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

Re: [GP] Request for help testing GP co-author graphics

Expand Messages
  • Langdon W B
    ... Dear Sean, Many thanks for your prompt and helpful reply about: http://www.cs.bham.ac.uk/~wbl/biblio/gp-coauthors/index.html The graphs are created by
    Message 1 of 8 , Feb 16, 2006
      > 1. Very interesting! Bill, I can't figure out what the function is
      > which defines the locations of the dots. Could you elaborate for us?

      Dear Sean,
      Many thanks for your prompt and helpful reply about:
      http://www.cs.bham.ac.uk/~wbl/biblio/gp-coauthors/index.html

      The graphs are created by graphviz' neato.
      www.research.att.com/sw/tools/graphviz/
      It tries to lay out the graph nicely, ie without stretching edges
      too much or with too many lines overlapping each other.

      > 2. What is the meaning of the graph that has no edges on it, just
      > dots of different sizes arrayed in a grid? What is the difference
      > between black and white dots? The very large center dot appears to
      > contain numerous people, sort of a Being John Malkovich thing!

      The top level graph is a summary of the whole bibliography.
      Each dot represents a connected subgraph (ie a bunch of people and
      their papers which have GP co-authors in common). The size of the dots
      is proportional to the number of entries in the bibliography.

      The fact their are so many dots is due to many singletons.
      Ie small groups who have published only a few papers together.

      I did not generate a graph for each case. In most cases the graphs
      would be quite trivial, linking together only 3 or 4 people. And there
      would have been hundreds of such graphs. Instead I applied an
      (arbitrary) cut off and only generated graphs with more than 15
      co-authors. These are indicated by red! Clicking on red dots pulls up
      the corresponding subgraph.

      Approx half of us are in the big red "John Malkovich" in the middle.

      > 3. Unfortunately, the program doesn't appear to run properly on
      > Safari. Works on Firefox on the Mac.

      Does anybody with access to JavaScript on Safari fancy porting it?

      Bill

      >
      > Sean
      >
      > On Feb 15, 2006, at 4:01 PM, Langdon W B wrote:
      >
      > > I have put a beta version which shows links between GP co-authors at
      > > http://www.cs.bham.ac.uk/~wbl/biblio/gp-coauthors/index.html
      > >
      > > If you have a moment to test it, I would welcome constructive
      > > comments.
      > >
      > > Thank you
      > >
      > > Bill
      > >
      > > Ps: I would also like help with adding and correcting the GP-bib.
      > > Please send _full_ citations (prefereably in BibTeX).
      > >
      > >
      > > W. B. Langdon,
      > > Computer Science,
      > > University of Essex
      > > Wivenhoe Park,
      > > Colchester CO4 3SQ, UK
      > > http://www.cs.essex.ac.uk/staff/W.Langdon/
      > > Spam blocking gmail
      > > Foundations of Genetic Programming
      > > http://www.cs.ucl.ac.uk/staff/W.Langdon/FOGP
      > > EuroGP Budapest 2006 http://www.evonet.info/eurogp2006
    • Langdon W B
      Dear Steve, Oh... its does, well it should do, it does on my machine. Which browser are you using? Placing the cursor over a dot should give the page to which
      Message 2 of 8 , Feb 17, 2006
        Dear Steve,
        Oh... its does, well it should do, it does on my machine.
        Which browser are you using?
        Placing the cursor over a dot should give the page to which it will jump
        (should you click) this usually includes the name of the author
        plus, after the #, the number of papers and number of co-authors.
        This uses the url not the alt text. I dont know if it could use alt.

        I guess its not as easy to use as I had hoped. I will expand the text
        description below the top graph.

        Bill

        > Hi Bill,
        >
        > I think it looks great. It would be really neat if you could somehow
        > more easily identify where an author might be... Perhaps, if you do
        > a `mouseover' on a dot (with a URL), the alt text gives a list of
        > last names, or some of them, that the dot represents?
        >
        > Cheers,
        > Steve
        >
        > On Thu, Feb 16, 2006 at 04:16:58PM +0000, Langdon W B wrote:
        > > > 1. Very interesting! Bill, I can't figure out what the function is
        > > > which defines the locations of the dots. Could you elaborate for us?
        > >
        > > Dear Sean,
        > > Many thanks for your prompt and helpful reply about:
        > > http://www.cs.bham.ac.uk/~wbl/biblio/gp-coauthors/index.html
        > >
        > > The graphs are created by graphviz' neato.
        > > www.research.att.com/sw/tools/graphviz/
        > > It tries to lay out the graph nicely, ie without stretching edges
        > > too much or with too many lines overlapping each other.
        > >
        > > > 2. What is the meaning of the graph that has no edges on it, just
        > > > dots of different sizes arrayed in a grid? What is the difference
        > > > between black and white dots? The very large center dot appears to
        > > > contain numerous people, sort of a Being John Malkovich thing!
        > >
        > > The top level graph is a summary of the whole bibliography.
        > > Each dot represents a connected subgraph (ie a bunch of people and
        > > their papers which have GP co-authors in common). The size of the dots
        > > is proportional to the number of entries in the bibliography.
        > >
        > > The fact their are so many dots is due to many singletons.
        > > Ie small groups who have published only a few papers together.
        > >
        > > I did not generate a graph for each case. In most cases the graphs
        > > would be quite trivial, linking together only 3 or 4 people. And there
        > > would have been hundreds of such graphs. Instead I applied an
        > > (arbitrary) cut off and only generated graphs with more than 15
        > > co-authors. These are indicated by red! Clicking on red dots pulls up
        > > the corresponding subgraph.
        > >
        > > Approx half of us are in the big red "John Malkovich" in the middle.
        > >
        > > > 3. Unfortunately, the program doesn't appear to run properly on
        > > > Safari. Works on Firefox on the Mac.
        > >
        > > Does anybody with access to JavaScript on Safari fancy porting it?
        > >
        > > Bill
        > >
        > > >
        > > > Sean
        > > >
        > > > On Feb 15, 2006, at 4:01 PM, Langdon W B wrote:
        > > >
        > > > > I have put a beta version which shows links between GP co-authors at
        > > > > http://www.cs.bham.ac.uk/~wbl/biblio/gp-coauthors/index.html
        > > > >
        > > > > If you have a moment to test it, I would welcome constructive
        > > > > comments.
        > > > >
        > > > > Thank you
        > > > >
        > > > > Bill
        > > > >
        > > > > Ps: I would also like help with adding and correcting the GP-bib.
        > > > > Please send _full_ citations (prefereably in BibTeX).
        > > > >
        > > > >
        > > > > W. B. Langdon,
        > > > > Computer Science,
        > > > > University of Essex
        > > > > Wivenhoe Park,
        > > > > Colchester CO4 3SQ, UK
        > > > > http://www.cs.essex.ac.uk/staff/W.Langdon/
        > > > > Spam blocking gmail
        > > > > Foundations of Genetic Programming
        > > > > http://www.cs.ucl.ac.uk/staff/W.Langdon/FOGP
        > > > > EuroGP Budapest 2006 http://www.evonet.info/eurogp2006
        > >
        > >
        > >
        > >
        > >
        > > Yahoo! Groups Links
        > >
        > >
        > >
        > >
        > >
        > >
        >
        > --
        > Dr. Steven GUSTAFSON
        > School of Computer Science & IT
        > University of Nottingham, UK
        > http://www.cs.nott.ac.uk/~smg
        >
        >
        > This message has been checked for viruses but the contents of an attachment
        > may still contain software viruses, which could damage your computer system:
        > you are advised to perform your own checks. Email communications with the
        > University of Nottingham may be monitored as permitted by UK legislation.
        >
      • Langdon W B
        Following suggestions I have made a number of changes: Per author pages containing all GP entries for that author (replaces just for first author). neato
        Message 3 of 8 , Feb 27, 2006
          Following suggestions I have made a number of changes:

          Per author pages containing all GP entries for that author
          (replaces just for first author).

          neato weights and edge lengths changed
          to give more stable co-authorship graphs.

          http://www.cs.bham.ac.uk/~wbl/biblio/


          Bill


          W. B. Langdon, Phone +44-1206-873291
          Computer Science, Fax: +44-1206-872788
          University of Essex
          Wivenhoe Park,
          Colchester CO4 3SQ, UK
          http://www.cs.essex.ac.uk/staff/W.Langdon/
          Spam blocking gmail
          Foundations of Genetic Programming
          http://www.cs.ucl.ac.uk/staff/W.Langdon/FOGP
          EuroGP Budapest 2006 http://www.evonet.info/eurogp2006
        • Sean Luke
          Hey, now I m even further away from the center! :-) Nice to see it working in Safari. Very cool. Sean
          Message 4 of 8 , Feb 28, 2006
            Hey, now I'm even further away from the center! :-)

            Nice to see it working in Safari. Very cool.

            Sean

            On Feb 27, 2006, at 7:48 AM, Langdon W B wrote:

            > Following suggestions I have made a number of changes:
            >
            > Per author pages containing all GP entries for that author
            > (replaces just for first author).
            >
            > neato weights and edge lengths changed
            > to give more stable co-authorship graphs.
            >
            > http://www.cs.bham.ac.uk/~wbl/biblio/
            >
            >
            > Bill
            >
            >
            > W. B. Langdon, Phone +44-1206-873291
            > Computer Science, Fax: +44-1206-872788
            > University of Essex
            > Wivenhoe Park,
            > Colchester CO4 3SQ, UK
            > http://www.cs.essex.ac.uk/staff/W.Langdon/
            > Spam blocking gmail
            > Foundations of Genetic Programming
            > http://www.cs.ucl.ac.uk/staff/W.Langdon/FOGP
            > EuroGP Budapest 2006 http://www.evonet.info/eurogp2006
            >
            >
            >
            >
            >
            >
            >
            >
            >
            >
            > Yahoo! Groups Links
            >
            >
            >
            >
            >
            >
          Your message has been successfully submitted and would be delivered to recipients shortly.