Flutterby™! : on LID and Trust 2005-02-09 18:32:05.365004+00

on LID and Trust

2005-02-09 18:32:05.365004+00 by Dan Lyke 3 comments

I did some ramblings on LID and Trust, if anybody cares.

You must keep your head in a vise. Or is it vice? Tight thoughts. Identity and Trust: You and I have defined each other as 'meet me in Des Moine with $50k cash, no questions asked.' people. Dispite both of our playings in Orkut, Tribes and such places, I see very little of this going on in the 'social networking world' - at least at my level. Yet I just got back from a Chamber trade show at the Convention center where lots of people were glad to see me.. and half of them asked: "What are you doing working for 'them'". - It was a lesson about how much people over the years have trusted me, and how hard I have worked to earn and keep their trust.. and how I was being put in a position to lend my identity and trust level to the people I am working for, Could they be worthy of such trust? In time, yes. But it's lots of hard work and (re-)training away...

I prefer to keep my head in a vice, but, yeah, if I don't keep my head in a vise I don't get anything done.

I guess what I'm thinking of here is that in your particular application, you're going to have one entity, call it the Forum, vouch for the paying ability of a group of people, the Forum Users, to read a Publication. Now the Publication can do that by seeing that the Forum controls the URL

of the LID

, can query the Forum and ask "will you pay for access by the user who has this _LID_?"

But the consequences of trusting people vary. Sometimes trust isn't an economic thing, it's a social inconvenience. If comment spam ever becomes an issue here, I'd love to have Flutterby be able to ask "has Meuon ever corresponded with this person?" or "is this person's karma on /. good?", and on the basis of that let them post immediately. If we can't trust them immediately, maybe we quarantine their messages until we know that they're not spammers. Not a life and death issue, but one of those things that through heuristics can gradually evolve to make the load of administration of a social space that much easier.

I guess that where part of this rambling is going is that in the non-virtual world, trust is earned. But trust is earned because we have an identity that's transferrable between people. If you say "Dan Lyke" and topspin says "Dan Lyke", there are mechanisms to make sure that you're both referring to me. But there's only a little social convention, one that many people are willing to circumvent, that the "Dan Lyke" who shows up on some new discussion forum is me. If we can tie identity to a unique token, even if I as an individual can make as many of those tokens, and therefore identities, as I want, over time I can give one of those identities enough of a history that it can earn trust.

#Comment LID's major point is made: 2005-02-13 05:08:24.456259+00 by: baylink

putting control of who knows what about you, and when you let them know it, in your hands.

This makes the whole "tying trustedness" thing difficult to do, I think inherently. Cause it's a question of early vs late binding, approximately. Or, more properly: public vs private binding.

See my thoughts on this on my blog tonight; I was going to expand the wiki front page with a "what does LID really do?" graf... and I got half way across it, and realized that the reason Johannes obfuscates that question so much is that I don't think even *he* knows what it really does, right now. :-}