Re: wm in default installation

Bernd Ahlers wrote:
> -f [Wed Jan 12, 2005 at 01:28:11AM +0100] wrote:
> 
>>i really don't want to start a religious war on wm's, because
>>it doesn't really matter what's in base, everybody adds
>>his/her favourite one after install anyway (so i dont mind
>>fvwm at all, and don't use it at all as well).
>>
> 
> That's not correct. I know many people which are using the "base"
> fvwm, including me.
> 
> 
>>but if license and depedencies matter, i would still suggest blackbox...
>>
>>pros:
>>-BSD licence
>>-well tested, well known
>>-lightweight (WANTLIB=ICE SM X11 Xext c m stdc++)
>>-development stopped
>>
> 
> I don't understand why "development stopped" is a pro?!
> 
> 
>>cons:
>>-development stopped
>>(patches can't go upstream, but there's only 3 of them)
>>-???
>>
> 
> Import a wm which isn't under active development anymore makes no
> sense. (IMHO) Nobody can/will fix bugs in blackbox because there is
> fluxbox. (it's derived from blackbox)
> The existing three patches are not the problem, but future bugs are.
> Or can you fix the blackbox bugs which maybe show up on sgi or zaurus?
> (if they ever get X support ;) )
> 

Blackbox (and fluxbox too) is more or less broken on all 64 bit arches 
(alpha, amd64, sparc64). Look at the XChangeProperty() manual page 
carefully, observe that in the case of a 32 bit value to be set/get you 
need to pass a pointer to a long or unsigned long (yes it's weird, but 
it was designed this way) and then look at the code that passes pointers 
to int or unsigned int. This means memory corruption.

Many other X applications share this kind of bugs BTW.
-- 
Matthieu Herrb

[demime 1.01d removed an attachment of type application/x-pkcs7-signature which had a name of smime.p7s]