[extropy-chat] Conquering the world with a POV wiki

Jef Allbright jef at jefallbright.net
Tue Dec 26 18:57:08 UTC 2006


brent.allsop wrote:

> The goal is to provide as many data sources to "canonizers" 
> as possible.  For example, ivy legue schools could provide 
> some kind of verification interface to validate the people do 
> indeed have a PhD in Physics from that institution.  (until 
> we get something like this, we could just have a way for 
> people to claim they have such...)  Then there could be a 
> "canonizer" that only counts votes have people that have such 
> PhD degrees - and so on.
> 
> Canonizers could use any and all of these data sources to 
> rank things any way they wanted.  If there isn't one that 
> "canonizes" things the way you want, you simply submit a new 
> one, or help us provide a new way to get whatever new data 
> source you require, into the system.
> 
> Another example might be some kind of "maverick" score that 
> ranks people that start out in very minority camps before 
> everyone else is converted to the camp - and so on.  I'm 
> keeping all the history so things like this will be very possible.
> 
> <<<<
> Have you thought about how your POV wiki idea might relate or 
> interoperate with prediction markets?
> >>>>
> 
> Absolutely.  As one example, someone's net worth or 
> performance scores in such prediction markets could be yet 
> another data source that canonizers could use to judge the 
> value or reputation of a person's vote.  The two systems 
> definitely fit well together and I definitely figure there is 
> great collaboration possibilities with these two types of efforts.
> 
> Another big patented idea is how it will be possible to 
> delegate your vote to someone you respect in a particular 
> field which you don't know much about, and that person 
> further delegating his support to yet another person and so 
> on.  Resulting in a complex dynamic tree structure of 
> delegated support.  So once you "earn" a reputation, you will 
> have lots of power to influence the reputation of any "camp" 
> and so on because you could carry a potentially huge tree 
> structure of delegated votes with you.  Of course, if you 
> screw up in any way, all that reputation via delegated 
> support could vanish in an instant...
> 
> And of course, if everyone in your delegated tree is a 
> declared "christian" or whatever, you could select a 
> canonizer that would ignore it all and so on.
> 
> Thanks and tell me more about your thinking!

Well, I think you're heading in the right direction as the value of the
web shifts: from effective access to pages of information, to effective
access to highly ranked pages of information, to where the information
in the links approaches or even surpasses the information in the pages
themselves.

This seems nearly congruent with the problem of building an effectively
autonomous artificial intelligence.  We know how to store and retrieve
representations of knowledge, we're getting better at extracting and
linking the salient features of knowledge, but we still lack a practical
artificial system for "evolving" the links that comprise the
"intelligence" of the system without having humans in the loop.

In the near term there's a lot of low-hanging fruit to be harvested from
the latent information implicit in the activities of humans interacting
via the net.

That's an important piece of my current business plan, with the
expectation that Google and their like continue to serve the mainstream,
leaving the long tail to provide many smaller economic niches for the
rest of us.

- Jef




More information about the extropy-chat mailing list