I, for one, am glad that Dick brought up the issue. I'd been having an
odd problem where having eclipse stop at a breakpoint in my swing
application caused my *entire* *desktop* to stop responding to all
input. The only solution was remote SSH and kill the swing app being
debugged.

I haven't experienced this problem again since setting
GDK_NATIVE_WINDOWS=1 before starting eclipse. Coincidence? Maybe, but
I'm glad the Posse gave me the idea, because the trouble was having
was driving me nuts!

Besides, it didn't sounds like "SWT Bashing" to me. It sounded like
they were reporting and the usual pissing contest one might expect as
to who was more at fault when a change in a library causes a library
client to misbehave. Not that it matters to an end-user.

I found the GDK argument that it's all fine because you can just set
an environment variable to work around the problem somewhat hollow.
Sure, that might be find for developers who can be expected to know
what the hell an environment variable is, but ... what about normal
users?

On the other hand, the observation that this change only seemed to
effect SWT apps indicates that SWT must be doing something with the
GDK APIs that no one else has thought to do. Possibly something
inadvisable, wouldn't you agree?

What's infuriating about this particular bug, whatever the proximate
cause, is how utterly in-transparent the symptoms are. I mean, a crash
and a stack dump would be better than just having the entire desktop
interface randomly stop responding to input.

For my POV both teams deserve some lumps for this one. I'm glad it was
mentioned on the podcast. I'm hopeful it won't be an issue in the
future and I'm glad I have an easy workaround for the interim.

// Ben

-- 
You received this message because you are subscribed to the Google Groups "The 
Java Posse" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/javaposse?hl=en.

Reply via email to