{"p":"","h":{"iv":"ROXSYW+cfvEbFHu5","at":"ocxplSQjdRC3tXEtB/9/wg=="}}

@cwebber @garbados Is this paradigm likely to make testing purely distributed applications easier? Because frankly that was always the nightmare for me when I was working with dat (activitypub being decentralized but not distributed managed to not run into as many problems)

4
Share
Share on Mastodon
Share on Twitter
Share on Facebook
Share on Linkedin
crows call me breadlady

@darius @cwebber AFAICT, yes absolutely.

as an example, i'm writing a system for portable encrypted storage -- one of the major components the Spritely whitepaper outlines but doesn't specify -- as a set of five behaviors, or capabilities. these capabilities can be provides by subroutines on your machine, on other machines, on other planets... but for the sake of my program, i need only prove it in one case (all on my machine) because all the rest are abstractions upon that case.

0
2y
Christine Lemmer-Webber

@darius @garbados It's pretty easy for that, yes

0
2y
tech? no! man, see...

@darius @cwebber @garbados ocaps feels a lot like going from C to a garbage collected language

like ... if you ask "what does this let me do that I couldn't do before" the answer is technically nothing

but

you no longer have to be constantly on your guard wondering whether you accidentally left some tiny innocent-looking thing open that's going to end up in a massive vulnerability

0
2y
Levi Ramsey

@darius @cwebber @garbados at least at first glance there's a certain similarity to what Akka Typed (handing out an ActorRef[A] is permission to send an A) is doing.

My personal experience has been that that sort of thing makes building reliable distributed apps a lot easier.

0
2y
Replies