On 24.05.2008, at 11:42, James Bunton wrote:
> On Sat, May 24, 2008 at 09:43:55AM +0200, Martin Costabel wrote:
>> At this occasion, I repeat my call to scrap that stinking compost
>> heap
>> called "stable tree" completely and replace it ASAP, at the same
>> time as
>> the unstable tree, by the pangocairo branch of the unstable tree.
>
> Is there any reason not to do this? The first thing anybody installing
> Fink usually does is switch to the unstable branch. Does anybody
> actually use the stable branch?
In general, the concept of a "stable" branch/tree is commonly found in
various projects. If there are enough resources to assure that it
stays more or less up to date and extra precaution is taken, then it
benefits its users.
For Fink, from my perception, the current "stable" tree is quite
outdated. I guess most active users (meaning they prefer to use
current versions of software packages) switch to unstable anyway. In
addition, when people provide packages for new software they need,
getting them into unstable already takes a while. As the process for
getting a package from unstable to stable is not very detailed and/or
predictable, people which want to promote their package (or rather the
software package they require or want to advertise on the mac) always
have to explain on their web page/their blog that their users have to
first install fink and then switch to "unstable" to be able to use
that single peace of software. (for an example, see my installation
guide for using the avr-toolchain on mac at
http://www.btnode.ethz.ch/Documentation/MacOSXInstall)
. Please note, I'm in now way try to complain about this, I'm happy
that many of us help out in keeping the quality of the package rather
high.
To me, maintaining both the stable and the unstable branch seems to
require resources which otherwise might be better used. I would, too,
vote for replacing the stable tree with the unstable one, as Martin
proposed. The current pangocairo-branch seems to a real "unstable"
tree which should only last until the pangocairo switch is done before
it would become the second next stable tree.
Instead of spending time in promoting packages from unstable to
stable, an automated clean test similar to the one proposed by Alex
Hansen (http://wiki.finkproject.org/index.php/Fink:CleanBuild) could
help to ensure that package in stable at least always build. I don't
know what I would expect more than "packages in stable build with
99.9% probability".
that's all from me, cheers,
Matthias
-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Fink-devel mailing list
[email protected]
http://news.gmane.org/gmane.os.apple.fink.devel
