November 2016 Gna! announced it would shut its doors. I was on the list that sent this message. Even here on HN someone raised the cry [3]. I didn't hear it until it was too late. Others started pinging me that Gna! was gone; what would happen to the code, did they have the last commits, what's next?
I turned to GNU Savannah, to GitHub - what if they closed their doors? Was my illusion of permanence toward code repos unfounded? It caused me to question the Internet in general, how we approach projects like the Linux kernel differently from smaller projects hosted by GitHub, GNU Savannah and SourceForge. Are all projects equal and are their repositories? Certainly not if Gna! could be allowed to close for good.
When they cried "Gna! is shutting down!" were they not heard? Was this a symptom of a disorder, something that could grow, potentially reach more repos? Do we need to act? Some will say Gna! was barely used, most people hadn’t heard of it, what’s the big deal? Some will remind us the Internet has no functional permanence - such expectations are counter-productive. But source code repos aren’t just Internet sites – they’re infrastructure (IMHO). Should infrastructure not hold permanence, such that code repos simply change hands, not close? Why wasn’t the Gna! baton passed?
[1] https://savannah.gnu.org
[2] https://en.wikipedia.org/wiki/Gna!
[3] https://news.ycombinator.com/item?id=13559215