Take a Stand and Understand the Standard - HTML with Style | WebReference

Take a Stand and Understand the Standard - HTML with Style

index123456

Take a Stand and Understand the Standard

The Web Travels Backwards in Time

It does. As anyone will tell you, the whole issue of "get these standards implemented" is something we've figured the wrong way around. The reason is, and this is an important point, that the W3C doesn't make standards at all. Let me explain.

Traditionally in the world of information technology, someone comes up with a bright idea, implements it, and gets it adopted, usually making a fat pile of money in the process. When an idea, process or technology gets adopted, people try to replicate it, integrate it, or, as the case may be, "embrace and extend" it. Whatever. The point is, you get someone's bright idea being used by a lot of people. That's when the issue of compatibility comes along and rears its ugly head. People want to be able to use the exact same technology, for various reasons that are too obvious to go into. So you get standardization; the people that use the technology most get together and agree on a common way of doing things in the interest of compatibility.

This is nice, but it has drawbacks. First of all, it stifles competition to a great degree; that's why standardization usually comes along long after the battle to get a process right is long ended. Every time you get a new technology, you get someone who thinks of it first, a number of people that add ingenious quirks to it, then you get a standard and it sort of fizzles and dies, but that's OK because it's grown up by then.

Take the C programming language, arguably one of the most popular ones in the world, definitely a revolutionary one, at least in terms of its heritage (read BCPL). C was good at a lot of things, and people were implementing and using it in all kinds of circumstances; the problem was, a few years after it was created, if a programmer "knew C," he didn't necessarily know anything useful at all; there were so many different and completely incompatible versions of C floating around that each and every one required you to learn everything from scratch, even though they all did mostly the same things. So people got together and made a standard (or two, or three) and things are mostly consolidated right now. OK, that's an overstatement, and writing cross-platform C is still a nightmare, but compared to the situation ten or even five years ago, before ANSI C and other similar efforts, it's a child's game and mostly well documented. Overall, and despite their drawbacks, C standards have been a good thing.

A standard is the computer equivalent of settling down and starting a family. You might miss the excitement of the single life, and some people might view it as a kind of stagnation or compromise, but it has its compensations and most people tend to find they welcome the release and security at some point.

The Web started out that way too. Tim Berners-Lee had the bright idea, organizations like NCSA and Netscape hacked on it and made it grow and prosper, but after a few years it's just too wide-spread and chaotic for comfort. Enough innovation, people are crying. Just make the damn thing work. So up comes the W3C and pumps out a couple of specifications. The W3C calls these "Recommendations," not standards. It's an important distinction, but not many people get it. The W3C is a bunch of companies that have a stake in the future of the Web, that come together and agree on things.

To a large extent, this process looks a lot like the standardization process. I mean, after all, standards will differ from existing implementations, that's the point. From a certain point of view, Cascading Style Sheets are not an innovation; they're just a standard way of doing what was accomplished by FONT elements and animated GIFs. The problem is, most of the W3C's recent recommendations are so different from current implementations in both what they do and how they do it. And they have trouble getting implemented.

index123456

URL: http://www.webreference.com/html/watch/standards/1.html

Produced by Stephanos Piperoglou
Created: August 25, 1999
Revised: August 26, 2089