On Standards
J. Morse Loyola, 24 January 2003

Note: some of the specific details mentioned in this article are now obsolete; for example, AOL Time Warner is now just Time Warner; as well, these pages now validate as XHTML 1.0, not HTML 4.01, and the GIF patent finally expired, removing that legal roadblock to using GIF images. I don't feel that these changes detract from the general idea discussed however, so with the exception of the validation link I'm presenting this article as it appeared on the old site. -- /i.

There are three ways to create web pages on this here World Wide Web, and each of them has their limitations. They are, in order of popularity:

1. It Looks Fine On My Monitor
2. You Need Hypertext Transponder Version 3.14159 Build 666 To View This Site
3. This Site Is So Standards Compliant It's Being Reccomended As An RFC

The first way listed here is, unfortunately, the most common. Its adherents don't bother to learn how to do proper markup, because they have pretty little drool-proof tools to do it for them. The more egregious offenders will even have "C:\My Documents\Stupid Image.jpg" as an argument to an IMG tag; yes, I've seen this even on commercial sites. Granted, no one can check their work with every browser out there; after all, when was the last time you fired up charlotte on OpenVMS? Still, I have to wonder - is basic competence too much to ask? The answer to that ought to be obvious to anyone who's voted recently.

A subset of the WYSIWYG jockeys, however, is arguably more annoying: the Browser Requirement people. Here we've built a scalable, public, worldwide network accessible from any kind of hardware with a TCP/IP stack, and some lazy, ignorant tosswit is going to demand that I run the latest Wintel virusware in some godawful resolution just to read whatever snivelling adolescent backwash they could be bothered to mash together? I don't fucking think so, boys and girls. As has been said before, "...by God I KNOW what this network is for, and you can't have it".

Which brings us nicely to number 3, the unbearable pedants. These folk are easy to spot, with their chest-banging braggartry and macho posturing: you'll find link buttons all over the place pointing you to the W3 Consortium's HTML Validator; no doubt the authors will even have a note somewhere talking about why the site doesn't display properly in AOL Time Warner's Netscape Communications Corporation's Communicator 4 Suite. What's even more frustrating about the Standards Pedants is that (repeat after me, class) THEY ARE RIGHT.

I'm sure you can see where I'm going with this, so we'll start off with a demonstration. Let's try to display a PNG image, shall we? Seems simple enough, or so you'd think.

What it should look like:

AOL Time Warner's Netscape Navigator 4:

Your browser:

What's the deal with that?! You'd think the damn thing could display a simple image correctly, wouldn't you. Well, guess what--Navigator 4.7¹ and higher can't seem to display the black pixels in a PNG graphic. Not only that, but earlier releases couldn't even get that far (which you'll notice if you use, say, Navigator 4.0). And we're not even going to talk about transparency; support for that is so far beyond fucked up it's not even worth a demonstration.

Why not just use GIFs, then? Why bother with this "PNG" thing at all, when a bright, shining, and established standard already exists--GIF! That answer is simple: I can't afford to. Unisys, who own the GIF patent, have been demanding that sites that use GIF images pay them licensing fees. See what happens when you try to create standards based on non-free code? PNG is an established standard that is not based on code that could get you sued, so that's what we use. For more info on the debate, have a look at burnallgifs.org.

But wait, there's more! Another WWW standard is Cascading Style Sheets, or CSS². CSS is a way of assigning the same value to a tag--a color, for example--across an entire web site. This makes it orders of magnitude easier for a designer to change, say, the background color on all the different pages. Instead of making the change on every page, he or she just has to do it once, and the change is applied to every page on the site. Sounds great, don't it?

Of course, neither AOL Time Warner's Netscape Communications Division nor Microsoft Corporation got that right. They both tried to extend control over it in an attempt to lock out the competition, and to a degree they were both successful: they just about managed to lock us out of the very same system we built for them. Microsoft Corporation may have won the browser war, but the rest of us were the real losers.

With that in mind, our stance on the Usability vs. Standards debate is that we will do what we can to make things show up properly in your browser, but we have to draw the line somewhere. Mind you, we don't require any browser more powerful than telnet to read the content of the site, but some of the more flashy bloat--like images--might look a little strange in older browsers. We are left with no other reccomendation than that the time has finally come to put the generation 4 browsers to rest, bite the bullet, and upgrade.

¹ I'd love to test Microsoft Corporation's Internet Explorer, but I can't. They don't make it available for any platform I have access to, nor have they made the source code available. Too bad, so sad, their loss.
² Some may argue that Cross-Site Scripting is also a web standard, and they're probably right.

Got something to say about it? Say it here.