Hi Paul, On Sat, Mar 26, 2022 at 8:54 PM Paul Wise <p...@debian.org> wrote: > > Debian's relationship with the various distributions derived from > Debian and approach to existing and new derivatives has had a wide > range of states. Most derivatives recieve indifference from Debian. > There has been animosity from Debian towards some derivatives. We have > welcomed the creation of derivatives. We have welcomed developers from > derivatives into Debian packaging teams. We have encouraged people to > start blends within Debian instead of starting derivatives. > > What do you think of Debian's current relationship with derivatives? > > What would you like to change about our relationships? > > What do you feel Debian's current approach to derivatives is? > > What would you like to change about that approach? > > What is your favourite derivative? > > Would you like to see it merged into Debian? > > Thoughts on identical Debian pure blends vs derivatives? > > Other thoughts welcome, as are thoughts from non-candidates.
First off, I am proud that there are a lot of derivatives. Last time I checked, over half of the entries on DistroWatch were based on Debian. [1] "Imitation is the sincerest form of flattery." Meanwhile, it is helpful to look at the individual heritage for each project. Here, I will focus on those that either (1) want to be different or (2) those who cannot achieve in Debian what they would like to accomplish. There is some overlap because Debian is an odd place to get anything done. In the first category, I see many of the desktop alternatives, like "Mint" and "MX". They love their independence. Some lesser known projects like Endless rely entirely on their own brand. I do not think we can do much about them other than learn from their marketing and maybe adopt some of their tools (although some of those world travellers may join us when we find an attractive long-term home for Debian). By the way, I know the desire to be different. As an early Linux user in 1993, I briefly considered starting my own distribution, called "Felix," but then got busy with school and stuck to Slackware. (I switched to Debian in 1996 or so.) In retrospect, I am glad that Ian Murdock (and Deb) got their names on our baby. I am incredibly proud to be a part of Debian today! The second category is populated by projects we know a lot better. Usually, there was some kind of a split. Among those, I count Devuan (initd) and Ubuntu (release cycle). They exist mostly because we can be jerks—although maybe the latter was not so much of an issue for Ubuntu. I used Ubuntu very happily and productively without contributing from 2005 to 2010, but their frequent release cycle was a burden for me. The day I switched back to tracking testing was one of the happiest days in my life. Otherwise, I have limited experience with our derivatives. If it's okay to broaden the scope of your questions for a moment, I love Arch and would like to merge with them (although I have never actually used their operating system). They have great energy, superb documentation, and are a lot younger. On the flip side, Arch would benefit from our technical acumen and our packaging experience. I'd love to see us join forces with them somehow. The pure blends question should really go to the folks on this list who actively maintain derivatives. Are human factors in the way where technical solutions exist? Let's come together! To improve our relationships with all derivatives—you might have guessed it—I would form an Integration Committee (aka as the "come home" committee). As the first member, I would approach Mark Hindley. Finally, I would like to draw some specialty derivatives nearer to Debian, although many are super friendly already. In the embedded space, for example, OpenWrt needs help with their archive. For unreleased builds, one has to reflash their base system when adding packages later. Maybe we can give them a snapshot.openwrt.org with old builds sorted by date. Thank you for a question with great peacemaking potential! Kind regards, Felix Lechner [1] Sorry, I can't find the reference right now.
