How to improve openSUSE?
Well I guess growbag meant the actual distro, but openSUSE is more than just the distro. The social/infrastructure of openSUSE is improving and that should be commended. Merging the forums, forming the Board, forming the openSUSE Boosters, revamping the wiki…all of these are good moves.
Looking at the distro itself, a number of my personal gripes have already been fixed but to contribute to the discussion…
-
I agree that a simpler USB media experience would be ideal moving forward, especially as more and more computers lack disc drives.
-
I’d love to see something akin to Ubuntu’s “paper cuts” project. Although as a tester I try to find a report small, easily fixed, things that would make the distro better (OpenOffice now has its grammar-checker installed by default for example) there are sure to be many things that I miss. I’d like to see a more formal effort in this regard.
-
Simplify the OBS repo structure. This is already underway in some corners…The KDE branch is a mess of options that could be pared down (this is happening by the way).
-
Continue the new version flexibility on bug-fix releases. OS 11.2 saw the standard update repo upgrade users from KDE 4.3.1 to 4.3.5, with similar moves for GNOME. In my opinion this makes perfect sense, if upstream is making the shipped software more stable then we should pass those bug-fixes on to users.
-
Move to an annual release with a mid-year bug-fix release. I’m not actually a die-hard proponent for this, but I do like it. I think there could be some advantages in say releasing a major version every May, and then every November releasing a respin that was the same just with all the patches to date. I believe most major upstream software is mature enough to make this feasible.
This has a number of repercussions:
PROS
a. Conservative users get their “super-stable” release which will include all the official bug fixes from GNOME & KDE along with the squashing of most/all major bugs that may have cropped up in wide testing. People who have bandwidth limitations would also benefit from a ‘maximum fixes included’ iso being available.
b. Gives openSUSE consistent release months that everyone knows rather than the current tumbling system.
c. Buys the artists, programmers and testers a longer period of time to work on things for each major release.
d. Gives all the support mechanisms a longer shelf life. Forum helpers don’t need to shift their support target as quickly, manuals and wiki pages stay valid for longer periods of time, etc. etc.
e. Given the support commitment of “two releases plus two months” the support cycle would expand from 18 months to 26 months - the respin doesn’t count as a proper release since it just a collection of bug fixes for the major release. This would reduce the feeling of being on an “upgrade treadmill”.
CONS
a. There’s supposedly a large segment of Linux users that want new software but are only willing to get it through brand-new releases of the whole distro.
I think this segment is misunderstood. The users who need new major releases to get new software are using the major releases to get around their frustration with distros not providing new software via online updates. The presence of the OBS means anyone who really wants the newest Chromium/GNOME/Amarok/whatever can get it - what we need to do is make the OBS more discoverable.
b. openSUSE will miss a major update to GNOME and KDE.
I don’t think this is that big a deal. The current system already guarantees that major versions will be missed (for example OS 11.3 will have KDE 4.4.x while OS. 12.0 will have KDE 4.6.x).