3 ms·

Wild guess (I've never programmed Javascript (pathetic, I know)): They've used something like http://code.google.com/closure/ http://code.google.com/closure/ on

Wild guess (I've never programmed Javascript (pathetic, I know)): They've used something like http://code.google.com/closure/ http://code.google.com/closure/ on it to make it more efficient, and any comments have been "optimized" away together with the intentions behind the web as a text based, open, readable format. That way they can make sure HTML5 really IS the next Flash, i.e. so unreadable that it might as well be a proprietary format. Your source code will be safe from prying, learning (i.e. stealing) eyes.

ableal 16y ago
Well, that link (http://code.google.com/closure/ http://code.google.com/closure/) invites one to "Try the Closure Compiler", leading here: http://closure-compiler.appspot.com/home http://closure-compiler.appspot.com/home So I pasted in the source (http://macek.github.com/google_pacman/src/js/pacman10-hp.3.js http://macek.github.com/google_pacman/src/js/pacman10-hp.3.j...), and tried a couple of the optimization levels. The first squeezes whitespace, the third mangles everything ("Saved 41.37% off the original size (17.94% off the gzipped size)"). But the original is pretty clean, readable source code. It even uses spaces for indentation. Clearly not the output of an optimizer. Code lines such as "g.currentDotEatingSpeed = g.levels.dotEatingFrightSpeed * 0.8;" are better than possibly-out-of-sync comments.
- gahahaha 16y ago
  According to the README, it seems like the nicely formatted version has been cleaned up before it was posted on github. ("Thanks to: SteD for providing a nicely formatted version of pacman10-hp.2.js) /I'll stop talking about things I know nothing about now...
  - ableal 16y ago
    I also dislike binary blobs, or obfuscated 'source', and hadn't tried the optimizer before (thanks for the link, btw), so I gave it a shot. But this one is the 'real' source. The mangled version is being served over at the now permanent URL (http://www.google.com/pacman http://www.google.com/pacman , http://www.google.com/pacman/pacman.2.js http://www.google.com/pacman/pacman.2.js), doing its job of saving bytes/watts.

ableal 16y ago

Well, that link (http://code.google.com/closure/ http://code.google.com/closure/) invites one to "Try the Closure Compiler", leading here: http://closure-compiler.appspot.com/home http://closure-compiler.appspot.com/home So I pasted in the source (http://macek.github.com/google_pacman/src/js/pacman10-hp.3.js http://macek.github.com/google_pacman/src/js/pacman10-hp.3.j...), and tried a couple of the optimization levels. The first squeezes whitespace, the third mangles everything ("Saved 41.37% off the original size (17.94% off the gzipped size)"). But the original is pretty clean, readable source code. It even uses spaces for indentation. Clearly not the output of an optimizer. Code lines such as "g.currentDotEatingSpeed = g.levels.dotEatingFrightSpeed * 0.8;" are better than possibly-out-of-sync comments.

gahahaha 16y ago
According to the README, it seems like the nicely formatted version has been cleaned up before it was posted on github. ("Thanks to: SteD for providing a nicely formatted version of pacman10-hp.2.js) /I'll stop talking about things I know nothing about now...
- ableal 16y ago
  I also dislike binary blobs, or obfuscated 'source', and hadn't tried the optimizer before (thanks for the link, btw), so I gave it a shot. But this one is the 'real' source. The mangled version is being served over at the now permanent URL (http://www.google.com/pacman http://www.google.com/pacman , http://www.google.com/pacman/pacman.2.js http://www.google.com/pacman/pacman.2.js), doing its job of saving bytes/watts.