Great podcast!

As an example of js semi-colon breakage:

x=3

switch(x){
case 1:
echo 'one'
break
case 2:
echo 'two'
break
case 3:
echo 'three'
break
case 4:
echo 'four'
break
case 5:
echo 'five'
break
}

This causes "SyntaxError: Unexpected string".

I think it's a good idea to not have semi-colons optional, ever, for the purpose of debugging things like this, and also because it follows a uniform flow. I love python (optional semi-colons) and there are very few cases where you should use semi-colons (run-on statements is the only example.) In a language like python, I say leave them out, but javascript is not uniform about where it requires them, so they should be used, if for no other reason, just because it's easier to keep track of. It's like using {} on one-line if statements. It's optional, but leaving them out makes code unnecessarily non-uniform and harder to follow.

I think Douglas Crockford was a jerk, though, in this situation. If it works in tier1 browsers, then it should work as minified code in tier1 browsers. The reason it didn't work is that semi-colons should have been added by jsmin, which in my opinion, is a bug.

I think "no one uses opera-mobile, anyway" is a super-common sentiment, and is probably the reason they are supporting other browser's prefix's. Part of the prefix's thing is that different browsers support different prefix's syntax for the same stuff (think gradients) before it becomes part of the un-prefixed syntax (browsers agree on syntax, according to standard.) If it cross-supports the other prefix's syntax, then it doesn't seem so bad, but it goes against the reasoning of using prefix's at all. I think developers should support un-prefixed + prefixed for whatever they want to support (in the case of mobile using webkit prefix & un-prefixed.)

Was the bird-browser "songbird" the firefox-based music-player?