Hi Yaron, On Wednesday 09 February 2005 07:11, Yaron Tausky wrote: > Therefor, I would like to suggest that > each ebuild would include a "Testing Period Expiration Date" -- a variable > containing a date after which the ebuild would automatically considered > stable, unless the maintainer chose otherwise.
I can understand the temptation, but the risk of breaking live systems is just too high. It only takes one widely used package to be broken by this to result in unhappy users and bad press for Gentoo. > This will take a lot of work > off the hands of package maintainers -- instead of having to look after > every ebuild and mark it stable, they would only have to take care of > malfunctioning ones, yet it would still allow a proper testing period > before release. A "proper testing period" needs to be measured in terms of the amount of testing activity, not simply elapsed time. There's no point in marking something stable after 30 days if hardly anyone has used it. What we really need are automated test suites, covering at least three areas: - core functionality - bugs filed and marked as fixed in bugzilla - upgrades from one version to the next If a new version of a package can pass at least the same tests that the last stable version did, then at least we'd have a sane basis for automatically marking stuff as stable. Best regards, Stu -- Stuart Herbert [EMAIL PROTECTED] Gentoo Developer http://www.gentoo.org/ http://stu.gnqs.org/diary/ GnuPG key id# F9AFC57C available from http://pgp.mit.edu Key fingerprint = 31FB 50D4 1F88 E227 F319 C549 0C2F 80BA F9AF C57C -- -- [email protected] mailing list
