5 ms·

Only certain FOSS is worth working on for me. Here's my heuristic as of today. * Don't work on FOSS that supports a SaaS or proprietary product. You're enrichi

Only certain FOSS is worth working on for me. Here's my heuristic as of today. * Don't work on FOSS that supports a SaaS or proprietary product. You're enriching someone else's ecosystem for no money. You take on the communication bandwidth and issue load of dealing with not only your issues but their issues too. Your software is useless if theirs fails. It's thankless. * Don't work on FOSS you aren't using yourself anymore. This heuristic isn't fully fleshed yet, but I've found if I stop using software I wrote, it just becomes a support burden for me. Especially frustrating are entitled users or entitled users who make money entirely using my software who want niche issues fixed. Not PRs, they want me to debug it all or put in the thought to get the features working. (Probably doesn't apply if you're immersed socially with a friend group working on a FOSS project though)

psychstudio 5y ago
Also: entitled users who ask if x can be done, and after a comprehensive answer which took a lot of time and thought say "can you make a demo and post the code. thanks"
- tonyedgecombe 5y ago
  To be fair that isn't limited to open source software, I get plenty of that with my commercial software as well.
- rs999gti 5y ago
  Why not reverse bounty or charge these entitled users for fixes and change requests? If they want the patch or improvement, they should pay for it. Log4J CVE might have been the perfect use case for asking for 50K-1M EURO for the maintainers to patch the bug, to any corporate clients this is a drop in the bucket.
  - voakbasda 5y ago
    That starts to sound suspiciously like an extortion racket.
    - LordDragonfang 5y ago
      Capitalism is just one big extortion racket. Pending to move to fully automated luxury space communism, that's what people need to do to succeed in the economic incentive structure we've created.
ChrisMarshallNY 5y ago
One skill that I’ve learned, over the years, is to say “no,” in a calm, respectful manner. Works a charm. The “calm and respectful” part is important. I can be right ornery, and learned, the hard way, that OS projects are more fragile than commercial ones, so enemies can do a lot of damage. I have written software for a fairly … pithy … demographic, and have dealt with feedback that would make many FOSS authors hide under their standing desks, whimpering.
- cobertos 5y ago
  I've not found a good heuristic for saying "no". Most issues are valid issues. I can see the benefits, they might even benefit future me. The hardest part for me is actually saying no, because it's never that black/white. I might work on it later because I want to help. I might stumble into a use case where I now need this feature and then do it. Someone might make a PR. But each "open ended issue" is still bandwidth, brain capacity, effort.
  - ChrisMarshallNY 5y ago
    I've learned the skill. I had to. However, you are correct about that "using bandwidth" thing. If we publish something, then I think it behooves us to make it clear what users/adopters of our work can expect from us. I tend to use the MIT license ("You can do what you want, but don't come crying to me."). I also make every effort to make the project as understandable and well-documented as possible. I appreciate feedback, and apply it, if I deem it fitting. I have only had one project "go viral," so to speak. It's a fairly big deal, nowadays. It was taken over by a new team, and they had very little trouble getting up to speed on it, because of the quality of the code and documentation. Thousands of folks around the world, use it every day. It's become a fairly vital infrastructure component, for a certain demographic. The best thing I ever did, with that project, was get the hell away from it, as soon as I was confident the new team was good.
KyleJune 5y ago
In response to your first point, I'd rather use and contribute to FOSS that supports a SaaS or proprietary product. Issues are more likely to be addressed if the maintainers depend on the software themselves for profit. I see it as trading helping them improve their software for the ability to use it in my own projects. It also gives some reassurance that the FOSS is going to be maintained if the creators depend on it financially. I agree with your second point though. The main reason I see for contributing to FOSS is to fix issues that affect me or creating issues that may lead to the maintainers improving the software in a way that would benefit me. I think it makes sense for companies to open source tools they depend on. They help others by providing the free software and in return they get help identifying and fixing bugs.
- southerntofu 5y ago
  > I'd rather use and contribute to FOSS that supports a SaaS or proprietary product. Issues are more likely to be addressed I understand the logic behind this assertion, but i don't believe it holds true at all. Businesses will favor fixing bugs related to main features that drive revenue in. Whatever's not hurting their business is invisible to them.
southerntofu 5y ago
> Don't work on FOSS that supports a SaaS or proprietary product. So true. I'd expand on that with a more specific example: game modders. When i look at all the energy spent on game mods which bring significant (gratis) value to commercial games, i can't help but wonder what the FOSS gamedev ecosystem would look like if all those dedicated people put their energy on free-software ecosystem. Minecraft for example made a fortune on promises of later open-sourcing the game and deriving value from all the volunteers building mods.
- js8 5y ago
  While I agree with the broader point, I don't think Minecraft example is accurate. Historically, most MC mods were freeware closed-source, only few years back (around 1.10) lot of them became open-sourced. So I don't think open source was a big consideration for modders.
  - southerntofu 5y ago
    You are correct. I meant that Mojang was saying they would consider open sourcing Minecraft itself when it's ready and has raised sufficient funding (have we reached this point yet?). My point was that modders are providing significant value for free to private companies (and are happy to do that as lon as they're credited) and i wish such energy was directed towards a FOSS game ecosystem instead of continuing to enrich the same crooks who derive money off dark patterns (subscriptions, lootboxes) and exploited workers (gamedev industry is famously harsh). Also, each software has its guidelines to build mods. I feel like if a game had modding guidelines suggesting to open source, modders would be more than happy to do so; if not, open-sourcing a single-person project can be considered more work (eg. setup git).
    - mathteddybear 5y ago
      I believe the phrase was "Once sales start dying" though Anyways, while there are some similarities shared between Open Source and video game modding scenes, they are substantially different. Instead of evidence to back up this claim, I implore you to imagine what it would be like if some wrote an article titled "Is It Even Worth Working on Videogame Mods Anymore?".
mpol 5y ago
> Don't work on FOSS you aren't using yourself anymore. I think there is still some middleground possible. It can still be enjoyable doing small support stuff for it and to keep it running and fixing small bugs, without any big rewrites or big features. I once cared about those things and I might still do. Just keeping it running might feel better than discarding it. It all depends on personal preferences, the type of users and their communication, if you still use the language, etc.
- cobertos 5y ago
  Yeah, I agree. I think https://news.ycombinator.com/item?id=29673524 https://news.ycombinator.com/item?id=29673524 captures it well
csdummi 5y ago
dspillett 5y ago
> Don't work on FOSS you aren't using yourself anymore. More-or-less agreed. > This heuristic isn't fully fleshed yet I'd perhaps extend it as: “Don't work on FOSS you or people you sufficiently care about aren't using anymore.” Who qualifies as those you care about and why is an entirely personal thing and I'll not judge anyone's choice there. They could be friends, they could represent a cause you support, they could be randoms and you just get off on feeling helpful, they could be important and you like to be associated, you could be being paid to care, etc. Also it goes without saying, but I'll say it anyway as I'm in a wordy mood, if you are enjoying working on something even if you don't use the output I'd still consider that to be finding some use in what-ever it is.
- cobertos 5y ago
  Yeah, exactly this. Enjoyment is a good signal for whether youre using your time in a way your brain tolerates. I personally strike a balance between maximizing happiness and also maximizing value, so when enjoyment dips, it takes a bit for me to get off a project still.
adam_arthur 5y ago
Or just wait for a SaaS to start using your project and then change the license to monetize. Lol!
chongli 5y ago
That first heuristic is very difficult. It means you can’t work on the Linux kernel or pretty much anything that could be used for server infrastructure. It also means you’re very restricted in license choices since you otherwise never know if your project might get picked up in the future and used in a proprietary service. This likely means you have to use something like AGPL.
- Retric 5y ago
  It’s fine if someone picks up and uses a project I worked on. That’s what OSS is for, but the Linux kernel doesn’t need my support to keep going as it’s self sustaining at this point. Plenty of other projects are in serious need of active support and keeping them afloat is simply a better use of my time.
  - chongli 5y ago
    Having your project picked up by commercial users is entirely the problem this heuristic is intended to avoid. It’s also the central problem from the linked article: commercial users take from open source projects and often don’t give anything back. They also tend to externalize any costs they have from use of the project onto the project maintainers. Maybe you’re okay with being treated the way the log4j maintainers are being treated right now. The whole point is that you shouldn’t be! People producing valuable software in their free time and giving it away for free should be treated with respect and that’s not happening right now. We need to find a solution!
    - Retric 5y ago
      “Being treated” implies a lot of things. Someone using something I have written doesn’t cost me anything, but it also doesn’t create any obligations for me. I don’t need to make my contact info public, or have any interactions with users whatsoever. log4j accepted donations which is really it’s own thing. Running a charity, which is what it means to get paid full time to maintain OSS, has a host of issues that have little to do with OSS and a lot to do with public relations.
- 5y ago
throwaway984393 5y ago
I learned long ago that I don't want to waste my life doing something I don't care about. I publish things as open source that I wrote because I needed them, and nothing existed like it. One of those things, other people actually use! I don't so much, but I enjoy keeping it going every so often because it gives me purpose. Another of those things, literally nobody seems to know about, even though it's a very useful thing which there is no oss alternative for. I'm a little sad nobody uses it, but it's fine because I still use it. When I stop using these completely, I'll shutter them. Somebody can fork them if they want.

psychstudio 5y ago

Also: entitled users who ask if x can be done, and after a comprehensive answer which took a lot of time and thought say "can you make a demo and post the code. thanks"

tonyedgecombe 5y ago
To be fair that isn't limited to open source software, I get plenty of that with my commercial software as well.
rs999gti 5y ago
Why not reverse bounty or charge these entitled users for fixes and change requests? If they want the patch or improvement, they should pay for it. Log4J CVE might have been the perfect use case for asking for 50K-1M EURO for the maintainers to patch the bug, to any corporate clients this is a drop in the bucket.
- voakbasda 5y ago
  That starts to sound suspiciously like an extortion racket.
  - LordDragonfang 5y ago
    Capitalism is just one big extortion racket. Pending to move to fully automated luxury space communism, that's what people need to do to succeed in the economic incentive structure we've created.

ChrisMarshallNY 5y ago

One skill that I’ve learned, over the years, is to say “no,” in a calm, respectful manner. Works a charm. The “calm and respectful” part is important. I can be right ornery, and learned, the hard way, that OS projects are more fragile than commercial ones, so enemies can do a lot of damage. I have written software for a fairly … pithy … demographic, and have dealt with feedback that would make many FOSS authors hide under their standing desks, whimpering.

cobertos 5y ago
I've not found a good heuristic for saying "no". Most issues are valid issues. I can see the benefits, they might even benefit future me. The hardest part for me is actually saying no, because it's never that black/white. I might work on it later because I want to help. I might stumble into a use case where I now need this feature and then do it. Someone might make a PR. But each "open ended issue" is still bandwidth, brain capacity, effort.
- ChrisMarshallNY 5y ago
  I've learned the skill. I had to. However, you are correct about that "using bandwidth" thing. If we publish something, then I think it behooves us to make it clear what users/adopters of our work can expect from us. I tend to use the MIT license ("You can do what you want, but don't come crying to me."). I also make every effort to make the project as understandable and well-documented as possible. I appreciate feedback, and apply it, if I deem it fitting. I have only had one project "go viral," so to speak. It's a fairly big deal, nowadays. It was taken over by a new team, and they had very little trouble getting up to speed on it, because of the quality of the code and documentation. Thousands of folks around the world, use it every day. It's become a fairly vital infrastructure component, for a certain demographic. The best thing I ever did, with that project, was get the hell away from it, as soon as I was confident the new team was good.

KyleJune 5y ago

In response to your first point, I'd rather use and contribute to FOSS that supports a SaaS or proprietary product. Issues are more likely to be addressed if the maintainers depend on the software themselves for profit. I see it as trading helping them improve their software for the ability to use it in my own projects. It also gives some reassurance that the FOSS is going to be maintained if the creators depend on it financially. I agree with your second point though. The main reason I see for contributing to FOSS is to fix issues that affect me or creating issues that may lead to the maintainers improving the software in a way that would benefit me. I think it makes sense for companies to open source tools they depend on. They help others by providing the free software and in return they get help identifying and fixing bugs.

southerntofu 5y ago
> I'd rather use and contribute to FOSS that supports a SaaS or proprietary product. Issues are more likely to be addressed I understand the logic behind this assertion, but i don't believe it holds true at all. Businesses will favor fixing bugs related to main features that drive revenue in. Whatever's not hurting their business is invisible to them.

southerntofu 5y ago

> Don't work on FOSS that supports a SaaS or proprietary product. So true. I'd expand on that with a more specific example: game modders. When i look at all the energy spent on game mods which bring significant (gratis) value to commercial games, i can't help but wonder what the FOSS gamedev ecosystem would look like if all those dedicated people put their energy on free-software ecosystem. Minecraft for example made a fortune on promises of later open-sourcing the game and deriving value from all the volunteers building mods.

js8 5y ago
While I agree with the broader point, I don't think Minecraft example is accurate. Historically, most MC mods were freeware closed-source, only few years back (around 1.10) lot of them became open-sourced. So I don't think open source was a big consideration for modders.
- southerntofu 5y ago
  You are correct. I meant that Mojang was saying they would consider open sourcing Minecraft itself when it's ready and has raised sufficient funding (have we reached this point yet?). My point was that modders are providing significant value for free to private companies (and are happy to do that as lon as they're credited) and i wish such energy was directed towards a FOSS game ecosystem instead of continuing to enrich the same crooks who derive money off dark patterns (subscriptions, lootboxes) and exploited workers (gamedev industry is famously harsh). Also, each software has its guidelines to build mods. I feel like if a game had modding guidelines suggesting to open source, modders would be more than happy to do so; if not, open-sourcing a single-person project can be considered more work (eg. setup git).
  - mathteddybear 5y ago
    I believe the phrase was "Once sales start dying" though Anyways, while there are some similarities shared between Open Source and video game modding scenes, they are substantially different. Instead of evidence to back up this claim, I implore you to imagine what it would be like if some wrote an article titled "Is It Even Worth Working on Videogame Mods Anymore?".

mpol 5y ago

> Don't work on FOSS you aren't using yourself anymore. I think there is still some middleground possible. It can still be enjoyable doing small support stuff for it and to keep it running and fixing small bugs, without any big rewrites or big features. I once cared about those things and I might still do. Just keeping it running might feel better than discarding it. It all depends on personal preferences, the type of users and their communication, if you still use the language, etc.

cobertos 5y ago
Yeah, I agree. I think https://news.ycombinator.com/item?id=29673524 https://news.ycombinator.com/item?id=29673524 captures it well

csdummi 5y ago

dspillett 5y ago

> Don't work on FOSS you aren't using yourself anymore. More-or-less agreed. > This heuristic isn't fully fleshed yet I'd perhaps extend it as: “Don't work on FOSS you or people you sufficiently care about aren't using anymore.” Who qualifies as those you care about and why is an entirely personal thing and I'll not judge anyone's choice there. They could be friends, they could represent a cause you support, they could be randoms and you just get off on feeling helpful, they could be important and you like to be associated, you could be being paid to care, etc. Also it goes without saying, but I'll say it anyway as I'm in a wordy mood, if you are enjoying working on something even if you don't use the output I'd still consider that to be finding some use in what-ever it is.

cobertos 5y ago
Yeah, exactly this. Enjoyment is a good signal for whether youre using your time in a way your brain tolerates. I personally strike a balance between maximizing happiness and also maximizing value, so when enjoyment dips, it takes a bit for me to get off a project still.

adam_arthur 5y ago

Or just wait for a SaaS to start using your project and then change the license to monetize. Lol!

chongli 5y ago

That first heuristic is very difficult. It means you can’t work on the Linux kernel or pretty much anything that could be used for server infrastructure. It also means you’re very restricted in license choices since you otherwise never know if your project might get picked up in the future and used in a proprietary service. This likely means you have to use something like AGPL.

Retric 5y ago
It’s fine if someone picks up and uses a project I worked on. That’s what OSS is for, but the Linux kernel doesn’t need my support to keep going as it’s self sustaining at this point. Plenty of other projects are in serious need of active support and keeping them afloat is simply a better use of my time.
- chongli 5y ago
  Having your project picked up by commercial users is entirely the problem this heuristic is intended to avoid. It’s also the central problem from the linked article: commercial users take from open source projects and often don’t give anything back. They also tend to externalize any costs they have from use of the project onto the project maintainers. Maybe you’re okay with being treated the way the log4j maintainers are being treated right now. The whole point is that you shouldn’t be! People producing valuable software in their free time and giving it away for free should be treated with respect and that’s not happening right now. We need to find a solution!
  - Retric 5y ago
    “Being treated” implies a lot of things. Someone using something I have written doesn’t cost me anything, but it also doesn’t create any obligations for me. I don’t need to make my contact info public, or have any interactions with users whatsoever. log4j accepted donations which is really it’s own thing. Running a charity, which is what it means to get paid full time to maintain OSS, has a host of issues that have little to do with OSS and a lot to do with public relations.
5y ago

throwaway984393 5y ago

I learned long ago that I don't want to waste my life doing something I don't care about. I publish things as open source that I wrote because I needed them, and nothing existed like it. One of those things, other people actually use! I don't so much, but I enjoy keeping it going every so often because it gives me purpose. Another of those things, literally nobody seems to know about, even though it's a very useful thing which there is no oss alternative for. I'm a little sad nobody uses it, but it's fine because I still use it. When I stop using these completely, I'll shutter them. Somebody can fork them if they want.