Peer-to-Peer (P2P) search debate, Part II




Every Tuesday night on AltSearchEngines we host a debate between two search engines in one particular area of search.  Tonight we have part two of the debate that began last week on Peer-to-Peer (P2P) search.  Here is the link to part one.



6) Participation: How do your users participate (By way of contribution and benefits)?

FAROO: Our users provide infrastructure, and ranking, and we provide technology, so in fact we are building our search engine together.  Therefore we decided that our users should also participate in the revenues.  Not some minor percentage, we are sharing revenues of up to fifty percent with our users.  They may donate their share for charity organizations, joining their forces not only for search, but also for helping other people.

Of course users also benefit from the collectively generated ranking and privacy protection. Currently we are about to define an API for FAROO’s distributed database.  That would allow everybody to use the distributed index, the collected information and ranking, both commercially and non-commercially, in mash-ups, with their own interfaces…

Probably there will be some contribution/usage-ratio, to keep the p2p-principle working.

ATLAS: Atlas is primarily for search implementers and services, not really end users. The model supports both free, attribution-based, and paid relationships, every entity must choose how they want to provide value into the network and what kind of reward they need for doing so. 

 7) Privacy: How do you protect the privacy of your users?

FAROO: FAROO does not collect any search log files. All search queries and the distributed index are encrypted.  Neither the other peers nor any intermediate party may observe searches or visited pages.

Due to the fully distributed architecture, FAROO is in fact working as a p2p anonymizer for searching, ranking and crawling.  No personal information is leaving the computer at any time. Even personalization is done client side.

We are giving back search privacy and censor resistance to the user, something what is more and more fading away with current search engines. 

ATLAS: There is no session data / cookies, IP addresses, or any user information whatsoever supported by Atlas.

8) Spam: How do you deal with search engine spam or SEO manipulation?

FAROO: With our attention-based ranking, SEOs efforts are not deciding the rank anymore, but the community of users itself.  Results which are of no interest to the users automatically lose rank and disappear from search results. With several statistical measures FAROO prevents bogus peer manipulation. 

ATLAS: Through competition, if any entity is injecting spam or being manipulated, competing entities have an opportunity to provide a higher quality.

9) When do you plan on having both a private beta, and a public one?

FAROO: During our presentation at TechCrunch40 September 17th we started our private beta. This is mainly intended to test and improve our distributed architecture.Using a dedicated crawler we are now increasing the number of indexed documents to prevent a “chicken or egg problem”, preparing our public launch later this year. 

ATLAS: Near the end of this year, being open source, they are one and the same :-) .

10) Partners: What, if any, is your relationship with the dozens of “alternative” search engines? Any partnerships past, present or in the future?

FAROO: We are just about to start our search engine, so there are no partnerships with “alternative” search engines yet.  Of course we are always looking at what others do, and some recent achievements of “alternative” search engines in result clustering, natural language search or image recognition are amazing.

On the other hand, with opening our distributed index through an API there will be a lot more options for partnering in one or the other way.  But already today we have a lot of strong partnerships, and a lot more are to come – with each of our users ;-)

ATLAS: It’s all open source so partnerships don’t apply, anyone can get involved to help if they want to be part of it.

That concludes our two-part debate on Peer-to-Peer (P2P) Search.

I want to heartily thank our participants, Wolf Garbe of FAROO and Jeremie Miller of Wikia.

If you have any questions or comments, please leave them here.

5 Responses to “Peer-to-Peer (P2P) search debate, Part II”

  1. University Update - Open Source - Peer-to-Peer (P2P) search debate, Part II Says:

    [...] Peer-to-Peer (P2P) search debate, Part II » This Summary is from an article posted at Alt Search Engines on Tuesday, October 09, 2007 [...]

  2. GadgetGadget.info - Gadgets on the web » Peer-to-Peer (P2P) search debate, Part II Says:

    [...] Danny Sullivan wrote an interesting post today!.Here’s a quick excerptEvery Tuesday night on AltSearchEngines we host a debate between two search engines in one particular area of search. Tonight we have part two of the debate that began last week on Peer-to-Peer (P2P) search. Here is the link to part one … [...]

  3. Relationship » Peer-to-Peer (P2P) search debate, Part II Says:

    [...] Brian Chavez wrote an interesting post today onHere’s a quick excerptThe model supports both free, attribution-based, and paid relationships, every entity must choose how they want to provide value into the network and what kind of reward they need for doing so. 7) Privacy: How do you protect the privacy … [...]

  4. FAROO Blog » Blog Archive » Great Debate: Peer-to-Peer (P2P) Search Part I + II Says:

    [...] you can read part I and part II of the [...]

  5. InternetActu.net Says:

    Après Google : les moteurs de recherche P2P ?…

    Quand on évoque un moteur de recherche P2P, on pense le plus souvent à un moteur de Torrents pour chercher des albums ou des films sur des plateformes P2P. Pourtant, depuis Skype ou Joost, on sait que la technologie P2P peut avoir d’autres appl…

Leave a Reply