gemini://sunshinegardens.org/~xj9/posts/2017-11-16-kill-the-web/

友/os[1] is a natively peer-to-peer operating system that tries to tie together a selection of technologies and design patterns that we think could be used to create some really cool user experiences powered by mesh networking. 友/os is designed for deep integration with our mobile hardware project[2].

defeating web in terms of development and user experience on a micro scale isn't all that hard. market share is nice, but what we really care about is having a community of people that is excited about building interesting stuff. we really like the approach haiku has taken where they are open to external tech, but they spend a lot of time integrating it so it feels like a native part of the system. now, we hope we can have a faster pace than haiku, but that's a different topic.

its also important to enable commerce in a way the respects the privacy conscious ideas we have embraced. we are very inspired by the elementaryOS indie dev market that has a "pay what you want" model build right into the app center. libre software with an easy option to support the developer if i want to. this is something that @gargron@mastodon.social[3] got exactly right. he created mastodon and gets tipped for his work. patreon has issues, but the basic model is fairly sound.

consider something like snapchat. one feature raised 3 BILLION USD. that's nuts. the problem is, they didn't win. facebook copied their idea and used their network to get traction really fast. SNAP still has a lot of capital, but they are facing a money making machine with effectively unlimited resources.

looking at this one might think, how do we level the playing field when the opposition is effectively god? like the key to all of this is resources. so maybe we consider who is paying for what? cloud operations are really expensive at scale, and, as a developer, you have to front all of that yourself if you want to get any users. this also means that your costs increase as your user base increases. this model dooms us to a world of acqui-hires and giant tech conglomerates. infrastructure cost is a real problem.

individually, users don't cost much. the flip side is that social networks aren't interesting without a lot of users. ok so we need scale-ability and low costs, like really low costs, like zero. if (as a developer) your infrastructure is free all you have to worry about is making cool shit. you don't have any pressure to monetize (but the option isn't off the table either!) and if FB or anyone else tries to copy your idea they can't outspend you because the only resource you are utilizing is time.

> if FB or anyone else tries to copy your idea they can't outspend you because the only resource you are utilizing is time

at scale, we could be attacking these mega-corporations from every angle and force them to bleed capital while our grassroots communities are happily using our p2p tech and tipping us with monero for doing a good job. there are a lot of unknowns and challenges here, but by having each user pay for their own slice of the network it frees developers to create much more interesting applications because they can just have a fun time.

/g/ kill the web

let's talk about infrastructure