| |||||||||||||
Distributed social networking is a hard problemDistributed social networking is a hard problemPosted Oct 5, 2012 21:43 UTC (Fri) by rillian (subscriber, #11344)In reply to: Distributed social networking is a hard problem by Baylink Parent article: Tent pitches a new social networking protocol
> The problem with distributed systems like this with no central trusted third party [...]
One idea the developers have, which wasn't very clear at launch, is to use copyright licenses to enforce this sort of thing. That is, recourse to the legal system is the 'trusted third party'; what cannot be done in computer code can be done in legal code.
"Tent posts can have licenses attached. Licenses are basically contracts, like terms of service on a specific post. [...] For example a license could be created that prohibited the receiving user from reposting any posts with the license attached or requiring all users who received the post to store it only on an encrypted server." -- http://tent.io/blog/tent-basics
Interesting idea.
(Log in to post comments)
Distributed social networking is a hard problem Posted Oct 5, 2012 22:08 UTC (Fri) by Baylink (subscriber, #755) [Link]
Perhaps, but it doesn't address the class of problems I identified.
Distributed social networking is a hard problem Posted Oct 7, 2012 0:19 UTC (Sun) by giraffedata (subscriber, #1954) [Link] It's an interesting idea, but since copyright was not designed, and has never been used, to protect confidentiality, I see nothing but frustration.A non-disclosure agreement (NDA) would make a lot more sense. In order to get Bob's posts, Alice would have to agree to restrictions on reposting them. I don't know how information travels around in these systems; maybe it isn't feasible to make an NDA a prerequisite for accessing the information.
Distributed social networking is a hard problem Posted Oct 7, 2012 14:58 UTC (Sun) by Baylink (subscriber, #755) [Link]
Yes, but this still requires me, Bob, to *tell* Alice whom I don't want to see my posts.
This is why a trusted third party, who *does* know the shape of the graph, but does not tell any of the participants, is essential to allow the sort of restrictions on distribution that we've become accustomed to on, say, Facebook.
The sort of *problem* we're trying to fix here, of course, is the one I have right now. At 12:26EDT Friday, I bought an ad on FB, to promote an FB event. Apparently, this is Suspicious Behavior; my account was locked out the same minute, and I'm told their "upload us a picture of your government ID" process to unlock it can take *three to 15 days*.
Oh: yes, they accepted the ad, 3 hours later, and they are, presumably, happily charging me for it, though I couldn't log in to cancel it if I wanted to. So, clearly, I encourage thought about how to get *around* this category of problem, as unfruitful as I think it might be...
Distributed social networking is a hard problem Posted Oct 7, 2012 17:34 UTC (Sun) by giraffedata (subscriber, #1954) [Link] Yes, but this still requires me, Bob, to *tell* Alice Yes, as you posted before, the copyright scheme isn't meant to address the problem of keeping the block confidential, and the NDA scheme is just a possibly effective alternative to what the copyright scheme does mean to address.
The sort of *problem* we're trying to fix here ... Apparently, this is Suspicious Behavior; my account was locked out the same minute. ... So you're saying as long as one company owns social networking, that company can exclude someone arbitrarily and unfairly, but if it were an open, ungoverned social network, that could not happen, right? I suspect the ungoverned version would be worse. Facebook is trying (probably effectively) to solve a real societal problem with these imperfect detections of suspicious activity. I don't know what it is, but it's probably preventing some kind of theft or other thing I'm against. So if we did have a viable open, distributed version of Facebook, I think people would insist on the same kind of protections.
|
|||||||||||||