Okay, I can finally write "new" versions of files in Crystal (tahoe-lafs / freenet like mutable p2p distributable files that *don't* have consistency guarantees) and then retrieve them again.

I think this will be a useful demonstration of both some cool things we can do in p2p systems and the ways CAP tradeoffs get in the way... and how those tradeoffs reappear everywhere throughout our systems.

Here's my rough expected schedule for Crystal:
- Code dump: this week
- Docs: next week
- Crystal Golem (a version of Golem that supports update'able content in a p2p system): week after next

Did you implement the filesytem in Racket or are you using something like the Tahoe client?


@AceNovo It's implemented in Racket. It's not as robust as Tahoe's, more of a demo of the core ideas, stripped down so that they're understandable.

Understandable is good 👍

I think what I want is to help you move these ideas from proof of concept to minimum viable product, put them in front of players, and see where the pain points are wrt technical vulnerabilities and social concerns. I'm spending way too much time on infrastructure for that to go quickly, but what can you do? 😎

@AceNovo I fully intend to move from the PoC stage to the MVP stage, but it's very important to me to make the ideas be well described and well documented enough so that I don't build some indecipherable alien artifact, which I think has been how people have felt about many p2p projects historically

@AceNovo wow I desparately need to expand those acronymns because they are PROBLEMATICALLY AMBIGUOUS

PoC here means Proof of Concept
MVP here means Minimum Viable Product

I normally kinda hate those acronymns because of their obvious naming conflicts so kinda embarassed on the re-read :X

Sign in to participate in the conversation

The social network of the future: No ads, no corporate surveillance, ethical design, and decentralization! Own your data with Mastodon!