February 28, 2009


Sometimes I forget how great Everything2 can be.

posted by Andrew

February 27, 2009

Safari 4 Hidden preferences

SwedishCampGround.com has a funky list of Safari 4’s hidden prefs such as restoring the old tab position. Haven’t tried them yet.

posted by Andrew


  • Why does Adobe Reader 9 come with AIR? Mmm, proliferation. Uninstalled immediately.
  • Why do so many installers assume I’m willing to grant them net access? What exactly are you doing? BTW, I love VLC’s approach to this: a message from the devs saying they hate apps connecting to the net automatically too.
  • Why do apps ignore my preference for standard antialiasing? This includes IE8 (which has an ‘always use Cleartype for HTML’ option buried in Advanced options), Acrobat Adobe Reader and the MS Live installer.
  • Silverlight install requests. I hate monopolies too, but I’m not encouraging another proprietary plugin just to break it.
  • Google Update’s strange install location + checking for updates every 10 minutes. At least it uses the task scheduler, but the apps seems to run two copies at once at 7MB RAM a pop.
  • c:\documents and settings\<username>\my documents\updater . What is this dir? Why is it always recreated empty?

Beginning to sound like the old guys from the Muppets

posted by Andrew

February 25, 2009

Safari 4

Some lovin’ and hatin’ for the Safari 4 beta today. This is just a cursory inspection from consumer and developer angles:

  • Nice visuals on the top sites & bookmarks features. Limited long-lasting appeal for essentially a glamour feature though, and it wasn’t immediately apparent how to edit the top sites, where they were stored, and how they came to get in my browser (I had the same data customisation issues with Flock a while back). Opera’s SpeedDial feels much more user-friendly at this stage.
  • Tab placement is horrendous on Windows; at least Chrome’s tabs make a pass at being in the browser window’s content area. Stop screwing with interface conventions — treating some tab mouse events as the app title bar is contextually confusing (double-click, dragging). Please stop innovating things that don’t need innovation.
  • Each Safari release reminds me how good Webkit’s renderer is. Awesome.
  • Thanks for leaving the ability to chose alternate font rendering in — I thought this might be on the chopping block. I hate XP’s rendering of generic web typefaces. Switching between the default and alternate actually exposed the cause of the header positioning bug on my Test Design 26 page, heh.
  • Please implement middle-click tab close.

I was surprised to see CSS Animations and CSS Effects mentioned on the Features page — these are Apple CSS specs that (as far as I know) are still exposed in Webkit with the -webkit- prefix. The CSS Working Group site lists these modules as an upcoming working draft (unlinked but available on the w3c dev server). Are they mature enough for use outside of an closed, internal ecosystem? I must admit that I empathise with anyone who tries to get specs through the W3C process.

Speaking of specs in progress, I was delighted to see the Safari site’s written in HTML5 (yes I’m a hypocrit for advocating unfinished specs here and questioning them above, but HTML5 is much more mature). IE demands sacrifice, so the page includes

<script type="text/javascript" charset="utf-8">

to make non-existent tags be treated as both non-void and styleable. The Safari startup page actually uses the <video> and <audio> tags too, and it sure was freaky to see video without Flash installed.

[Update: eek, the page isn’t valid HTML5. <section>, <header>, <footer> and <nav> are misused. For the <footer> links they’d be better off using a parent UL instead of <h3>s. Not sure what’s triggering the <script> error though.]

posted by Andrew

February 22, 2009

HTML 5 Reality

Great quote from Ian Hickson during the discussion on Rob Sayre’s reduced-scope spec:

> You’re the dictator. Am I reading the WHATWG process wrong?

Apparently. A dictator, by definition, has total power. I have nearly no power; I am contrained by legacy content, by the whims of implementators, by rational and logical argument, by the needs of authors and users, and by research.

On another note, Anne van Kesteren links to an awesome CSS thread which provides much opportunity for facepalming to any pragmatists reading.

posted by Andrew

February 20, 2009

The IE6 resistance

(Cleaning up an unfinished-post queue, for those wondering. This one’s a little skimpy but statistics are always fun.)

A while back I held a small workshop at TAFE on browser usage and the importance of verifying designs in multiple browsers. Lucky I don’t get sick of complaining about IE6; even now, seven years after its release, market share keeps it on our testing matrix:

  • IE6: ~36%
  • IE7: ~32%
  • Firefox: ~19%
  • Safari: ~3%
  • Opera: ~2%

(Stats tallied from several public sources late last year)

Of course these stats are a rough average from several sources widely varying demographics (and the numbers don’t even add up to 100%).

One of the few serious benefits of MS’s Vista push is IE6 being erradicated in its wake. XP’s latest service pack still doesn’t include IE7 though — they must be facing off with a lot of corporate intranet admins whose business software still depends on IE6 Javascript bugs.

Curiously, Microsoft will be ‘pushing’ IE8 to Vista and XP via OEM channels (think Dell) as an optional component. Maybe web devs should start petitioning OEMs to enable it by default? Imagine writing only one stylesheet per website.

posted by Andrew