Friday, December 29, 2006

Picture perfect

BBC NEWS has published "The best pictures sent in by our readers during 2006", including one by my sister. Congratulations Christine!

Wednesday, September 20, 2006


Ajax is at the heart of a quiet revolution that is sweeping the Web. Many people have seen, and been impressed by, its effects; fewer have heard of it.

Ajax is not a single technology; rather it is a cobbling together of several disparate tools, adapting them in ways they were not designed to work. In other words, it's a kludge, but it's one that makes dynamic Web pages look a lot more like desk-top applications than they ever did before. It will never be what Java could have been without the Sun/Microsoft pissing contest, but it's the best we've got for now.

The term "Ajax" seems to have been coined by Jesse James Garrett, and one of its earliest uses was in his February 2005 essay for Adaptive Path, where he says it "is shorthand for Asynchronous JavaScript + XML". He cites Google Maps as one of the best-known examples of Ajax technology, and references Joel Webber's blog as simple as possible, but no simpler for an explanation of how it works.

Throughout 2005, the Web was abuzz with tales of Ajax. In September of that year, venture capitalist Dan Grossman in his blog A Venture Forth listed his Top 10 Ajax Applications, followed swiftly by his
Top 10 Ajax Applications (Part 2).

In 2006, the novelty wore off, but quietly in the background people have continued developing their Ajax applications. "Applications" is the appropriate word, because of the way Ajax brings the Web and the desktop closer together than ever before. When Protopage featured in Dan Grossman's original list , it was very close to being the Web application I had always wanted. It has been steadily refined ever since, and my own Protopage (with its public interface at is now one of my favourite Web portals.

People who care about accessibility may have concerns about Ajax. The most obvious one is that it relies on JavaScript, which has a different flavour for every browser on the market. Microsoft's version isn't even called JavaScript, but JScript. But there is a standard, ECMAScript, which provides cross-browser compatibility for a significant subset of JavaScript and JScript.

Another concern is that Ajax applications sometimes "break the Back button", but I fear there is an element of cargo cult thinking behind this taboo. When using a browser to step through a series of static pages, it is obvious what the Back button should do. When running an application within a browser, it is no longer obvious: should it go back to the previous page, or to some previous state within the current application? Ajax developers should give careful thought (and user testing) to what the back button should do in any given situation.

Further reading

Ajax (programming) in Wikipedia
Ajax in SWiK

Wednesday, August 30, 2006

The wacky world of the Web

There's been a bit of a kerfuffle lately about Steorn's perpetual motion machine. On their Web site, they proudly proclaim:

"We have developed a technology that produces free, clean and constant energy.

This means never having to recharge your phone, never having to refuel your car. A world with an infinite supply of clean energy for all.

Our technology has been independently validated by engineers and scientists - always off the record, always proven to work."

They're obviously doing something right, to get all of this publicity, but exceptional claims call for exceptional evidence. If you want to write off a century and a half of thermodynamics, that calls for very exceptional evidence indeed. And their evidence? "independently validated by engineers and scientists - always off the record". In other words, none at all. I won't be holding my breath.

Reading about Steorn reminded me of another proponent of alternative energy sources, Gary Johnson. On the face of it, he seems like a sensible chap. His book "Wind Energy Systems" (on that page) is a pretty decent text. But scroll down to the bottom of the page to find his two treatises on a whole new energy source. The evidence for its existence? The gist of the argument seems to be:

1. All energy sources known to physics are running out.
2. God wouldn't let us run out of energy.
3. Therefore physics is incomplete.

It's logic, Jim, but not as we know it.

Browsing through The Daily WTF, I stumbled across a reference to Alex Chiu. Clearly an out-and-out fruitcake. Or is he? Maybe he's found some suckers to pay for his snake oil. Even a brutal interview on Slashot doesn't seem to have perturbed him.

For a real loony, how about Gene Ray? Or is he some experiment in artificial intelligence?

Usenet newsgroups seem to attract their fair share of nutters. From, there's Mike Corley (Just how incompetent do you think MI5 are? They've been trying to kill you for fifteen years?) And from sci.military.naval, there's Alan Yu, who has been persecuted for years by "invisible tiny (like small ant size)" government agents.

Friday, June 30, 2006

Quis custodiet ipsos custodes

My attention has just been drawn (thanks Edward) to Design by Fire and, in particular, its critique of Jakob Nielsen's Guidelines for Visualizing Links. Well I can't pass up the opportunity to pass comment on a site passing comment on a site that passes comment on all other sites.

First a few comments on Design by Fire itself. It looks good. Very good. Better in Firefox, but not bad in IE. But it relies on Javascript for navigation. Minus several million for accessibility.

Now for their comments on Nielsen.


XHTML? Where have you been? Try Googling in ciwah. Internet Explorer is broken. It cannot handle XHTML without some serious hackery. As long as 90% of Web users cannot deal with XHTML qua XHTML, it would be stupid (or ignorant - I've done it myself) to use it.

HTML Strict is the only way to go.

Verdana? ciwas is the place to Google for that. Verdana is produced by Microsoft. In an astonishing act of magnanimity, it is available for Mac as well as Windows. It is not, nor is it ever likely to be, available for Linux, FreeBSD or other systems. That's not a major problem though, is it? After all, CSS provides for graceful degradation by allowing a list of alternative fonts to be suggested. Unfortunately, Verdana lies about its size - at a given nominal size, it is taller and much wider than other faces. That means it is impossible to include Verdana in a list of fonts that will look even vaguely similar.


Flash? There's a saying; well okay, it's not widely known, indeed I coined it myself, but I've been using it in sigs and elsewhere for some time:

Flash doesn't make Web sites inaccessible; trained monkeys make Web sites inaccessible.

Flash isn't necessarily bad. And Macromedia have given serious thought to accessibility issues. But Flash attracts trained monkeys.

The problem with Flash goes deeper than that. I'm willing to entertain the possibility that a competent designer could use Flash to enhance a Web site, but I have never seen an example. Pause a few moments to let that sink in.

I usually use the Web to find information. That usually means reading text. I have never seen a use of Flash that allowed me to find what I was looking for more quickly or easily.

Of course there's more to the Web than text. Occasionally, as with Flickr, images are central to the purpose of a page or an entire site, but a few well-chosen graphics can add life to almost any page. I'm prepared to believe that Flash offers potential benefits that go way beyond simple line graphics or photographic images. but those benefits have a cost. Even with today's broadband links, Flash takes time to download. And I have never seen a Flash movie that was worth waiting for.

And yes, that's a challenge.