Re: [unrev-II] Lucid Thinking

From: Jack Park (jackpark@thinkalong.com)
Date: Mon Sep 24 2001 - 11:54:52 PDT

  • Next message: Jack Park: "[unrev-II] Fwd: [issues] An American voice in the midst of our shared crisis"

    At 01:57 PM 9/24/2001 -0400, you wrote:
    >At 09:37 AM 9/24/01 -0700, Jack Park wrote:
    > >Incremental + decentralized fashion. You mean, a sourceforge.net project
    > >with lots of builders?
    >
    >This would be nice. I was thinking more along the lines of several
    >cooperating sf.net projects.
    >By what I vaguely called incremental, I meant that the project could be
    >released in stages, each of which would be useful. This, as opposed to
    >waiting until all the functionality was in place before letting users try
    >it out.
    >

    Cooperating sf projects is fine with me. To make that work, it seems to me
    that there will need to be some kind of core API around which all projects
    operate. That implies that there is a core project, and all the others are
    some kind of plugin or whatever. Heck! In some sense, it seems to me that
    NODAL would make the ideal core project since it deals with the data --
    http://sourceforge.net/projects/nodal. Good place to start, though there
    does not exist any software yet to play with. Acute "fire-em-up-itis" makes
    me want to look elsewhere.

    Nexist has a plug in architecture: you just write to a particular api, then
    drop the classes into a particular directory and add a reference to that
    new directory to an XML properties file and reboot the system. There are
    probably other drop-in architectures out there (Protege, Jext, and JEdit
    come to mind) that may be more refined than what I have written.

    > >I'm all for adding stuff to Lucid itself; I just read the Lucid source code
    > >-- looks like it can be done. In fact, that would be the fastest way to
    > >get such a prototype running.
    > >
    > >In the end, however, OHS-dev is committed to a non-GPL license. Starting
    > >fresh, not infringing on Steve's code, etc, seems appropriate.
    >
    >What's your philosophy of open source licenses? I chose the Apache one,
    >because I sensed that there was no turning back from gnu. Also, I didn't
    >want a licence that was too long, or that prevented the code from being
    >reused. Is there some anti-GPL propaganda out there that I could read?

    Google "OHS license" and you will find lots of discussion on this topic.

    > >Personally, I can imagine my BLOG at such a portal having both private and
    > >public data -- that which I think others might enjoy or find useful, not to
    > >mention contributing to the mining activity, and that which is private, not
    > >available to the mining tools. It may be that my client application will,
    > >at once, log into the web, and log into a local database where the private
    > >stuff resides. I can then work offline and later log in and synch
    > >databases as a way of publishing to the portal.
    >
    >Sounds like a good idea, but I bet people would be paranoid about privacy
    >issues.

    Privacy, authentication, etc, are all enormous issues. I'm inclined to
    start thinking about using one of the J2EE platforms (e.g. JBoss) as a
    starting point in order to inherit the necessary security tools. JBoss is
    LGPL, which means its license will not infect ours. Enhydra recently
    pulled their J2EE offering from open source because of licensing
    issues. There are several other J2EE platforms, including uPortal which is
    completely free. Grant Bowman has compiled a list of these projects at
    http://nexist.sourceforge.net/groupware.html

    > It's easiest to keep a blog public. (why the all caps by the way,
    >it's not an acronym, but derived from 'weblog')

    I knew that :-)

    >Also, check out http://www.advogato.org/ it's kind of a slashdot with
    >personal journals.
    >
    > >One area that the mining activity will contribute will be in the area of
    > >discovery of "loosely related" links.
    > >One area the mining activity will have to pay attention to is that of
    > >detection of cycles in the graph. Graph cycles, as most software
    > >developers know, result in the "Branch to South Succotash" effect.
    >
    >Cycles don't scare me. Check out my graph layout example
    >http://www.touchgraph.com I've got a first draft of locality working.
    >
    > >As to the "global brain" nature, I think there is merit in that thought. If
    > >you read the papers and activities of that group, as I have forwarded here
    > >before, you can see that the network-like configuration is favored.
    > >
    > >Jack
    >
    >It's all about the network-like configuration.
    >
    >--Alex

    Yup!

    Jack

    ------------------------ Yahoo! Groups Sponsor ---------------------~-->
    Get your FREE VeriSign guide to security solutions for your web site: encrypting transactions, securing intranets, and more!
    http://us.click.yahoo.com/XrFcOC/m5_CAA/yigFAA/IHFolB/TM
    ---------------------------------------------------------------------~->

    Community email addresses:
      Post message: unrev-II@onelist.com
      Subscribe: unrev-II-subscribe@onelist.com
      Unsubscribe: unrev-II-unsubscribe@onelist.com
      List owner: unrev-II-owner@onelist.com

    Shortcut URL to this page:
      http://www.onelist.com/community/unrev-II

    Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/



    This archive was generated by hypermail 2.0.0 : Mon Sep 24 2001 - 11:49:32 PDT