Useful site for practical information about HTML5.

Robert Nyman created a proper test suite + compatibility table for JavaScript 1.6-1.8 features. Useful!

Stefan Kolb continues his series of tests of JavaScript libraries on mobile phones. This time he did the TaskSpeed tests on ten Nokia S60 phones.

Conclusion: Dojo again the fastest library; this time Prototype is the slowest.

Microsoft will not include IE in the Windows 7 version for Europe, which will be released in October. Computer vendors may add a browser to the system on their own, though.

Still, there may be a few computers that will have no browser installed. I'm curious what will happen next. How will these people get their browsers?

Some information about the Microsoft widget system. Looks a lot like the W3C one, which is good.

More news on the Opera vs. Microsoft EU fight: the EU might want to go for the "five" option, in which new Windows computers come with all five browsers installed. I doubt the wisdom of this move.

Zeldman gives a useful overview of where we stand with regards to web fonts.

John discusses some new ECMAScript 5 features: strict mode and native JSON parsing. (It should be noted that IE8 is said to contain the latter already, though I haven't tested it yet so I can't say whether it adheres to the proposed implementation completely.)

A useful comparison between jQuery and MooTools.

Andy Clarke discusses possible solutions to the IE6 problem, and then highlights his own: create one Universal Internet Explorer 6 CSS file that will give any page a nice-but-basic typography and margins, but no layout or grid at all.

Interesting idea. We'll have to see what clients want (or whether they'll notice), but it sounds like an excellent intermediate solution between "don't bother with IE6" and "make IE6's rendering pixel-perfect".

Blackberry to support W3C Widgets in the future? Of course this article doesn't give technical details, but the inevitable figure with the boxes and arrows pointing anywhere does seem to sketch a widget context.

Apple refuses iPhone apps that are written with the Phonegap library.

If they don't rescind this ban I'm forced to conclude that Apple is Evil.

Steve Souders discusses situations in which it might be useful to spread your assets over multiple (sub)domains, even though that will lead to extra DNS lookups.

Kangax writes about testing methodology.

Initially he proposes reading out support for certain CSS declarations in JavaScript (i.e. if the browser supports element.style.marginLeft it supports margin-left). I disagree strongly: you're only testing whether a certain JavaScript object exists, and that does not necessarily say anything about CSS.

Fortunately Kangax is aware of this limitation:

A mere existence of a CSS property doesn’t tell us about an actual implementation and its conformance to a specification. A browser might have “borderRadius” property with a proper string value; it could allow to assign to that property and even set its value to a specified one after assignment; yet, it could never make borders rounded.

As far as I'm concerned this means that the testing method Kangax proposes is unusable in practice.

If you want to test border-radius or margin-left, create a CSS test case and look at it. You'll quickly see if it works or not. Don't try any automated testng, though. It's just not reliable enough. Ever.

This rule is especially important on mobile phones, where we can't be sure of anything. I will continue to test exclusively by hand.

More information about the Opera/T-Mobile widget manager. Something that's not clearly explained is that this widget manager is only available for Windows Mobile.