Why are you using sproutcore for a static plain ol html page? There is no need to use much js on such a page, is there? And then you are ending up breaking the page on old browsers. Whats the point? Are you using the right tools for the right job?
First, I wanted to test drive SproutCore 2.0. It's always good for framework developers to be app developers themselves; and just from the feedback and collaboration today I've fixed several bugs in SproutCore. So, that's a net positive, even though I'm sorry there was temporary frustration while we got to it, and it didn't help that I was in a plane for part of the day.
The second reason is that this is just the start. I plan to add many more browsing and filtering capabilities over the next few weeks and months. I hope this can be a good demo for how to build data-driven applications in SproutCore.
3.6 barely runs on my machine as it is; a number of clients are asking me to downgrade them from 4 to 3.6 because of stupid issues like not saving your tabs by default when you close the browser. I have about thirty tabs open in my browser right now, and the last time I upgraded Firefox, it didn't bother to save any of them the next it launched.
I just can't be arsed to go to that much trouble over one website.
Wow, that's bad. I accidentally shipped the version of SproutCore 2.0 with property accessors turned on; however, Safari 5 has a bug where this blows up when called on an element, and we've been testing on Lion.
Just FYI: the version of SproutCore being used here is SproutCore 2.0 developer preview, which made aggressive use of new browser features, but missed a couple of places to polyfill. Those places are now plugged, mostly due to this exercise.
SproutCore 1.x, the current stable line, is in use for a number of major sites that work across a large number of browsers, including IE7 and IE8.
I'll assume that there's a reason why you wrote "across a large number of browsers, including IE7 and IE8" in reply to a comment about Opera?
I must admit I haven't checked right now, but at least a few months ago the site was all about how awesome Sproutcore is, but accidentally neglected to mention that there are common browsers on which it does not work. I had to install and try to use it to figure that out.
IMHO this is a serious attitude problem; launch first, make work later. That's all right if you're open about what's missing, but the Sproutcore site makes promises ("it works") that it can't keep.
That's what I get now using Opera 11.11 32bit on Linux.
Uncaught exception: ReferenceError: Security error: attempted to read protected variable: toString
Error thrown at line 9, column 22266 in x(a, b, c) in http://www.everyjs.com/js/libs/sproutcore-2.0.a.3.min.js:
In line with tomdale's quite constructive multi-framework suggestion (and yeah I gave it props on the pull's comment page too), everyjs site looks ready for a third pulldown at the top: "Render this page using <JS-library-list>"
Maybe with some loadtime stats and other statporn too
The explanation of MooTools' advantages should mention inheritance. MooTools is indispensable if you want to build a robust class-and-instance structure that more closely matches the architecture you're using in other client technologies (e.g. java, flex, or cocoa)