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

6046Re: Decentralised Meta-Data Search Strategies

Expand Messages
  • samrhjoseph
    Aug 6, 2002
      --- In decentralization@y..., Lucas Gonze <lgonze@p...> wrote:
      >
      > Sam Joseph wrote:
      > > I
      > > think the ease of publishing data in the p2p environment means
      that
      > > the need for decentralized search is even greater than on the
      web,
      > > but that the churn makes it all the harder to provide ....
      >
      > churn is a brutally hard problem to solve. I'm surprized there are
      no
      > projects working on reducing it instead of accepting it.

      well there probably are, but I think they are at a lower level than
      p2p. I think there are many projects that are working on just trying
      to make more stable connections between two end points, independently
      of whether they are participating in a p2p network.

      >
      > > > Does it matter whether metadata search is an overlay on the
      > > network, as
      > > > with FASD, or the main linking structure, as with Plesh?
      > >
      > > I'm not sure that I see the difference you are pointing to here.
      I
      > > mean both FASD and Plesh work as overlays, I think.
      >
      > From http://semplesh.com/faq:
      > *******
      > Can I simply retrieve files by their hash/URI, and how does this
      work in
      > general?
      >
      > The Plex is built upon a hash-to-plaintext system which will allow
      you to
      > retrieve files, but this is not the Plex's main usage. Instead, the
      Plex
      > will allow you to retrieve all sorts of information about a URI.
      One type
      > of information, of course, is a representation of the URI, like
      what you
      > might get if you typed that URI into a Web browser.
      > *******
      >
      > ...meaning that they're focused on the same issue of metadata in
      > decentralized networks that you are, so there's probably relevant
      > insights.

      I'm sure there are relevant insights, but 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, perform
      a search and then retrieve a URI, that URI can then be used in
      another network to retrieve a document. In my understanding of the
      term "overlay network" it applies to both Plesh and FASD.

      I'm not sure what you mean by "main linking structure". Sorry to be
      a stickler on these terms, but I think that it is my understanding of
      them that is at fault, and I am keen to clear up my own confusion.

      To me the difference between FASD and Plesh is just the form of the
      metadata. In Plesh I specify parts of an RDF statement. In FASD I
      specify elements of a TFIDF Vector. In both systems I receive a set
      of matches to the meta-data I have specified, both of which may
      include URIs that I can then access from somewhere else ...

      CHEERS> SAM
    • Show all 9 messages in this topic