Re: Release-critical Bugreport for May 12, 2000
On Fri, May 12, 2000 at 04:31:05PM -0400, Daniel Burrows wrote:
> On Fri, May 12, 2000 at 11:07:50PM +0400, Michael Sobolev <email@example.com> was heard to say:
> > $ gnomeicu
> > ** ERROR **: Can't create applet!
> > aborting...
> > Aborted.
> Interesting. What package versions?
The latest in frozen.
> For me the following works:
> gnomeicu 0.90b-1
This very version.
[ ... snip ... ]
> No locale set
I tried it with unset locale as well.
> > This does not look to be working flawlessly. I can't even try to reproduce
> > what experience Alexey. I am pretty sure this is RC. :-(
[ explanation of critical and grave are skipped ]
You are right, it's not critical but grave.
> This is the one you could apply here. However, I and other people that I know
> are able to use this program. I'm sorry that it doesn't work for you, but
> I really think that "unusable" should mean "unusable for most users", not
> "unusable for Michael Sobolev".
Well, this is an arguable issue, is it not? :) We do not know how many users
are using this package, right? I'd expect that "most" means at least 50% of
users of this package. From other hand, if I use this program on several
computers may I count for that many users? If the latter is what we take into
account, then me, Michael Sobolev, counts for 3 (I have three computers on
which I tried the program) and Alexey Vyskubov (who is a friend of mine, BTW :)
counts for 2. Since I only heard from two (or three) users in response to bugs
in gnomeicu, (I do not know on how many computers they use this program on, so)
I count every that person for 1. It gives 3+2=5 "does not work", 3*1=3
"works". What is more? :)
Please do not take this too seriously. :)
> So do we have any information on just how widespread these problems are? Are
> they widespread enough to justify pulling a program which works acceptably for
> other people?
I spent a bit more time (as I said in my previous message, I do not use this
program at all, and all I wanted to see if the problem Alexey had described can
be reproduced on my computer) on investigation and found, that either there
is a bug in libpanel-applet0 or gnomeicu MUST depend on gnome-panel: as soon as
I installed the latter, it started (as I expected, BTW) to work.
I rechecked the bugs Alexey submitted, these are against 0.92 which seems to be
available only in woody... :) So we may put this discussion aside till woody
is frozen. :)
> Has anyone done backtraces to find out where it's segfaulting?
It works for me in 0.90b-1, which is potato.
So I think this bug should be appropriately marked in that special database.
Whom this request should be sent to?