5 ms·

This makes me think about Rails, how successful it was/(is?). Basecamp or formally 37 signals showed how to not only give back to the open source community but

This makes me think about Rails, how successful it was/(is?). Basecamp or formally 37 signals showed how to not only give back to the open source community but also greatly benefit from it. They are funded by their own service and the success of Rails is complimentary. Babel feels like it could live at a big corp with deep pockets or at least a small indie corp with deep pockets. Does it need 3 full time engineers or are there enough volunteers to keep it successful? Maybe incentivizing people to donate in order to achieve milestones that they can see making their companies more money could work too.

syl_sau 5y ago
On OSS projects, I think that we need to find a more robust model than what we have in most cases. To me, the most successful and efficient models when it comes to building open, capable, stable and friendly software (all those four) seems to be delegating all the work to a team of a few but very experienced developers. I'm thinking Blender, GIMP, Csound, Ardour, many video codecs, etc. It also seems all these projects have in common highly technical fields (3D, audio or video), which implies a far steeper learning curves than your average github project. Maybe we could even infer that one solution to drive contributions up is to make it less easy and less ostensibly accessible. I personally pay for Blender because I want them to improve some features. Plus I'm not sure if getting pull requests from random strangers all the time is a good way to take a software in the serious direction -- I'd wager reducing complexity by lowering the amount of people involved and the extensive need for communication is what keeps the core dev team focused, fast and flexible. Anyway, my 2 cents.