Re: OT: Package stability (was: Re: safe to upgrade sid?)

2005-07-17 Thread Josh Metzler 
On Sunday 17 July 2005 11:35 am, Anders Breindahl wrote:
...
> All that is well known. The morale is, that I still shouldn't suggest
> Unstable to users I help install, and that I really should consider
> Testing myself. I assume that such ``errors'' as kmail breaking would be
> considered rather important to fix straightaway, if they should emerge in
> Testing..?
>
> Regards, Anders Breindahl.

Well, if such an error emerged in testing, it would take even longer to fix, 
as the fix would first have to go through unstable.  This is exactly the 
reason testing is often considered safer in regard to package stability, 
though - hopefully all errors such as this will turn up in unstable and 
therefore the broken package will not make it into testing.

Josh


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

OT: Package stability (was: Re: safe to upgrade sid?)

2005-07-17 Thread Anders Breindahl 
On Saturday 16 July 2005 21:00, Adeodato Simó wrote:
 > * Anders Breindahl [Sat, 16 Jul 2005 19:37:40 +0200]:
 > > It is rather disturbing, that errors in libraries in unstable is not
 > > prioritized any higher than the ongoing transitions.
 >
 >   Sorry, but if this fuckup has not been fixed already is because it
 >   can't be fixed without major pain, due to GCC 4 being the default
 >   compiler now.

Thanks to those who pointed this out. That had passed my attention.

I was merely providing the (thought-up) view of a user, who didn't understand 
how his or her system broke because of a software update. The frustration is 
real, but if I worried about stability of packages, I shouldn't have gone 
with Unstable.
I know that fact, but I just didn't know that I worried about the stability of 
packages. As is, I have become afraid of dist-upgrading: What is going to 
break this time?

And I suppose that I am not the only wannabe-dev, who runs Unstable simply 
because of the version numbers it supplies.

All that is well known. The morale is, that I still shouldn't suggest Unstable 
to users I help install, and that I really should consider Testing myself. I 
assume that such ``errors'' as kmail breaking would be considered rather 
important to fix straightaway, if they should emerge in Testing..?

Regards, Anders Breindahl.


pgp6LchxHU8Pb.pgp
Description: PGP signature