Thursday, September 13, 2007

Software engineering best practices

I submitted a bug via Apple's bug tracker 4 months ago. The bug is a data corruption issue in the kernel/libc and it is very easy to reproduce. I also suspect it is easy to fix, since it is a case of saving/restoring some registers in signal handlers. Apple has not fixed it, or even acknowledged that it exists. I'm willing to bet $100 that nobody at Apple has even read the bug report. It is very likely the entire public bug reporter is just for show, sort of like Sun's "bug parade".

As for the infamous gcc x86 backend bug, well the one I submitted was closed as a duplicate... of a bug which has been open since 2003.

Update: my friend at Apple tells me my bug was fixed in June, but I'd have to be a Premium or Select ADC member to see this, since apparently even bug fix notifications are covered by the Leopard beta NDA cloud.

5 comments:

Anonymous said...

Ah, the joys of arrogant software/hardware companies, and the joys of open source.

All right there.

lypanov said...

i've had responses on a few app bugs. they had fixed the issues.

if you pay for support... they will probably be much more likely to fix :P

Vladimir said...

Oh, it's so easy to bash "just for show" bug databases, but as somebody who actually works with Sun's database, and as somebody who knows tens of engineers actively working with it, and taking the bug reports seriously, I can definitely say - Sun's bug parade IS NOT FOR SHOW. The world is not ideal, and there is only so much resources to spend on bugs evaluations/fixes, nothing new here. Now that Java is open source, let's see if any of the screams "if only Java was open sourced, then all bugs will be fixed by the community, and fast" would have anything with reality. (Hint: It won't. To evaluate, reproduce and to fix most of the bugs requires quite a lot of effort, and most of the efforts are not that fun, and of little interest to most).

Not that I entertained idea that I can actually change your opinions about Sun or Java... :)

Anonymous said...

why being so suspicious, and guessing the worse when you don't have all the facts to make sure your claims are true?
it might be just other circumstances, and isn't it better to give someone the benefit of the doubt and be more positive / optimistic about the whole situation?

george said...

With GCC, at least you can fix it your self probably.