http://www.zeldman.com/dwws/pdfs/0735712018C_08.pdf
Especially insightful is the part that goes:
A Transitional Book for a Transitional Time
To the kind of standards geek who spends hours each week arguing about the
evils of presentational markup on W3C mailing lists, what we’ve done here is evil and harmful. For that matter, we’ve also sinned by using tables as anything other than containers of tabular data, by specifying widths and heights in our table cells and by setting image margins to zero in markup. In fact, in the eyes of some, this entire chapter is sinful. Some standards geeks might not think much of this book, quite frankly. In their view, we should be telling you how to write semantic markup instead of letting you think it’s okay to sometimes use tables for layout.
But the thing is, it is okay. Maybe it won’t be okay some years from now, when designers use and browsers support purely semantic future versions of XHTML and rich future versions of CSS and SVG. But this is a transitional book for a transitional time. “Web standards” is not a set of immutable laws, but a path filled with options and decisions. In our view, people who insist on absolute purity in today’s browser and standards environment do as much harm to the mainstream adoption of web standards as those who have never heard of or are downright hostile toward structural markup and CSS.
```````````````````````
><p> is a semantic structure, it instructs a user agent that this is the end
>of a series of thoughts. How it is respresented visually can be
>accomplished via CSS. I can indent a <p>. I can make margins or padding
>bigger or smaller. Heck, I can even place a paragraph above another one
>using CSS. Every user agent has its own style sheet as it were to affect a
>presentational representation. User Agents that are aural (ie, screen
>readers, will bring a small pause at the end of a sentence (as denoted by a
>period) and a slightly longer pause at the end of a paragraph. The semantic
>structure of the document is important, but it isn't necessarily
>presentational.
>
>Not all agents reading your web page will rely on the visual. Semantic
>structure is a growing area. Does your page make sense without style sheets
>or any outside visual indicators? There is a HUGE reason for the separation
>of content/presentation in this regard.
>
>As for layout grids being under every design. I've seen some way cool
>designs at CSS Zen Garden, that mix things up. Right now, for the most
>part, we are still totally in a mind set that our web pages must mimic
>print. 3 column layouts with a header and footer are newspaper/magazine
>layouts. They don't necessarily mimic a table, they mimic what we are used
>to. People are used to something familiar. Heck when television first came
>out, there was usually an announcer announcing what the actors were doing
>even though people could see it happening. This was from radio, where people
>couldn't see what was happening and had to be told.) Same thing.
>
>As to CSS not being able to handle specific things, CSS was never intended
>to do one thing, CSS is in use not only for Web Pages, it can be used for
>XML, Flash uses CSS for presentation now. If you are going to blame someone
>or something for not having style sheets show up the same in different
>browsers, blame the browser makers. I'm trying to create a class on CSS
>right now, when I get to certain features, I can only show them on specific
>browsers (usually Opera). Not because the spec is wrong, but because the
>browser makers have not chosen to either implement the spec in its entirety
>or accurately. Heck IE6 is the LEAST compliant browser right now and
>Microsoft makes NO BONES that it doesn't give a rats' patootie.
>http://westciv.typepad.com/dog_or_higher/2004/05/plus_ca_change.html
>
> _____
>
>From: Irvin Gomez
>
>The BIG problem with the theory of "separating logic/content from
>presentation" is that you just can't do it! Take the <p> tag, for example.
>It neatly separates chunks of information in both a structural AND VISUAL
>WAY. How can we deal with that? Besides, most experts are slowly coming to
>the conclusion that there's not only no way to separate content from
>presentation, there's also no need for it. Presentation is part of the
>content, anyway!
>
>Another point frequently missed by "purists" is that every design has an
>underlying layout grid, that basically mirrors the behavior/attributes of a
>table. Unfortunately, most of the initial CSS proponents didn't have a
>design background, so these issues escaped them (that also explains their
>ugly websites!). To the point that they ignored the simple fact that CSS is
>a concept (down to the name of "stylesheets") derived from print design.
>
>I guess until CSS 3 (or future versions) takes care of these issues, we are
>best adviced to go with hybrid layouts, as you point out. But then again,
>once that's done, then new issues will crop up: what about text wrapping
>around pictures (like the "runaround" feature in Quark Xpresss)?...
>
>> <http://www.houseoffusion.com/banners/view.cfm?bannerid=40>
[Todays Threads] [This Message] [Subscription] [Fast Unsubscribe] [User Settings]
