choosing non-free platforms is an individualist, short-term investment which prioritizes your project’s apparent access to popularity over the success of the FOSS ecosystem as a whole
Nah, I don’t give a shit about stars or followers. Had the argument been the endowment effect, I probably wouldn’t have commented at all. I have very limited free time and choose to spend it on things I enjoy. Migrating every single one of my hundreds of repos from more than a decade of collection is not something I’m interested in. I know this because I started the migration to GitLab when MS bought GitHub and it was a huge time suck that brought me no joy. Realistically no one is going to use my code after I die so who the fuck cares.
On the other hand, choosing FOSS platforms is a collectivist investment in the long-term success of the FOSS ecosystem as a whole, driving its overall growth.
I don’t think the FOSS ecosystem could scale quickly enough to handle mass exodus. If all the MIT and Apache 2 code left GitHub for Codeberg, I think Codeberg would die. But what do I know? If Drew DeVault wants to use their free time to migrate my code to the open ecosystem and put their money where their mouth is, I’d be happy to move. I just need all of the servers and random computers that I do dev on to have the remotes updated too.
Cool! Does he migrate all the code and update references or does he just make sweeping generalizations without understanding common user personas for the experiences he claims to own?
Going after the wrong dude my friend, this guy is a friend of the FOSS movement.
As for you, its alright to keep all your project codebases on github or gitlab etc. I think the article is majorly talking about large scale codebases that aim to replace existing closed source functionalities. Either way, if you plan and wish to implement a large project that you think will have many contributors, perhaps you could consider codeberg and similar open source devops projects to host and run your new project on, from the start. That way you won't have any migration pains. If it doesn't end up working out, hey, thats also a useful report for others who might be thinking about doing the same.
I’m going after the right paradigm. There’s an attitude in FOSS that if you don’t blindly follow everything someone from FSF says (or someone who looks up to Stallman), you’re a bad person. See my first pull quote from above. If you’re going to say something like that and not offer solutions to the pain points of your customer journey, you deserve ridicule (“you” being the author and the slrpnk user who dropped the important first part from their pull quote). The only reason I commented was because of the complete mischaracterization of people like myself, who I know many of, where the endowment effect is a more realistic description than this narcissistic spin.
Your argument is also very tenuous given the outage I called out. I code every day. Literally not figuratively. I code on at least two different devices work days and sometimes mobile (not laptops) weekends. I require a stable remote. An outage of 170hr is not something I care to handle. That’s just me solo. If I wanted to coordinate more than me, it would be a complete dealbreaker.