@media impressions - part 2

The conference was split into two tracks, and there have been quite a few discussions about whether this was a good idea. I think it is because it allows for more specialisation. In any case, here are a few notes on some of the presentations I attended.

Unfortunately I missed Tantek's on microformats—I planned to attend it, but when the time came I was just too lazy, and besides I was having an interesting talk about the preceding "Yahoo! vs. Yahoo!" presentation.


That presentation, by Nate Koechley, was my favourite one, because it's the only one that taught me some new geeky stuff. Among many other things, Nate reported fascinating findings by Yahoo's performance team, the team that actually worries about making a page load 1/4 of a second faster because that'll save 14,000 man-years when multiplied with the 5.2 billion visitors Yahoo receives every month.

For instance, Yahoo puts its <script> tags just before the </body> tag because the browser stalls while it's requesting and interpreting a new JavaScript file. From a user experience point of view it's far better to have it stall while the content has been loaded and is visible than before. The actual download time won't change much, but the perceived download time goes up.

Jeremy's presentation on "Using DOM Scripting To Plug Holes in CSS" (slides) went smooth as usual, and it basically started where his presentation from last year stopped. Just when I was about to go "No! Don't change styles like this!", he explained the problem and the solution in his usual elegant and simple way. He also enriched my knowledge of fifties country music and difficult Japanese names.

I had the pleasure of meeting James Edwards, aka. Brothercake, who has done immensely important work on combining screen readers and Ajax, and who's even published a compatibility table for events support in screen readers. As far as I know this is the first major JavaScript compatibility table that was published without me having to do anything—and that's an unusual bit of luxury I could get addicted to.


I also loved Eric Meyer's historical overview of CSS. Many people said that it resembled Zeldman's historical keynote from last year, and although that's true in a way, Zeldman and Eric have different viewpoints and treated different topics. There was no real overlap, as far as I'm concerned; just a discussion of different aspects of Web history.

In fact, there should be an effort to document this sort of interesting historical stuff—but since I'm one of the few people on the planet who's a qualified Web developer and a qualified historian, I'll probably have to do it myself. Duly noted on my increasingly huge To-Do list.

Speaking of history, I must admit I expected Jeffrey Veen's presentation "Designing the Next Generation of Web Apps" to be different than it was. When he introduced the tulip mania that rocked Holland in the mid 1630's, I nodded sagely and thought I understood what he was aiming at.

Back then people were buying a kind of options to tulip bulbs for idiotic prices that kept raising and raising. Not only the upper classes did so, but the cheaper variations were even within reach of the middle classes, which enthousiastically participated. The point, however, the one that Jeff forgot to mention, is that this mania ended in a complete bust that impoverished many middle class buyers, and even a few upper class ones. As such it's not a good example of a hype that results in economic growth: it just didn't. His other historical examples were spot-on, by the way.

The Wikipedia article doesn't mention it, but Simon Schama, "The Embarassment of Riches", chapter V.3 contains an excellent explanation of the economic underpinnings of the tulip mania.


During her "Internationalisation: Awakening The Sleeping Giant" presentation, Molly was her usual exuberant self, even though I later learned that she wasn't feeling all that well. She remarked on that when I met her just before my own panel, but I thought she was speaking metaphorically, and besides I was too caught up in being nervous. In any case, she's in my thoughts.

During her presentation a member of the audience remarked on Brussels, and that convinced me that Brussels is the ideal city to try out internationalisation theories.

Not only does Brussels have both Dutch (Flemish) speaking and French speaking inhabitants, but they hate each other for historical reasons, and therefore choosing a language is an important social and political statement. Speaking the wrong language at the wrong place can cause trouble and embarrasment, as innocent Brits who want to try out their French may notice; and even supposedly international conferences can get bogged down by local language wars.

Since I'm only an ignorant notherner I can't really explain the problem—it's just too insanely complicated. The only thing I'm sure of is that the Flemings want us Dutch to keep on speaking Dutch as long as possible instead of proudly showing off our knowledge of French—it's a sort of social best practice to keep up good relations with our fellow Dutch-speakers.

The language situation in Brussels is far more complicated than that in Belgium in general—and the Belgian situation is pretty convoluted to begin with. Besides, Brussels is the capital of Europe with its 20 official languages. Therefore it's the ideal testing ground for any general internationalisation rule: if it works there, it'll work everywhere.

Later on I had an intriguing conversation with Bruno Girin, who has spent several years of his life making complicated multi-lingual applications for the finance world, and who gave quite a few gory details of the problems in bilingual English/Arab interfaces. I hope he'll publish a bit about his experiences to enrich everybody's internationalisation knowledge.

In fact, it's time for a more active European presence in the internationalisation discussion—especially the non-English speakers. We've got more experience than the Americans in dealing with this stuff.

Tomorrow I'll describe a few thoughts I had about one of the questions that came up during the Hot Topics panel. Also, quite a few people asked me what my book is about. Although the short, obvious answer is "JavaScript!", they wanted a bit more detail. I'll write something about that, too.

This is the blog of Peter-Paul Koch, web developer, consultant, and trainer. You can also follow him on Twitter or Mastodon.
Atom RSS

If you like this blog, why not donate a little bit of money to help me pay my bills?



Comments are closed.

1 Posted by Michaël Guitton on 22 June 2006 | Permalink

What is your experience of handling UTF-8 localized content with the DOM JavaScript API? I had mixed results with browsers when trying to duplicate a node and attach it somewhere else in the document tree (e.g. most diacritics would be garbled!?) so I had to use numeric character references (NCR) instead as a safe work-around...

2 Posted by Marc van den Dobbelsteen on 22 June 2006 | Permalink

Hi PPK, this is a very nice excerpt for the ones who couldn't find the time to attend this years @media, like myself. By the way, i very much like the YAHOO! libs.

3 Posted by nate koechley on 13 July 2006 | Permalink

Hey PPK, thanks for the very kind words. I wanted to let you and your readers know that I have finally released my presentation slides: http://nate.koechley.com/blog/2006/07/12/my_atmedia_2006_slides/