not sure that is supposed to happen though - so I await your thought on this.

Oh yeah, and I publicly apologise to you for saying you were wrong. I honestly thought that particular layout was not possible this way... ah well, I will have to try and think of another example now! :rolleyes:

::] krycek [::

jkd
03-03-2003, 11:52 AM
No worries. Until the column model works its way through its phases at W3C, multi-column layouts aren't obvious in how to implement.

When I have some time, I'll ponder your overflow question.

krycek
03-03-2003, 12:27 PM
The oddest thing is, I just checked my code that I was trying a couple of months back, and it is using the same principle as what you have done. But, it does not work. Hmmmm. Then I noticed that I had done a couple of things to the container div, so I wrapped it all in another one and, hey presto! it worked :)

You are right though - the solutions are not always obvious and I feel it should be more straightforward.

::] krycek [::

whammy
03-03-2003, 04:59 PM
This turned out to be very interesting, after all. I will have to try some of those techniques as an alternative to using tables. I just loved how simple tables were. ;)

Alex Vincent
03-03-2003, 07:33 PM
Personally I think there are times when the W3C doesn't fully see the picture...

http://www.mozillazine.org/weblogs/weirdal/archives/002590.html

cg9com
03-03-2003, 08:02 PM
Originally posted by whammy
This turned out to be very interesting, after all. I will have to try some of those techniques as an alternative to using tables. I just loved how simple tables were. ;)

as did i whammy ;)
and this is a great thread topic!

and jkd margin:0 auto; doesnt work for me on IE5, is 5.5 as far back as it goes? not much for support.



EZ Archive Ads Plugin for vBulletin Copyright 2006 Computer Help Forum