On Sat, Sep 02, 2023 at 10:19:54AM +0200, Richard Brown wrote:
Hi all,
I've been looking at the results from the recent contributor survey to gauge the interest and feasibility of replacing openSUSE Leap with a new community-built offering.
For those who may not have been keeping up with the efforts, their are proposals to build two very different distributions to replace Leap:
"Linarite" - a regular old fashioned release desktop distribution, likely with a narrower package selection than we're used to with Leap unless we find significantly more contributors to be able to support everything
"Slowroll" - a derivative of Tumbleweed, built automatically as much as possible, using automation and metrics to copy packages from Tumbleweed only after certain conditions (max age, X weeks without change, etc). Basically an attempt to provide something less scary than full speed Tumbleweed.
Rather than bombard everyone with all the data, I've decided to take the approach of asking specific questions and seeing what data from the survey best helps answer that question.
--- Q1: Did the survey reach a representative sample of our contributor base?
A: The survey received 251 responses in total, 101 from people who self identified as a 'Contributor', either to the distributions or the Project more broadly. 72 identified as a distribution contributor, which is larger than the 61 people who submitted packages to Leap or Backports for the last release. Therefore I think it's clear the survey reached a broad enough audience to draw some meaningful conclusions from.
--- Q2: Between "Linarite", "Slowroll" and not replacing Leap, which is the most popular with our contributors?
A: Not replacing Leap was the most popular option when distribution contributors were asked which options would they contribute towards (39%)
Not replacing Leap/using Tumbleweed was the most popular option when distribution contributors were asked which would they use on their desktop/laptops (54%)
Those same contributors did collectively suggest that Linarite should be the direction of openSUSE (37%) and what they'd use on their servers (43%)
Q3. Between "Linarite", "Slowroll" and not replacing Leap, which is the most popular with our users?
Slowroll was the most popular option when users were asked which options they'd contribute towards (28%). Linarite was 3rd (19%) after not replacing Leap (25%)
Slowroll was also the #1 option when users were asked which option was the best direction for openSUSE (39%) and which would they use on a server (48%).
Just using Tumbleweed was also the most popular option with our users for their desktop/laptops (41%), followed by Slowroll (36%)
Linarite was not the #1 choice for our users in any of the questions asked.
Q4. Do we have enough contributors/Can we get enough contributors to make any option work?
24 distribution contributors said they would contribute to Linarite 13 said they would contribute to Slowroll 49% said they wouldn't contribute to any Leap replacement
26 users said they would contribute to Linarite 38 said they would contribute to Slowroll 53% said they wouldn't contribute to any Leap replacement
--- Some thoughts/analysis
Our users seem to overwhelmingly favour rolling releases, with 51%-64% expressing a preference for either Tumbleweed or Slowroll regardless of whether they're being asked if they'll use it for Server, Desktop, or whether they'll contribute or think it's the best direction for the project.
This preference increases when contributors are asked, with the preference ranging between 55%-71% depending on the question.
In the light of these results, it is my suggestion to the community that if we are to build something to replace Leap, then the option we should focus on is Slowroll.
It is the most popular with our users, and the option more closely aligned to what our contributors use themselves.
I also think its more important for the Leap replacement to focus on Desktop use cases, as openSUSE will also be hosting 1:1 copies of SUSE's ALP products. Those products should be awesome for folk who want Enterprise-like server distros.
Given that it seems silly to spread ourselves too thin trying to make a Leap replacement that is both a Server and Desktop OS.
That said though, I am still concerned that we do not have enough contributors to make any Leap replacement viable.
Leap has struggled even with 61 folk contributing directly to the codebase and backports/PackageHub. And this is when we've had the SLE codebase to borrow from, which dramatically reduced the work required.
Either Slowroll or Linarite would require considerably more packaging and maintenance work than Leap.
Hello, why do you think that Slowroll would require more packaging effort than Leap? Just as Leap derives from SLE, Slowroll derives from Tumbleweed. I could even argue that Slowroll would require less effort when all packages come from Tumbleweed while Leap is built on top of SLE and has a large number of packages of its own. Thanks Michal