"That reddit-killing distributed self-hosted tool you’re building probably won’t kill reddit, sorry. [...] Make something personal for a dozen people, because that’s the one thing those massive vendors will never do and never even understand that they could do."

- Graham: "My proposal for scaling open source: don’t"

@sungo I see where he is coming from, but the "design for that one small group and others will notice" won't save you from the fact that you have to design for scale than at a later point, *if* you ever become popular.

@spaetz if. "if" is kinda rare. and "if" is always a point where you can bail out. and even if tons of folks start digging on your software, doesn't mean you have to sell out. it is possible for lots of folks to adore software that optimized for a niche or a small group without that software having to change. it is fully possible to say "this small community is my target market and this will remain my focus regardless how much paparazzi is at my door"

@sungo Sure, fully agreed there. Also, solving someone's specific problems is way more satisfying :-)

@sungo I miss the catch here: “but make sure you do not block potential paths for efficiency”.

That said, the most useful corona overview page in Germany is running on Python and created by a single person:

@sungo I think this is why I like Mastodon so much. It feels like a community. At least the instance that I'm on. It's a great blend of sharing personal details with anonymity. We get a blend of all perspectives in life. Sometimes we find someone from a foreign land that puts stuff in perspective and highlights the irony of it all.

Sign in to participate in the conversation

This is a single-user instance, namely for @sungo