4 ms·
My problem is that protocols and software are not trees with their own imperatives for existence. They're just tools for people, and a useless tool gets neglec
by Semiapies 5y ago
My problem is that protocols and software are not trees with their own imperatives for existence. They're just tools for people, and a useless tool gets neglected. Even a fun tool that's not as capable as others gets used less and less over time, once you get past a period of infatuation. And with a small user base, that trend turns into a death spiral.
I'm curious to see how long Gemini keeps an active community.
- selfhoster11 5y agoGopher seemed to have a healthy (if small) adoption for years after it was declared dead. If Gemini becomes a Gopher+, then it has a strong chance at staying around.
- Semiapies 5y agoGopher has actually evolved since, though. The Gemini community doesn't want their protocol to evolve.
- spc476 5y agoThe Gopher+ protocol doesn't seem to be that popular in practice, and the only evolution of the gopher protocol since the mid-90s is the introduction of a few new selector types not found in RFC-1436 (like 'i' and 'h'), and the switch to UTF-8 (instead of ISO-8859-1 mentioned in RFC-1436). There are a few servers and clients that support TLS for gopher, but it's not exactly a good fit [1]. [1] http://boston.conman.org/2019/03/31.1 http://boston.conman.org/2019/03/31.1
- selfhoster11 5y agoMy bad, I didn't realise that Gopher+ was an actual protocol. I simply meant to use that name as a placeholder for something that's slightly more than basic Gopher, not to refer to any specific proposal.