--- Log opened Tue Sep 06 00:00:37 2011
03:33 < mcallan> conseo: i'm online again c, i'm afraid it's rather late for you
03:35 < conseo> no prob. my sleeping habits drive me crazy :-D
03:40 < conseo> i think i prefer option 1)
03:40 < conseo> the html 5 cache is not really necessary yet, so we can still postpone that, right?
03:40 < conseo> + i really like the resource stuff
03:41 < mcallan> ok, it's fairly simple to describe.  i can give you instructions
03:42 < mcallan> are you going to work on it now?  or get some rest first?
03:46 < conseo> i am going to work on it now, but i am tired
03:46 < mcallan> let's skype then
03:47 < conseo> ok
03:47 < conseo> btw. in VotespaceBiterSS you write the loop with a break command
03:47 < conseo> wouldn't for( Person person : persons ) do the trick better?
03:58 < mcallan> http://zelea.com/project/votorola/a/count/resource/
04:06 < mcallan> http://zelea.com/project/votorola/a/count/resource/SacJS.java
04:09 < mcallan> http://zelea.com/w/Site:Null_count
04:09 < mcallan> http://zelea.com/w/Category:Counting_method
04:22 < mcallan> http://zelea.com/project/votorola/a/xf/TheatreToolbar.java
04:29 < mcallan> http://zelea.com/project/votorola/a/xf/vote/_/
04:53 < mcallan> http://zelea.com/w/Category:Account
05:42 < mcallan> conseo: note there is a little snippet of css for it, too: http://zelea.com/project/votorola/a/web/base/count/resource.css
05:42 < mcallan> (at bottom)
05:44 < conseo> ok
05:49 < mcallan> conseo: there it is, i committed it
07:39 < conseo> ok
07:39 < conseo> thx
10:19 < conseo> mcallan: foreach seems to use an iterator: http://download.oracle.com/javase/6/docs/api/java/lang/Iterable.html
10:54 < mcallan> true, i was wrong about the array (thinking of something else).  so if the compiler actually constructs the iterator (which seems probable), then construction and garbage collection will slow it down.  biters must run fast and frequently, which is why they try to construct no objects
10:55 < mcallan> if you look at profiler of vm in operation, you will know what i mean by garbage.  it builds up, and then the vm pauses to clear it away
10:56 < mcallan> (in normal code, you should not worry too much about that)
10:56 < mcallan> (always optimize later, only at known bottlenecks.  but biters are *designed* for speed)
11:06 < mcallan> (i could be wrong.  when using library iterables like ArrayList, the compiler might optimize away the iterator and access the array directly.  it's possible to worry too much about this kind of thing, i guess)
11:34 < conseo> hmm, ok. the only other information (besides profiling which i haven't done yet) is this:
11:34 < conseo> http://stackoverflow.com/questions/256859/is-there-a-performance-difference-between-a-for-loop-and-a-for-each-loop
11:35 < conseo> i am off, cu later
11:38 < mcallan> cu later
11:41 < mcallan> they assume the last thing i suggested, that the compiler optimizes away the iterator for an ArrayList (if that's what your list is). the people who say it *always* faster are simply wrong, because not all Iterables can be optimized by the compiler
11:43 < mcallan> i meant, "*always* the same"
17:52 < conseo> ok
--- Log closed Wed Sep 07 00:00:53 2011