Saturday, March 15, 2008

WebKit succeeds where Mozilla has failed

I was annoyed when Apple announced that their new Safari browser would be based on KHTML from Konqueror. I mean, it was great that they were using and supporting open source software, I just thought that KHTML was an odd choice. It just seemed like a huge snub to the Mozilla community. Why not throw your weight behind the project leading in mindshare and clearly with the best future ahead of it? I mean, who the hell was using Konqueror anyway?

When they announced that they were porting Safari to Windows last year I thought that they really must have it in for Firefox. Here we had a browser that seemed to be very consistently gaining in market share against IE and Apple had to go and undermine it. Not that choice is a bad thing, but it seemed to me like it had the potential to further balkanize the market just as the carnage from the "browser wars" had seemingly gotten under control.

Lately I feel like I have to eat my words after seeing the successes of WebKit in the embedded space. It's freakin' everywhere now. It seems to have taken Opera's one hope away from it by pretty much owning the future mobile browser with incorporation into S60, Android and of course the iPhone. It's now in AIR, which is huge. At this point if you need browser functionality in your app it seems crazy not to use WebKit.

What happened? Was Apple right from the start about Mozilla/Gecko being too bloated and unwieldy? Well, all I know is that Firefox is painful to live with these days. It's slow. It's a memory pig. It crashes. FF3 is supposed to fix all that, and the latest beta is fast, but it's gotten to the point where if I need to quickly check something and FF isn't running already, (even if it is), I've been going for Safari or even, gasp, IE.

This is bad for Mozilla in multiple ways. If FF3 fails to deliver on promised memory and speed improvements people are going to be looking for alternatives and Safari is the next logical choice. Seems like all the cool kids are using Macs these days anyway so that just makes it even easier. Your phone's browser will probably be running it. If AIR takes off, (and I think it will), it will just be further represented there.

I do have high hopes for Prism, but will running potentially dozens of instances of FF really solve any resource consumption problems? I used to think that the whole Mozilla stack, (Gecko, XPCOM, XUL, etc.), had huge potential on the desktop. Where has that gone? Have you seen any popular/widespread/successful apps written in that besides Firefox and Thunderbird? Yes, I know about Komodo, and even Crocodile Clips, but you know what I mean. Anyway, if you were going to start writing a desktop app now which platform would you consider more, Mozilla or AIR?

Of course the huge thing that FF has that nothing else can compare to currently are the extensions. I can't live without Better Gmail. That alone will keep me running FF for the foreseeable future. If Apple opened up a Safari plugin API though? Hmm...

It does seem like the blossoming of non-IE browsers is having a positive impact on standards. Microsoft just announced that IE8 will default to standards-mode. Perhaps the future is not a victory for one side or another in browser wars 2 but the rendering, (sorry, pun intended), of all of them commodities.


Anonymous said...

You are a stupid loser, Firefox is awesome. Loser. I hate losers.

Simon said...

Mr. Anonymous, the poor guy didn't really say that FF's bad or anything... and even if he did, he'd probably been able to sound convincing. ;)

Very good article... what I'd be looking for is a Prism alternative for KDE, using Konqueror. Especially now that Konqueror in KDE4 renders pages just like Safari again.

BTW. One thing that I'm missing from all of these Prism/AIR-like things is a way to use the operating systems' native menus. A standard "unordered list"-style menu would be great standard to specify that.