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

Re: [decentralization] de-centralized data patterns

Expand Messages
  • Paolo Massa
    I guess big de-centralized data patterns can be found in scientific literature. Decentralized because every journal or author can make available informations
    Message 1 of 14 , Jan 14, 2003
    • 0 Attachment
      I guess big "de-centralized data patterns" can be found in scientific
      literature.

      Decentralized because every journal or author can make available
      informations about some papers.
      Data patterns are deriving from citation, linking, reading activities.

      Eprint.org [1] is a big project dedicated to opening access to the
      refereed research literature online through author/institution
      self-archiving.

      The key idea is: Authors self-archive their eprints papers in
      open-access archives that comply with the Open Archive Initiative (OAI).

      Opcit [2] is a sub-project that is
      * Defining the operational semantics of documents (digital objects)
      to allow perfect linking

      There is a "A wide range of more or less centralized tools and
      initiatives that take advantage of this cloud of data".
      All the software (GNU EPrints) [3] is available as free software (GNU
      licence).

      Open Archives Initiative (OAI) [4] develops and promotes
      interoperability standards that aim to facilitate the efficient
      dissemination of content. The Open Archives Initiative has its roots in
      an effort to enhance access to e-print archives as a means of increasing
      the availability of scholarly communication.

      It seems a lot of universities are converging to this formats (and other
      such as OpenURL), or at least this paper (Open citation linking - the
      way forward) [5] say so.

      Conclusions of this paper:
      - OAI is gathering momentum.
      - Software for building OAI repositories is available
      - Institutional archives are being created, but need to be filled by authors
      - Attracting authors requires evidence of services that will improve the
      visibility, usage and impact of their works.
      - Citation-ranked search and reference linking are examples of OAI
      services that do this
      - The infrastructure supporting OAI services continues to be enhanced
      - Resource discovery and current awareness are exemplar OAI services
      now. Future services may be citation mapping and co-citation
      visualisation, preservation management, and personalization

      Paolo

      [1] http://www.eprints.org
      [2] http://opcit.eprints.org/
      [3] http://software.eprints.org/
      [4] http://www.openarchives.org/documents/FAQ.html
      [5] http://www.dlib.org/dlib/october02/hitchcock/10hitchcock.html
    • cefn.hoile@bt.com
      Not to forget Citeseer http://citeseer.nj.nec.com/cs - linked in some way to the google search engine I think. Cefn ... From: Paolo Massa [mailto:massa@itc.it]
      Message 2 of 14 , Jan 14, 2003
      • 0 Attachment
        Not to forget Citeseer http://citeseer.nj.nec.com/cs - linked in some way to
        the google search engine I think.

        Cefn

        -----Original Message-----
        From: Paolo Massa [mailto:massa@...]

        Eprint.org [...] The key idea is: Authors self-archive their eprints papers
        in
        open-access archives that comply with the Open Archive Initiative (OAI).
      • Lucas Gonze
        ... Hm, now that you mention it this strikes me as a nameable thing about the pattern.
        Message 3 of 14 , Jan 14, 2003
        • 0 Attachment
          Julian Bond wrote:
          > For one of these standards to succeed there does need to be a virtuous
          > circle of publishing, tools, and standard.

          Hm, now that you mention it this strikes me as a nameable thing about the
          pattern.
        • S. Mike Dierken
          ... From: Julian Bond ... What do you mean by relatively centralized tools? Google is an example of widely distributed data and a
          Message 4 of 14 , Jan 14, 2003
          • 0 Attachment
            ----- Original Message -----
            From: "Julian Bond" <julian_bond@...>

            > 1) Has anyone got a good name for this pattern? Widely distributed,
            > standardised, machine readable data combined with relatively centralized
            > tools and data processors. "de-centralized data pattern" is much too
            > clunky.

            What do you mean by relatively centralized tools?
            Google is an example of widely distributed data and a central
            tool/processor - does that fit the pattern?
            Is another way to put it "distributed data, centralized access"?
            It seems you are getting at more then the storage pattern, but both the
            storage and usage.
            How is it different from aggregation or integration?
          • Julian Bond
            ... Look at RSS. We ve got all of it. We ve got everything from very de-centralized desktop readers and publishers to very centralized 2nd level meta sites and
            Message 5 of 14 , Jan 14, 2003
            • 0 Attachment
              S. Mike Dierken <mdierken@...> wrote:
              >----- Original Message -----
              >From: "Julian Bond" <julian_bond@...>
              >
              >> 1) Has anyone got a good name for this pattern? Widely distributed,
              >> standardised, machine readable data combined with relatively centralized
              >> tools and data processors. "de-centralized data pattern" is much too
              >> clunky.
              >
              >What do you mean by relatively centralized tools?
              >Google is an example of widely distributed data and a central
              >tool/processor - does that fit the pattern?
              >Is another way to put it "distributed data, centralized access"?
              >It seems you are getting at more then the storage pattern, but both the
              >storage and usage.
              >How is it different from aggregation or integration?

              Look at RSS. We've got all of it. We've got everything from very
              de-centralized desktop readers and publishers to very centralized 2nd
              level meta sites and all points in between.

              The key was that it was so simple that anyone could play and get a quick
              win. So they did.

              I think the big difference from html, web servers, browsers, proxies and
              google is that html was always meant to be human readable data while
              RSS, FOAF, SMBData et al are about machine readable metadata. That opens
              up more possibilities for intermediaries.

              The reason I find this interesting is that there's a tension here
              between de-centralizing and centralizing. It forces you to confront
              "Where should the tool be", "Where should the data be", "should I copy
              the data or read it fresh each time" and similar questions, at every
              step.

              One lesson I think we can learn (that didn't actually come from RSS) is
              that including pointers to other instances in the files themselves can
              really help to get initial critical mass. Having every file have seeAlso
              references to other files of the same type makes it much easier to build
              the first spiders.

              --
              Julian Bond Email&MSM: julian.bond@...
              Webmaster: http://www.ecademy.com/
              Personal WebLog: http://www.voidstar.com/
              CV/Resume: http://www.voidstar.com/cv/
              M: +44 (0)77 5907 2173 T: +44 (0)192 0412 433
            • Bill Kearney <ml_yahoo@ideaspace.net>
              ... RSS) is ... can ... seeAlso ... build ... This reminds me of the old George Carlin joke, You can t have everything, where would you put it? What s the
              Message 6 of 14 , Jan 15, 2003
              • 0 Attachment
                > One lesson I think we can learn (that didn't actually come from
                RSS) is
                > that including pointers to other instances in the files themselves
                can
                > really help to get initial critical mass. Having every file have
                seeAlso
                > references to other files of the same type makes it much easier to
                build
                > the first spiders.

                This reminds me of the old George Carlin joke, "You can't have
                everything, where would you put it?"

                What's the progress (if any) on getting someting like a freespace://
                protocol handler described? Something that allowed pulling elements
                out of freespace (whichever one is suitable) would really help
                advance it's use.

                -Bill Kearney
              Your message has been successfully submitted and would be delivered to recipients shortly.