CSS purity vs. table hacking

Our buddy Dave Slusher makes excellent points about why using tables (the pre-CSS structure for web design) is often more efficient than using CSS for a lot common Web tasks. Even as a CSS-purist (in theory, anyway... in my own work, I still often use tables) I found his latest post on the subject both funny and thought-provoking.

CSS fundamentalists are exactly as big a pain as any other form of fundamentalist and for the same reason - disagreeing with them pokes a big hole in the purity of their belief system and they feel the need to “correct you”.

While I'll cede the point, I do want to make one case for exception: The larger and more complex your site is, the more benefits you accrue from semantic web purity. Not today, of course, but moving into the unknowable future. If you're archiving a lot of data and documents, the extra effort you put into keeping your content / presentation division pure could be the difference between profiting from your work and facing an enormous and expensive data-recovery project.

Having worked for a news website, in which previous management had used tables and frames and inline styles and -- well, just about anything that struck their fancy, apparently -- it became obvious that starting over would be easier than trying to upgrade the existing site. And, since I believe the real value of big sites in Web 3.0 will be their machine readability (not just their searchability), trying to convert non-semantic sites to Semantic Web functionality will become a big issue during the 2010s..

CSS works poorly today because some browser makers (not to mention any names, MICROSOFT) refuse to work together on compliance. Consequently, EVERYTHING about web design remains a series of hacks and compromises. But this is a temporary condition: XML functionality is going to drive the next wave of Web profit, and for that to pay off, browsers are going to have to start rewarding sites that translate code properly.

The idea of a Semantic Web begins with the notion that content and presentation are separated for a reason. We're about 10 years in, and we're still waiting for the consistent browser support that will make that effort worthwhile. Since most companies focus on the now, there's really just no reason to fret about Web standards compliance. But if you're thinking ahead to the unpredictable future, here's one thing I think we can know: Whatever that future brings, it's likely to be based on XML.

If you build around that, I suspect it will wind up being worth it. But it's an investment, and investments involve risk and suspended gratification.

Comments

[Grand Rapids Web Design]

CSS Purist? How do they differ with the Linux Puritans and the Open-Source Fundamentalists?