RE: [unrev-II] Augment + categories = OHS v0.1

From: Gil Regev (
Date: Thu Jun 22 2000 - 23:42:41 PDT

  • Next message: "Re: [unrev-II] Economics and the Garden of Eden"


    The problem with categories is that they tend to rigidify the knowledge
    repository, in which case I wonder how dynamic it will be. I was in a
    conference on trans-disciplinarity a few months ago and one of the
    presenters said that the the real threat to trans-disciplinarity was the
    hardening of categories. I a collaborative software (not Knoware) we did in
    our lab we moved from relying on categories to find information to using a
    good search tool. We figured that whatever categorization scheme we could
    come up with, it would be obsolete pretty fast. Having said that, I totally
    agree that categories are essential but they should be implemented in a way
    that preserves the dynamics of the system.

    In Knoware, relationships have no meaning for the software but they do have
    meaning to the users. The search tool searches for text in relationships as
    well as in concepts.

      -----Original Message-----
      From: Eugene Eric Kim []
      Sent: vendredi, 23. juin 2000 05:34
      Subject: [unrev-II] Augment + categories = OHS v0.1

      Based on the discussions we had at our meeting today and some of the
      discussions afterwards with Eric, Howard and others, I came up with the
      following conclusions.

      Fundamentally, our system consists of two things: nodes and links. The
      difference between our system and the previous two revisions of Doug's
      work is that our system allows us to categorize nodes. Categorizing nodes
      is crucial, because it provides an additional semantic layer that is
      crucial for knowledge management. Both the users of the system and the
      system itself can take advantage of this additional layer.

      For example, in order to adapt our system to accomodate IBIS-style
      discussions, we add categories such as "question," "answer," and
      "alternatives." With this information, we can search for all the
      questions pertaining to a particular topic of discussion. Or, we can
      write a system module that automatically compiles and presents an IBIS
      view based on the current question and answer nodes.

      Categorizing nodes allows us to provide some structure to discussion, even
      if the discussion itself is unstructured. For example, if we're trying to
      come up with a Use Cases document for a piece of software, I may propose
      five different Use Cases in three different e-mails. However, if these
      nodes are properly categorized in each node, then I can easily create a
      Use Cases view that shows all of the Use Cases in one view, regardless of
      when or where these were proposed.

      I think there's still one open question that needs to be resolved for
      version 0.1 of the OHS. That question is, are links categorizable? This
      relates to some of Eric's previous questions regarding relationships. In
      my opinion, a link is the way you represent a relationship. The question
      is, should you be able to specify categories for a link.

      My suspicion is yes. Even if this doesn't immediately affect system
      behavior, I think it is a useful attribute to have in the data
      structure. In Gil's Knoware, which he demonstrated last week, you can
      label relationships, but those labels have no meaning. However, it should
      be fairly easy to give those labels meaning in the system in the future,
      which is one of the things Gil said he planned on doing. I see no reason
      why we shouldn't take the same approach.


      +=== Eugene Eric Kim ===== =====
      | "Writer's block is a fancy term made up by whiners so they
      +===== can have an excuse to drink alcohol." --Steve Martin



    ---------------------------------------------------------------------------- -- Community email addresses: Post message: Subscribe: Unsubscribe: List owner:

    Shortcut URL to this page:

    This archive was generated by hypermail 2b29 : Thu Jun 22 2000 - 23:52:38 PDT