7 ms·

Lessons I learned from working full-time on a FOSS project for 503 days

140 points by willm 3y ago

KiranRao0 3y ago
To me, these lessons feel very similar to the lessons learned working at a midsized company. Was there any particular takeaways that are exclusive to working at a FOSS?
- magicalhippo 3y ago
  While I agree that it's not terribly unique to FOSS, it makes it even more useful IMO since relatively few work on FOSS. I think the lessons learned and advice given are great for a large number of developers.
- simonw 3y ago
  The lessons about working with external pull requests feel unique to FOSS to me.
quantum_state 3y ago
Very valuable … thanks for sharing
candiddevmike 3y ago
I looked all over Textuals website and I have no idea how they make money. Donations? Consulting?
- smokel 3y ago
  Ads in TUIs are an underdeveloped market, I suppose.
- CubsFan1060 3y ago
  I think the eventual plan might be this? https://textual.textualize.io/blog/2023/09/06/what-is-textual-web/ https://textual.textualize.io/blog/2023/09/06/what-is-textua...
- zeckalpha 3y ago
  Why must everything make money?
  - lmz 3y ago
    Usually people don't have enough savings to dedicate themselves full time to a non money making project.
    - hackernewds 3y ago
      if the project pays them, it works
  - tsuica 3y ago
    Because people have to eat.
  - 7bit 3y ago
    Why do you not work for free?
    - zarathustreal 2y ago
      Contributing to OSS is a hobby, not a job. You explicitly work for free on purpose. Whether or not you’re able to pay your bills is unrelated to the OSS contribution, it’s very clear that you’re not going to get paid for writing code that you’re giving away for free.
      - willm 2y ago
        You only had to read the title to know that the author of the post was paid for 503 days to work on OSS. It was literally his job.
        zarathustreal 2y ago
        An example of one person being employed and contributing to open source as part of his job is not a refutation of the statement. You don’t have to read anything to know that, just use logic. To provide a counter-example you’d need to reference an open-source project that makes money selling the source code which, hopefully obviously, doesn’t exist, by definition
- SushiHippie 3y ago
  https://www.willmcgugan.com/blog/tech/post/from-open-source-maintainer-to-founder-ceo/ https://www.willmcgugan.com/blog/tech/post/from-open-source-... > Additionally, I would work on a cloud service using both those projects which at some point would become a business. > Now this was a good plan. The money from GitHub sponsors which I had previously been donating to charity would take the sting out of not having an income for a while. While far from a salary (where I live), at around $1000 a month it would help pay a few bills. > I have far more confidence in that vision now that I have raised seed funding. Predicated on that cloud service, I have enough funding to hire a few developers to work on it with me. Last month I was an unemployed code monkey, today I am a Founder / CEO of Textualize. Go figure.
  - candiddevmike 3y ago
    Good find! Getting seed funding to try and monetize a bunch of MIT libraries using a "cloud service". Wild. Wonder if OP was always 100% dedicated to FOSS for 503 days or if they worked on the "cloud service"...
jacksonhacker 3y ago
[dead]
bionhoward 3y ago
I love Rich! Using it for tree_plus and other projects, and it’s beautiful and hackable. Great Python package for terminal visuals. Definitely recommend it if you have a Python CLI, can really improve readability and visual niceties.
righthand 2y ago
This is all pretty ideal and I think these outcomes are more likely if the project has a substantial base. Comparatively I was trying to contribute to another FOSS project only to get push back. I started contributing to the docs to build the package which were way out of date. The project switched from qmake to cmake. The project did not have a correctly working cmake setup so I assisted with that. Every time I contributed the maintainers would check my PR then ping another maintainer who was supposedly working on the specific issues. Eventually I managed to get some docs and code merged but after that the maintainers returned harder to blocking my PR’s with their own work which they happily merged in. They would not update my PR that they had already gone in and solved the problem, I had to repull and figure it out myself, which left me with open PRs/tickets just lingering in the space. They would happily clarify my questions with obscure non-revealing conversation about their project. Eventually I gave up. Similarly I have tried to contribute to things like KDE in the past as well and been met with similar walls on contributing. Some developer somewhere is in charge of a FOSS project as a hobby and only really cares about their use cases or providing a simple demo app or rewriting an old app for their needs because the old app was too complicated or probably run by another hobbyist maintainer. This leads to opening tickets and looking for ways to contribute or discuss implementation with little to no response. Or in the case of KDE, having your ticket closed and shifted over to their bug system which requires rediscussion and retriaging. At some point you just lose momentum and interest trying to push a project along due to the sheer uneventfulness and refocusing of it all. I think we need fewer hobbyist project owners and more hobbyist project managers.
- CoastalCoder 2y ago
  I had similar frustrations with a well-known deep-learning framework. I'd get well-vetted PR's for the build system, CI system, etc. committed. Only to have them later undone, without discussion, by people in the inner circle. It doesn't take many experiences like that to lose all motivation. I think some open-source projects deceive themselves (and therefore, outsiders) about their contribution processes.
  - ozim 2y ago
    Not wondering people rather start their own thing rather than contribute to someone else projects. Cambrian explosion of JavaScript frameworks explained. For me at least as I never really tried to contribute to any open project.
RojerGS 2y ago
The title of this posting was edited and no longer matches the title of the article linked. I wonder if anybody knows why?

140 points by willm 3y ago

To me, these lessons feel very similar to the lessons learned working at a midsized company. Was there any particular takeaways that are exclusive to working at a FOSS?

magicalhippo 3y ago
While I agree that it's not terribly unique to FOSS, it makes it even more useful IMO since relatively few work on FOSS. I think the lessons learned and advice given are great for a large number of developers.
simonw 3y ago
The lessons about working with external pull requests feel unique to FOSS to me.

quantum_state 3y ago

Very valuable … thanks for sharing

candiddevmike 3y ago

I looked all over Textuals website and I have no idea how they make money. Donations? Consulting?

smokel 3y ago
Ads in TUIs are an underdeveloped market, I suppose.
CubsFan1060 3y ago
I think the eventual plan might be this? https://textual.textualize.io/blog/2023/09/06/what-is-textual-web/ https://textual.textualize.io/blog/2023/09/06/what-is-textua...
zeckalpha 3y ago
Why must everything make money?
- lmz 3y ago
  Usually people don't have enough savings to dedicate themselves full time to a non money making project.
  - hackernewds 3y ago
    if the project pays them, it works
- tsuica 3y ago
  Because people have to eat.
- 7bit 3y ago
  Why do you not work for free?
  - zarathustreal 2y ago
    Contributing to OSS is a hobby, not a job. You explicitly work for free on purpose. Whether or not you’re able to pay your bills is unrelated to the OSS contribution, it’s very clear that you’re not going to get paid for writing code that you’re giving away for free.
    - willm 2y ago
      You only had to read the title to know that the author of the post was paid for 503 days to work on OSS. It was literally his job.
      - zarathustreal 2y ago
        An example of one person being employed and contributing to open source as part of his job is not a refutation of the statement. You don’t have to read anything to know that, just use logic. To provide a counter-example you’d need to reference an open-source project that makes money selling the source code which, hopefully obviously, doesn’t exist, by definition
SushiHippie 3y ago
https://www.willmcgugan.com/blog/tech/post/from-open-source-maintainer-to-founder-ceo/ https://www.willmcgugan.com/blog/tech/post/from-open-source-... > Additionally, I would work on a cloud service using both those projects which at some point would become a business. > Now this was a good plan. The money from GitHub sponsors which I had previously been donating to charity would take the sting out of not having an income for a while. While far from a salary (where I live), at around $1000 a month it would help pay a few bills. > I have far more confidence in that vision now that I have raised seed funding. Predicated on that cloud service, I have enough funding to hire a few developers to work on it with me. Last month I was an unemployed code monkey, today I am a Founder / CEO of Textualize. Go figure.
- candiddevmike 3y ago
  Good find! Getting seed funding to try and monetize a bunch of MIT libraries using a "cloud service". Wild. Wonder if OP was always 100% dedicated to FOSS for 503 days or if they worked on the "cloud service"...

jacksonhacker 3y ago

[dead]

bionhoward 3y ago

I love Rich! Using it for tree_plus and other projects, and it’s beautiful and hackable. Great Python package for terminal visuals. Definitely recommend it if you have a Python CLI, can really improve readability and visual niceties.

righthand 2y ago

This is all pretty ideal and I think these outcomes are more likely if the project has a substantial base. Comparatively I was trying to contribute to another FOSS project only to get push back. I started contributing to the docs to build the package which were way out of date. The project switched from qmake to cmake. The project did not have a correctly working cmake setup so I assisted with that. Every time I contributed the maintainers would check my PR then ping another maintainer who was supposedly working on the specific issues. Eventually I managed to get some docs and code merged but after that the maintainers returned harder to blocking my PR’s with their own work which they happily merged in. They would not update my PR that they had already gone in and solved the problem, I had to repull and figure it out myself, which left me with open PRs/tickets just lingering in the space. They would happily clarify my questions with obscure non-revealing conversation about their project. Eventually I gave up. Similarly I have tried to contribute to things like KDE in the past as well and been met with similar walls on contributing. Some developer somewhere is in charge of a FOSS project as a hobby and only really cares about their use cases or providing a simple demo app or rewriting an old app for their needs because the old app was too complicated or probably run by another hobbyist maintainer. This leads to opening tickets and looking for ways to contribute or discuss implementation with little to no response. Or in the case of KDE, having your ticket closed and shifted over to their bug system which requires rediscussion and retriaging. At some point you just lose momentum and interest trying to push a project along due to the sheer uneventfulness and refocusing of it all. I think we need fewer hobbyist project owners and more hobbyist project managers.

CoastalCoder 2y ago
I had similar frustrations with a well-known deep-learning framework. I'd get well-vetted PR's for the build system, CI system, etc. committed. Only to have them later undone, without discussion, by people in the inner circle. It doesn't take many experiences like that to lose all motivation. I think some open-source projects deceive themselves (and therefore, outsiders) about their contribution processes.
- ozim 2y ago
  Not wondering people rather start their own thing rather than contribute to someone else projects. Cambrian explosion of JavaScript frameworks explained. For me at least as I never really tried to contribute to any open project.

RojerGS 2y ago

The title of this posting was edited and no longer matches the title of the article linked. I wonder if anybody knows why?