Page style:

A Blue Perspective: Future-proof

17 March 2004

This site gave me a good laugh when I clicked on the splash image in IE 6.

For anyone whose user agent actually does identify as "Netscape", I'll save you some trouble and tell you that up came this message: "Sorry, but this presentation cannot run on your browser. We recommend using Netscape Navigator/Communicator 4.x+".

As laughable as installing Netscape 4 is now, back then (HTML 3.2!!!) it would have seemed perfectly reasonable. But then again, restricting year dates to 2 digits in software programs seemed reasonable too. There's no guaranteed way to future proof anything you make, but there are ways you can make sure that they at least have a chance. Restricting content based upon something which you think you can detect is one guaranteed way to go out of date, and at worst make your content inaccessible to users in the present.

Often I'll visit a Flash based site and have the site's scripts say that I don't have the correct version of the plugin, then offer me no way of getting any further. Sometimes I do have the correct version; sometimes I've got a lower but compatible version. I might even have a plugin from an entirely different company which somehow lets me read Flash. Either way, I'm not going to spend a couple of minutes downloading yet another plugin just to view your site, particularly when I don't even have any idea what your site can offer me yet. So, you can warn me that I might not be able to view your content properly, then give it to me to make heads of, or you can lose a visitor entirely. Is it really a choice?

This isn't just my usual Flash beat up though. I can see the same problems occurring with CSS Hacks. When you specify a rule in a stylesheet that is designed to cause errors for one particular set of browsers, your code immediately becomes obsolete. You can't predict what future versions of the browsers will do with that hack. They might apply it properly, they might not. If they don't, you'll have to modify your code – find a new hack, nest hacks, re-write rules – it's a downward spiral. Then, when the next version comes out, you'll do it all again.

All this is totally counter-intuitive to Standards. Standards are there to provide a stable, compatible platform on which to build web pages. A Standards-compliant page written now should be readable by an agent in 10 years time, that's the beauty of them – no compromises, no special treatment; standardisation. In the current climate of frozen browser development it's tempting to ignore the lessons from the past, but if you do, you'll get burnt, and I don't want to be cleaning up your code.


1/4. 17 March 2004 @ 04:56, Mark Wubben wrote:

You might want to read this:

2/4. 17 March 2004 @ 07:47, Ben wrote:

I find that in my experimental work I much prefer to do without hacks and use pure CSS, and then if it doesn't work, so be it. A little warning message is often enough to indicate that the version being viewed may not be as intended.

3/4. 18 March 2004 @ 22:59, The Man in Blue wrote:

... and for quite simply the most stupid use of hack-type circumvention I've ever seen, go to:
in IE, or just take a look at the first couple of lines of content in the code.

It's not as if the design's even that cutting edge anyway ...

4/4. 20 March 2004 @ 06:14, Trent wrote:

Wow, that rad-e8 site is very user hostile.

I still think is the most idiotic example of browser blocking. "You must be on a Windows Operating System using Internet Explorer version 5.0 or higher". Oh really? No Mac or Linux users? No Firefox/Opera users? Fine, I'll take my money elsewhere.

Sheer idiocy. I once was able to access it using a non-IE browser and an altered user-agent string, and the site looked fine. Not that I'd buy anything from them.

Post your own comments

Fields marked with an asterisk* are mandatory. All HTML tags will be escaped. http:// strings in comments will be auto-linked.