6 ms·

Open-source client software with strong public-key cryptography (built to a spec, like XMPP clients) Web of trust for key verification Content hosted on your

by qb 16y ago

Open-source client software with strong public-key cryptography (built to a spec, like XMPP clients) Web of trust for key verification Content hosted on your friends' computers (similar to Freenet and Wuala)

Maro 16y ago
I'd rather avoid XML-based protocols like XMPP. My co-founder wrote an XMPP server in his previous job and always complained it's a terrible protocol and the people responsible should be punished by implementing their own fully-compliant XMPP server =)
- codeslinger 16y ago
  I couldn't agree more. XML sounded like such a great idea for this back in '99. Attempted implementation will disabuse you of that feeling very quickly ;-)
arethuza 16y ago
Replication of content would be important - you need to be able to see your friend's data when they are offline. Doing that securely would be interesting - but I suspect not impossible with current cryto. Would need to supporting work both as as server application and a thick client. Sounds a bit like an open Lotus Notes...
- codeslinger 16y ago
  I was thinking along the same lines, actually, in terms of a "distributed Facebook"'s similarity to Lotus Notes. However, one of the big niceties of SaaS services is that you don't bear the cost of storing and transforming all the photos, videos, etc that you upload. Replication of this data would also get quite expensive as you gained followers. And then there's the problem of bandwidth in the US. Seems to me there are some infrastructural problems with a distributed Facebook model that require some more thinking than just "lets run a server on people's home machines" (e.g. Diaspora)
  - arethuza 16y ago
    I'm not sure I'd be too worried about bandwidth issues - once content is replicated it could be served up from people who already have it rather than always going back to the original source. In fact, I suspect things would have to be done this way to provide some kind of resilience. I think getting the security model for distributed content would be the crux of this kind of architecture, and I suspect that is probably solvable and would really benefit from an open architecture with multiple implementations.

by qb 16y ago

Maro 16y ago

I'd rather avoid XML-based protocols like XMPP. My co-founder wrote an XMPP server in his previous job and always complained it's a terrible protocol and the people responsible should be punished by implementing their own fully-compliant XMPP server =)

codeslinger 16y ago
I couldn't agree more. XML sounded like such a great idea for this back in '99. Attempted implementation will disabuse you of that feeling very quickly ;-)

arethuza 16y ago

Replication of content would be important - you need to be able to see your friend's data when they are offline. Doing that securely would be interesting - but I suspect not impossible with current cryto. Would need to supporting work both as as server application and a thick client. Sounds a bit like an open Lotus Notes...

codeslinger 16y ago
I was thinking along the same lines, actually, in terms of a "distributed Facebook"'s similarity to Lotus Notes. However, one of the big niceties of SaaS services is that you don't bear the cost of storing and transforming all the photos, videos, etc that you upload. Replication of this data would also get quite expensive as you gained followers. And then there's the problem of bandwidth in the US. Seems to me there are some infrastructural problems with a distributed Facebook model that require some more thinking than just "lets run a server on people's home machines" (e.g. Diaspora)
- arethuza 16y ago
  I'm not sure I'd be too worried about bandwidth issues - once content is replicated it could be served up from people who already have it rather than always going back to the original source. In fact, I suspect things would have to be done this way to provide some kind of resilience. I think getting the security model for distributed content would be the crux of this kind of architecture, and I suspect that is probably solvable and would really benefit from an open architecture with multiple implementations.