Gambler wrote:Okay, sorry. Giving such link was stupid. Hm...
Click on "Show Google's cache" to see what I meant.
Aha. Okay, thats not the page's problem. Thats Google's problem. Google is returning *invalid code*.
Gambler wrote:Well, it's not a real website, more like a technology showcase.
Hate to get picky, but I don't know what your definition of "Real" is. However,
Chevy's site most definitely is "Real". Its dynamic, on a real-life site, and its using 0 tables and plenty of CSS.
Gambler wrote:That image thing I "drawn" in ASCII was an example of this. It seems easy to do, but it really isn't.
As another reader commented, please be more specific in where you want the text to go. It doesn't just seem easy to do - it is easy to do, unless you are leaving out some details.
Gambler wrote:Both of them do not scale horizontally.
http://www.csszengarden.com/?cssfile=181/181.css <- That one does.
Gambler wrote:Both use 2-column layout, where one column does not stretch to the bottom of the page.
The one I posted above uses 3-column, and one does stretch to the bottom of the page.
Here's a second one.
Sounds like your only complaints about "variety" are disproven. Any others?
Gambler wrote:There are over a dozen articles on how to do so in CSS, and several examples in the CSS Zen Garden. Are you defining full-height columns differently?
It does not scale horizontally, and it does not scale in terms of complexity. You would need to draw more and more images to create different layouts.
I gave an example of one that scales horizontally (which has nothing to do with a full-height column, thats a totally seperate requirement), but never the less, its there. As to scaling in terms of complexity, you need to draw more and more table cells (and yes, images too!) to create different layouts. Both tables and css have that same problem.
Your challenge was to show a definitive value that tables have that css doesn't. So far, I'm not hearing any that are accurate.
Gambler wrote:Browser-specific hacks enable the content to be displayed correctly, and accurately to most people.
BTW, did you try to open csszengarden on Windows Mobile PDA? Looks like you-know-what. On the other hand, PHPbb works perfectly fine there. So CSS is not much more portable.
A single example. However, when you say "Windows Mobile PDA", you aren't being specific at all. Which browser? There is Pocket IE, IE for CE, and more. Not to mention multiple versions of each. Not to mention you didn't specify which settings you used (images on? text wrapping on?)..
Regardless, as someone who has developed for mobile devices for over 5 years, I can state definitively that css is much more portable than tables. I worked at Verizon Wireless for 4 years, and have continued to develop for a variety of platforms, from the Treo600 (Sprint) to the Sidekick (T-Mobile).
Tables are simply not even close to CSS in terms of flexibility and support.
But I'll give a counter example -
MacSlash. On pretty much every portable device I have, its equally horrible in layout - all thanks to locked width tables.
Gambler wrote:You've given 7 examples. CSS Zen garden alone shows over 200 counter examples.
I'm not a tables zealot, but if you want to hear one giving examples, than you could visit
http://www.decloak.com/Dev/CSSTables/. The guy is a bit aggressive, but he has a sense of humor.
He's been (deeply, horribly) debunked on dozens of sites (including
metafilter), but I think
Zeldman says it best..
Zeldman wrote:The article is a metric pantsload of distortions, half-truths, and self-serving assertions, sprinkled with an occasional out-of-context fact to create the appearance of reasoned opinion.
Gambler wrote:I've given numerous examples, and I'd appreciate seeing a counter-example of a well-known site that advocates using tables instead of css consistently (not as an exception to the rule).
I named Intel, IBM and other folks only as counter-examples of your statement about "most of the best" developers.
Right. Those aren't developers discussing how to develop properly. Those are commercial pages that have been developed, which may have had a number of reasons for going with tables - up to and inlcuding supporting ancient browsers, or being driven from output by software which still produces tables.
None of which is a counter-example to the majority of web designers on the net today, all endorsing css over tables. Apples to apples, or concede the point.
Gambler wrote:Personally, I think that goodness of a technology has nothing to do with people/companies who use it. Majority uses IE, thinks Star Wars is the best movie ever, and considers Lord of the Rings to be the finest book on Earth. So what?
The difference is that its not "the majority of people". Its the majority of the *authors* of books on how to properly design for the web now and into the future. Its not "the majority of people" that are
saving terabytes of data each year by switching to css - its the leaders, and the brightest of the bunch.
You've given a grand total of *one* site as a counter example. Each of the points you've brought up as an advantage for tables (uniqueness, simplicity, portability) has been debunked.
Unless you've got some respected webdesign sites with informative articles talking about real issues with css that tables solve, I think the conversation is mostly over.
Regardless, feel free to use tables as you like, and advocate them as you like. I'm done with this conversation. The evidence is literally a mountain, and my career is strong and thriving in part because of my use of standards (including css). If you can do well without them, good luck to you.
I suspect when push comes to shove, you'll lose contracts that css developers won't. Your loss.