On Tue, 11 Apr 2006, coderman wrote:
On 4/11/06, Emin Gun Sirer <[EMAIL PROTECTED]> wrote:
> Let me interject two factoids to make sure no myths are propagated:
>
> - The current Credence implementation uses explicit feedback. There is
> no reason why you couldn't use implicit indications of trust, if
> your application had such indicators.
thank you for pointing this out. do you know offhand how
easy/difficult it would be to extend the feedback mechanism to support
arbitrary qualifiers?
One of the earliest wish-list items was to allow more specific voting.
Maybe I am out of touch, but I was pretty surprised at how many people
wanted to be able to say things like "The file name is bogus, but the
bitrate, artist, and file type are all correct." Our most recent release
has a pretty general framework already in place to handle aribtrary
statements of this sort.
The user interface can now generate statements about file types, bitrates,
and file names, and I don't see any reason not to add other things too.
The main issue is trying to keep the GUI simple, and being careful about
the schema. Details are in our nsdi paper due out in a few weeks.
> It turns out that there are no such good implicit
> indicators in p2p filesharing - sharing a file
> is not a good indicator that the user would vouch for that file. Our
> paper has the details.
indeed. this is a hard problem and the good solutions are very
invasive and carry significant privacy concerns (for example,
feedbackfs monitors what files you open, how far you read into them,
if you copied them, deleted them, read them end to end many times,
etc. these actions are used to build implicit feedback (positive or
negative) associated with distinct file based resources. the privacy
concerns of such "file system profiling" should not be understated and
is why i've been detoured into strong security for so long)
Exactly right. I had though of trying to extract info from Windows Media
Player (which lets users rate items from 1-star to 5-stars, and also has
implicit automatic ratings based on usage), or adding a similar feature to
LimeWire's player. I didn't like the privacy implications, and I expect
others wouldn't either.
But that is not to say that implicit metrics are always bad, especially in
other domains. They just seem to be for file sharing.
- Credence computes a "very multidimensional" trust metric for each
[snip]
i mean to highlight this limited aspect only because what most people
want is "relevant" resources, and not necessarily "accurate meta
data", although the two often intersect.
I guess the problem would be to define "relevant". In the existing
networks, queries tend to be short, vague, and have no context. Since it
is not at all obvious what the user is looking for in the first place, it
would be kind of hard to decide what is "relevant" in the file sharing
world.
I'm not too sure what Philippe's bounty is looking for, though. He doesn't
mention files, or sharing, but does mention "family functions on a global
level". Can someone clue me in to what that is?
-Kevin
_______________________________________________
p2p-hackers mailing list
[email protected]
http://zgp.org/mailman/listinfo/p2p-hackers
_______________________________________________
Here is a web page listing P2P Conferences:
http://www.neurogrid.net/twiki/bin/view/Main/PeerToPeerConferences
