This is a bad argument. They are affected by microseconds and kilobytes, in proportions insignificant compared to the memory usage and parsing time already required.
As you're the article author, if you're even remotely interested in being taken seriously (because right now there's less-than-no reason to) it behooves you to actually demonstrate a perf or memory delta that anybody would notice in the first place.
In this particular context, it shouldn't be the developers task to minify code anyway. If the browser engineers deem that there is reasonable speedup to be gained from minifying extensions, then they should do it as part of chain from developer to user.
Do you have any benchmarks to offer as evidence that the effect is significant?
I thought people minified code in extensions just for the two reasons: force of habit (they do it for all their JA that is served online) and obfuscation.
(not sure why you are getting downvoted: your point is valid, though probably insignificant)
The code is compiled (once) and injected into all pages loaded.
The minified code and original code will look exactly the same at this point unless the minify process does something really clever.
Is there something I'm missing, or should someone file a bug report with Opera if their lexer is taking noticeably more time and memory to strip whitespace?
I’d go so far to say that someone should file a bug report with Opera if they are lexing extension code for each page individually, rather than compiling it once on startup and then using the compiled code for the individual pages.
