On Jul 12, 2005, at 8:54 PM, Elliott Hughes wrote:
i do have a problem with the system, though. it's terrible. and that's
something that Apple's developers should try to fix, because it
reflects badly on you.
If you haven't filed bugs on them, we don't know about them. We live
and breathe by what's filed in Radar (the bug tracking system).
but engineers *don't* communicate via the bug reporter.
_I_ do, and I'm reasonably sure that I'm an engineer. That
communication is often filtered through someone else (usually developer
relations, I think), but that's they way the Apple system works.
i haven't even seen any fields in which they could do so.
I have no idea how this works. I put into the bug something like "Can
you ask the developer what happens when they do X?" and hand the bug to
developer relations. Eventually, I get the bug back with some comment
from the developer in it. I don't know what this looks like from the
developer's POV.
comment or suggest work-arounds, other developers can. in a case like
my Robot.createScreenCapture crash, i can submit a bug *and* warn
other developers *and* tell other developers about my work-around, all
at the same time --- but only if it's a Sun problem rather than an
Apple one.
With Apple it takes 2 steps. One to post to java-dev, and another to
file a bug. I understand it's quite annoying, but I don't have any
control over that aspect. I _can_ tell you that filing a bug report is
the _only_ guaranteed way to make sure an engineer looks at it, even if
that's only to say "it's a duplicate of XX".
things are especially bad in the common case where a defect is marked
as a duplicate. if you didn't submit the original, that's it: complete
silence from then on.
This is something to take up with developer relations; I'd agree that
it would be nice to track at least the status of the duplicate bug.
and of course, you can't search for duplicates beforehand.
This is likely to be intentional, the more duplicates that are filed,
the more likely the bug will be fixed.
but this is another problem with the lack of insight we get. we don't
even get an idea of what was fixed after the fact from the release
notes. reading
i count 6 RFE bugs fixed, and 32 defect bugs fixed. which would be
pathetic *if* that represented all that had been done.
I think the intent is to try to only point out things that are
generally relevant to you as a developer. Lots of radar #s followed by
the explanation "we crash less" or "we are faster" aren't so useful.
Nor is listing all the duplicate or related bugs - the "AWT threading"
bug, for instance, covered 100+ unique radars.
so i have this crashing bug with Robot.createScreenCapture. i can't
see if it's already fixed. i can't see if it's already known about.
the only thing i know won't be a waste of my time is to post to
java-dev for other developers to see.
I'm trying to tell you that filing a bug _will not be a waste of your
time_. It is the _only_ guaranteed way to let Apple know about a bug.
It's a little bit like voting in a political election - just because
your favored candidate didn't win a given election doesn't mean the
voting process is a waste of your time.
Posting a bug to java-dev might get you workarounds, and might give
notice to other developers, but unless an apple engineer happens to see
it (in their free time) it will not get the bug fixed.
Karl
--
In the force if Yoda's so strong, construct a sentence with words in
the proper order then why can't he?
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Java-dev mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/java-dev/email@hidden
