What percentage of code in a Free and Open Source software project should be written by the core maintainers?
-
What percentage of code in a Free and Open Source software project should be written by the core maintainers?
@evan i think a lower % can be a positive sign for the health of the project (lots of contributors, functioning line of succussion when maintainers move on), though it can mean a lot of things
-
@evan as opposed to contributors? Or bots?
@manchicken or aliens!
-
@manchicken or aliens!
@evan I would genuinely love to meet an alien. Even if they have to kill me, I’d love to chat with them.
-
@evan I would genuinely love to meet an alien. Even if they have to kill me, I’d love to chat with them.
@manchicken make some software they would like with an annoying but easy-to-fix bug!
-
@dneary This is why you're an Open Source guru, Dave.
@evan I'm no guru, Evan! "Self-described guru" maybe. Or "The Forrest Gump of open source" maybe - like James Monroe, always in the scene for historic events. Rarely the main character, always an extra.
-
What percentage of code in a Free and Open Source software project should be written by the core maintainers?
@evan I reject the premise of the poll.
I'd like core maintainers to supply project direction and consistent purpose.
-
What percentage of code in a Free and Open Source software project should be written by the core maintainers?
@evan it depends. I voted 25-50 because that's where I imagine medium sized projects which aren't thrown-over-the-wall and have some impact on society should probably be.... But anywhere from 1 to 100 based on the type of program, complexity of the code and approachability of the core team. Basically a big fat "it depends", but I know I can't vote for that, so I didn't.
-
@evan I reject the premise of the poll.
I'd like core maintainers to supply project direction and consistent purpose.
Completely agree. The percentage depends a lot on both the age and scale of the project.
If a project has 100 active contributors, the mean contribution of new code will be 1% per contributor. How many of those are core? Maybe 10 or 20? So maybe there each writing 2-3% of new code and you have a decaying distribution? But that’s a moderately large project.
One of my projects started at 100% code written by me, but we did a release last year where none of the new code was written by me (it was all reviewed by me) and even the release was done by someone else, I just approved it. That’s a project with 2-3 moderately active contributors and a load of tiny drive-by contributors.
-
@manchicken make some software they would like with an annoying but easy-to-fix bug!
@evan I’m pretty sure I already have.
-
@evan I reject the premise of the poll.
I'd like core maintainers to supply project direction and consistent purpose.
@LyallMorrison what premise are you talking about? That code exists? That it is written by someone? The idea of percentages?
-
Completely agree. The percentage depends a lot on both the age and scale of the project.
If a project has 100 active contributors, the mean contribution of new code will be 1% per contributor. How many of those are core? Maybe 10 or 20? So maybe there each writing 2-3% of new code and you have a decaying distribution? But that’s a moderately large project.
One of my projects started at 100% code written by me, but we did a release last year where none of the new code was written by me (it was all reviewed by me) and even the release was done by someone else, I just approved it. That’s a project with 2-3 moderately active contributors and a load of tiny drive-by contributors.
@david_chisnall @LyallMorrison so, which of those is best? For code quality, wellbeing of contributors, sanity and satisfaction of the maintainers?
Poll FAQ
I do a lot of polls on my account on the Fediverse. I get the same questions or requests multiple times, so I made this FAQ to make it easier to reply. Q: Why do you do so many polls? A: I like to think about topics big and small, from the things we wear…
Evan Prodromou's Blog (evanp.me)
-
@LyallMorrison what premise are you talking about? That code exists? That it is written by someone? The idea of percentages?
@evan that core maintainers must write more than 0% of the code.
I'm being unnecessarily sassy about it, but I subscribe to a software development lifecycle where writing as little code as possible (but no less) is a virtue.
If a project maintainer _merely_ maintains scope, plans release timelines, manages technical contributors, writes comms and documentation, enforces quality standards... that's plenty of software development.
Or from another perspective, a healthy project should still be viable if they lose core maintainers for whatever reason. If there's only one person who could possibly maintain part of the project that's a red flag.
-
@evan that core maintainers must write more than 0% of the code.
I'm being unnecessarily sassy about it, but I subscribe to a software development lifecycle where writing as little code as possible (but no less) is a virtue.
If a project maintainer _merely_ maintains scope, plans release timelines, manages technical contributors, writes comms and documentation, enforces quality standards... that's plenty of software development.
Or from another perspective, a healthy project should still be viable if they lose core maintainers for whatever reason. If there's only one person who could possibly maintain part of the project that's a red flag.
@LyallMorrison lol. Someone grilled me about having 0-25% in my last poll. Anyway, next time I'll include 0 for you.
-
@david_chisnall @LyallMorrison so, which of those is best? For code quality, wellbeing of contributors, sanity and satisfaction of the maintainers?
Poll FAQ
I do a lot of polls on my account on the Fediverse. I get the same questions or requests multiple times, so I made this FAQ to make it easier to reply. Q: Why do you do so many polls? A: I like to think about topics big and small, from the things we wear…
Evan Prodromou's Blog (evanp.me)
Each one, at different sizes and ages of a project.
-
R relay@relay.infosec.exchange shared this topic
