-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On 9/23/11 11:31 AM, Hanspeter Niederstrasser wrote: > On 9/23/2011 10:56 AM, Alexander Hansen wrote: >> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 >> >> On 9/16/11 2:35 AM, Martin Costabel wrote: >>> On 15/09/11 16:13, David R. Morrison wrote: [] >>>> But there are also lots of unmaintained things in unstable, >>>> many of them very old, so I think it would be dangerous to >>>> just dump everything to stable. I like Alexander's >>>> approach. >>> >>> Why would this be more dangerous than the current situation, >>> where almost everybody is using the unstable tree anyway? Do >>> you know anyone, except some innocent newbies, who is really >>> using the stable tree? >> >> That's more of a question of "Do we bother to keep this old, >> broken stuff or update it?" In principle it's probably just as >> easy to decide on that after a rollover. > > The 10.5/unstable buildworld from back in August 2010 took care of > a lot of brokenness in unstable. And those packages that still > remain broken are not necessarily the ancient abandoned ones, but > instead many belong to people who are still in communication with > the project. > > Here's the list of the buildworld results. Important: this is from > August 2010. A lot of things marked broken there have been fixed > in the past year. > > http://www.snaggledworks.com/fink/buildworld/2010-08-14/out/maintindex.html > > I think the primary thing with a lot of downstream effect that has > gotten updated since then was Qt4 and I know it broke a number of > things, so it might have a wider effect than is known. If > MaxBuildJobs gets turned to on by default in the next Fink release, > that will probably introduce some more brokenness, but running a > new buildworld should find them. > > I do _not_ have the ability to run a buildworld again for the > foreseeable future. However, if anyone has a machine with > available CPU cycles (expect 2-3 weeks), I can easily walk them > through the process of getting it going. > > Hanspeter > > ps. the earlier 10.5/stable buildworld (in March?) showed about > the same proportion of broken packages, so just stable was not > necessarily stabler. I think this was one of the good things to > come out of the new 10.7 tree. It'll be stable on 10.7 ;) >
This suggests a more automatic way to start phasing unstable out. 1) If a package is in the 10.7 tree and has a nominally identical counterpart in the 10.4 tree (same version, revision | dependencies, etc.) then the 10.4 counterpart can be moved to stable, too. 2) As these counterparts get added to 10.4/stable, we _remove_ them from 10.4/unstable. That way we don't have to worry about doing diffs between the trees. We'll do the same thing if there are identical package descriptions in 10.4/stable and 10.4/unstable. - -- Alexander Hansen, Ph.D. Fink User Liaison http://finkakh.wordpress.com/ -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.11 (Darwin) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAk6AZSsACgkQB8UpO3rKjQ/q+ACgl9ypqiUsoBC5keoMGpwdmkLs X1gAnROZ+jE8vAFIpUD5Owjwl7fAk5cE =vaXy -----END PGP SIGNATURE----- ------------------------------------------------------------------------------ All the data continuously generated in your IT infrastructure contains a definitive record of customers, application performance, security threats, fraudulent activity and more. Splunk takes this data and makes sense of it. Business sense. IT sense. Common sense. http://p.sf.net/sfu/splunk-d2dcopy1 _______________________________________________ Fink-devel mailing list Fink-devel@lists.sourceforge.net List archive: http://news.gmane.org/gmane.os.apple.fink.devel Subscription management: https://lists.sourceforge.net/lists/listinfo/fink-devel
