Loading ...
Sorry, an error occurred while loading the content.

6050Re: Decentralised Meta-Data Search Strategies

Expand Messages
  • samrhjoseph
    Aug 7 5:20 PM
      --- In decentralization@y..., Lucas Gonze <lgonze@p...> wrote:
      > Sam said:
      > > I'm still not sure which
      > > difference between FASD and Plesh you are trying to highlight. I
      > > mean both FASD and Plesh allow you to specify some meta-data,
      > > a search and then retrieve a URI, that URI can then be used in
      > > another network to retrieve a document. In my understanding of
      > > term "overlay network" it applies to both Plesh and FASD.
      > The difference I had in mind is that Plesh is a network dedicated to
      > carrying metadata (even though it can also carry other data), while
      > has to cooexist with other uses of Freenet.

      Right, so here I think we have a small misunderstanding, which I will
      try to correct in the paper. FASD does not have to exist with other
      uses of Freenet. One can set up a dedicated network to support FASD
      that does nothing but support meta-data search, and then use a
      completely different network to actually retrieve the documents.

      Thus it would be quite possible to create a FASD network dedicated to
      carrying meta-data.

      I was wondering whether that
      > focus would let Plesh optimize the topology, and the answer is no.
      > Actually, FASD seems to have a more flexible search strategy.

      Yes, not sure about that. FASD and Plesh both can't support
      substring matching at present, which seems to be a fairly fundamental
      function of using hashes. As I note in my paper there may be some
      work around that involves hashing all available sub-strings, but
      superficially this would seem impractical in both cases. The main
      difference from a user point of view would appear to be the type of
      query supported, RDF or TFIDF Vector.

      > But I'll leave off with this until I get my real job taken care of
      > read the FASD paper all the way through during the weekend. Pretty
      > interesting paper so far...

      Thanks again for the feedback.

      Good luck with that real job stuff :-)

    • Show all 9 messages in this topic