On 9/14/07, Andrew Lentvorski <[EMAIL PROTECTED]> wrote: > Bob La Quey wrote: > > On 9/14/07, Andrew Lentvorski <[EMAIL PROTECTED]> wrote: > >> Bob La Quey wrote: > >> > >>> Well it does raise an intriquing question. Where do you > >>> distribute search? I see no reason for search to be centralized. > >>> The net was originally intended to be peer to peer. I see no > >>> real reason that it is not improved by returning to those > >>> roots. > >> I do. There is no good method other than "flood" to search p2p. (Who > >> has it? Dunno. Who is closer? Dunno.) > > > > Perhaps a heirarchy would help? A bit like DNS. Your search term > > goes into a heirarchical system and is only sent to those boxen > > that have the relevant index. > > Okay, so that's an index to an index. So, how do you maintain the > accuracy of the index to the index? > > It's turtles all the way down ...
I will return to this. > Kademlia attempts to answer this. However, again, the problem is that > Kadmelia assumes altruism, equality and accuracy in the nodes. These > properties are very absent in a system in which there is high incentive > to game. An alternative to altruism is "trust but verify." In other words certain archives (of say indices) have authorized maintainers. Is this labor intensive, yes. Can the labor be seriously subdivided, probably. Can algorithms help, almost certainly. > And we haven't even gotten into malicious nodes or nodes that choose not > to cooperate. Essentially the same answer. Distribute human intelligence and labor. This has been the solution that libraries have been using with at least some success for centuries. > The problem is that the only person with incentive to have you search > accurately is the content provider. Not true. "I" have an incentive to search accurately. All of those "I"s makes up a very large group of people. >To everybody else, you are a drain > on resources (at best) Not if I never access their site. >or someone to be messed with (at worst). I am harder to mess with in the system that I am suggesting will be required. > P2P vs. centralized have different strengths in different domains. When > searching for something *specific* we tend to use a centralized index > (libraries developed card catalogs for a *reason*, after all). When > searching for something vague, we tend to use P2P (You ask another > person, "I liked book X. What other book might I like?") Well libraries have at least two systems in operation: the central index and the physical layout, where books are usually sorted into subject categories. I often use a library first finding one book that fits my criteria then going to the stacks and browsing. Google is trying to support this idea with the "similar" link. I have not tried it much. Maybe I should. So back to the "turtles all the way down" as promised. It does not have to be. (What keeps the DNS from having this problem?) As with Wikipedia and with some similar problems we can use humans more. After all, we have six billion of them. Even mighty Google has only a few million servers. So we have lots of humans. The puzzle is how to qualify them. Again if we look to old fashioned libraries a lot of labor goes into catalog maintenance. Maybe that is the only way out. It seems clear that the maintenance of knowledge access is going to be a growth industry. Indeed, maintenance will likely outgrow the creation industry. When is it easier to recreate an answer than to find an existing one? BobLQ -- [email protected] http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list
