Re: [unrev-II] Digest Number 108

From: Nancy_Cooley@ferris.edu
Date: Sun Apr 30 2000 - 15:04:32 PDT

  • Next message: Eric Armstrong: "Re: [unrev-II] Experimental Zope Unrev-II Server (Was Re: Thoughts)"

    I have been a lurker on the unrev-11 listserv, after realizing that my
    administrative duties would prevent me from participating fully in the the
    on-line colloquium. Today, however, I could not resist responding to the
    posting by Henry. I've never before seen such tautology, and as I read it
    aloud to my husband, we roared with laughter. Henry's was truly an
    excellent "tongue-in-cheek" treatise, reinforcing the stereotype of
    academics with too much time on their hands who overanalyze and obscure the
    meaning of even the simplest concepts. Thank you for providing such
    levity!!

    unrev-II@egroups.com on 04/28/2000 10:28:52 AM

    Please respond to unrev-II@egroups.com

    To: unrev-II@egroups.com
    cc:

    Subject: [unrev-II] Digest Number 108

    ------------------------------------------------------------------------
    Make music with anyone, anywhere, through FREE Internet
    recording studio software. FREE software download!
    http://click.egroups.com/1/3734/3/_/444287/_/956932133/
    ------------------------------------------------------------------------

    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
    ------------------------------------------------------------------------

    There are 13 messages in this issue.

    Topics in this digest:

          1. Knowledge Representation (wasRe: Jack Park's "10 Step" Program)
               From: "Jack Park" <jackpark@verticalnet.com>
          2. Re: GOOD: Traction, by Twisted Systems (on browser neutrality)
               From: cjn@twisted-systems.com
          3. Re: Towards an atomic data structure.
               From: Henry van Eyken <vaneyken@sympatico.ca>
          4. Use case scenarios for OSS development
               From: Lee Iverson <leei@ai.sri.com>
          5. Re: Re: Towards an atomic data structure
               From: "Sandy Klausner" <klausner@cubicon.com>
          6. Re: GOOD: Traction, by Twisted Systems (on browser neutrality)
               From: Eric Armstrong <eric.armstrong@eng.sun.com>
          7. Re: Re: Towards an atomic data structure
               From: Eric Armstrong <eric.armstrong@eng.sun.com>
          8. Re: Towards an atomic data structure.
               From: "Sandy Klausner" <klausner@cubicon.com>
          9. ResearchIndex http://www.researchindex.com/
               From: NABETH Thierry <thierry.nabeth@insead.fr>
         10. Re: Use case scenarios for OSS development
               From: Eric Armstrong <eric.armstrong@eng.sun.com>
         11. Re: Re: Towards an atomic data structure
               From: "Sandy Klausner" <klausner@cubicon.com>
         12. All Colloquium transcripts available
               From: "Henry van Eyken" <vaneyken@sympatico.ca>
         13. A small experiment to help students
               From: "Henry van Eyken" <vaneyken@sympatico.ca>

    ________________________________________________________________________
    ________________________________________________________________________

    Message: 1
       Date: Thu, 27 Apr 2000 09:04:12 -0700
       From: "Jack Park" <jackpark@verticalnet.com>
    Subject: Knowledge Representation (wasRe: Jack Park's "10 Step" Program)

    From: Eric Armstrong <eric.armstrong@eng.sun.com>
    <snippage/>
    > There is an interesting possibility that there is. Looking over the
    > Traction offering and comparing that with IBIS concepts led to the
    > minor epiphony that the simple act of categorizing information nodes
    > according to some (agreed upon) schema is in essence a knowledge-
    > abstraction process. I'm convinced that Traction is absolutely on the
    > right track with respect to categorization -- IBIS is an
    > easily-definable subset of their system. Where they fall down is with
    > respect to document hierarchy, but they've made a contribution (to my
    > thinking, at least)
    > with respect to categories.
    >

    Kindly take the time to elucidate that which convinces you.

    > What is interesting, here, is the concept that the whole "knowledge
    > management" domain exists in the realm of the categories, where
    > "documents" are found among the information nodes. If it makes sense
    > to think of knowledge management in those terms, then we can conceivably
    > apply some interesting abstract manipulations to "knowledge", where
    > knowledge means a common (or possibly standard) set of categories, and
    > where the underlying information is unique to each domain.
    >

    Standard categories: Ha!
    Lakoff wrote a book called "Women, Fire, and Dangerous things," which, as I
    recall, were the primary categories of some aboriginals. The Maoris include
    reproduction in Earth Science. They do this because they see the
    cycle of life and always bury a placenta next to a tree. So much for
    logical categories.

    > For example, the category "argument for" can be applied to information
    > nodes in a biological sciences domain, or to one in an art analysis
    > domain. The category is a form of meta-data that is independent of the
    > information content.
    >
    > Now, given a standard set of categories, it might be possible to begin
    > describing category-relationships. That would produce the property of
    > abstract reasoning, that was independent of the problem domain.
    >
    > I keep thinking in terms of "implies". If there is some way to add the
    > meta-data "implies" in the category space, then automated reasoning
    > becomes possible.
    >
    > Example: at the initial writing, node A is written, as well as node B,
    > with the "implies" attribute linking the two. Later, someone adds C as
    > an implication of B. The system can now deduce that A implies C --
    > regardless of the information content contained in the nodes.
    >
    > Perhaps the "category" for such a system is "implication". Categorizing
    > B as an implication then requires pointing to A, to identify the node
    > from which B was derived. The symettric relationship can then also be
    > added -- call it "motivator", or some such. (If there is a logic term
    > for it, I've forgotten it.)
    >
    > There might also be categories for preconditions, requirements, and
    > what have you, all of which would allow for fairly sophisticated
    > reasoning engines to be built on top of the fundamental structures.
    > [There are also evaluations -- of node content as well as the logic
    > employed...]
    >
    > The interesting point to all this is that the "DKR" becomes a layer
    > of abstraction built on the OHS, where the categorization-capability
    > is already built into the OHS.
    >

    Actually, as I see it, the DKR is an API that the OHS supports. Thusly,
    most of the use cases should discuss what one wants to do at the OHS, and
    DKR use cases will define that API.

    In all of the above, reference is made to categories, relations,
    implication, and so forth. It seems to me that mankind has been discussing
    this since Aristotle, maybe before. It also seems to me that nobody has
    achieved "the solution tres grande." It further seems to me that nobody
    ever
    will. Therefore, we must take pains to define just what the DKR is
    intended
    to represent and manipulate, then give it our best shot. Lenat has made an
    enormous effort along these lines with CYC. There is even a public domain
    version of CYC evolving. This all stems from the fact that Eurisko,
    powerful
    as it was, never went very far simply because it lacked common sense. Guha,
    McCarthy, and lots of others are making careers trying to figure out how to
    represent common sense. Guha is not quite the champion of RDF, McCarthy
    remains the champion of a variety of logical formalisms.

    Automated reasoning is well documented. It does work for limited domains.
    In
    fact, just about anything will work when operating in a sufficiently
    constrained domain. DKR, however, wants to take on the universe, and
    everything. Not particularly constrained, IMHO.

    Humans talk in qualitative terms for normal conversation. We will use
    qualitative descriptions of probabilistic issues (often, seldom, etc), we
    use qualitative descriptions of fuzzy issues (tall, fat, ...), and we use
    crisp terms to describe other things (absolutely, never, ...). IMHO, the
    game is to invent a KR scheme that lets us partially automate handling of
    all kinds of representations. Zadeh has recently (at KR 2000) proposed
    something he calls "precisified natural language." (PNL) This appears to be
    a highly constrained natural language. Telling jokes with PNL would not be
    easy, but describing the evolution of an hiv infection would.

    Daphne Koller at Stanford has developed what appears to be a seamless
    integration of bayesian and description logics. One of her students, now
    at
    Harvard, is adding linguistics to that mix. Maybe, just maybe, they are on
    to something we need to understand better.

    I think that I am saying that the DKR warrants a deeper look at KR than is
    suggested by appeals to categories, relations, and logic. We are trying to
    represent things which are complex. Newtonian mechanics and reductionist
    thinking will not get us there. Indeed, there may not exist an atomic
    structure capable of supporting our dream.

    ________________________________________________________________________
    ________________________________________________________________________

    Message: 2
       Date: Thu, 27 Apr 2000 16:07:58 -0000
       From: cjn@twisted-systems.com
    Subject: Re: GOOD: Traction, by Twisted Systems (on browser neutrality)

    > I didn't say it to everyone but I'm also bothered that its not
    browser
    > neutral. Designing for one browser is just lazy, stupid, or both.

    I realize this reply is fairly belated, for which I apologize.

    I agree with Jon entirely. But we regularly use Traction with
    Netscape, Lynx, Proxiweb and Avant Go (on Palm), Opera and Internet
    Explorer. And occasionally with w3 mode in Emacs. It supports all
    these browsers.

    However, we saw an opportunity to improve responsiveness and
    information density using DHTML features (such as layers and
    JavaScript); Traction uses the client in the HTTP header to determine
    which interface to serve. For browsers that support DHTML features,
    such as Netscape and IE, Traction defaults to using those features.
    Like many options within Traction, this can be disabled as a
    preference.

    During the demo I gave, I used IE because its full-screen feature
    lets people concentrate on the web interface, and lets me use larger
    fonts to display on the projector. I didn't realize I was conveying
    the impression that IE was the only browser Traction supports.

    -Chris

    ________________________________________________________________________
    ________________________________________________________________________

    Message: 3
       Date: Thu, 27 Apr 2000 14:33:56 -0400
       From: Henry van Eyken <vaneyken@sympatico.ca>
    Subject: Re: Towards an atomic data structure.

    Or "Can a DKR bridle unbridable thought?"
    Or "Loom of frustration."
    Or "Re: Knowledge representation."
    Or "Should I really inflict this piece on anyone?"

    Wednesday morning, April 26. -- I just began reading this thread and almost
    immediately something began to revolt in me. And yet, I must (as indeed I
    do)
    respect the opinions of people who have spent much of their lives informing
    themselves in the best tradition of educated society.

    Before moving on to the next paragraph, let's share the observation that my
    opening sentence contains, contrary to the dogma of "one paragraph, one
    notion,"
    at least eight nodes of information, i.q. I; just; "began"; "to read";
    "almost
    immediately"; "something"; "to revolt"; "in me" -- a breakdown which, to be
    sure, is just one way of splitting the atomic ideal. And I haven't yet
    completed
    the paragraph. Also observe that this very last sentence, just written,
    implies
    a node of information that is hidden right after the word "completed." That
    implied node of information is the action to which the noun "paragraph" was
    subjected to. (Tiresome, am I not?)

    Eric's first post on the subject matter of atomization of language, or
    thought
    .... Oh, I must stop again. There mostly does not seem to be a one-to-one
    relationship between the components (free radicals, atoms, molecules,
    crystals,
    etc.) of thought and the components of language. Language might be seen as
    a
    conduit of thought, and, like my back-country telephone line, a most
    important,
    but far from a perfect conduit. Take. for example, the word "mankind"
    which,
    when taken out of context, may well signal an antifeminist attitude.
    "Mankind,"
    therefore," is an instance where we find at least two meanings within a
    single
    word. Isotopes, anybody?

    What upset me immediately is the top-down approach embodied by the class
    object-oriented terminology. As applied to text (and maybe to the potential
    of
    computing as well), I question the usefulness of a Simon-pure
    object-oriented
    approach. A top-down approach, it seems to me, is bridling the unbridable,
    a
    tool that communicates a hard-to-discipline melange of logical order and
    emotions, of what wells from the conscious and the levels of the
    less-than-conscious. Let me show my concern by trying to recapture some of
    the
    fleeting thoughts that went through my mind when reading the paragraphs
    under
    the heading "Text Nodes."

    I quote: The fundamental unit of a DKR is an item of information. Since the
    ideal in writing is to have "one idea per paragraph", an "information node"
    can
    be thought of as a paragraph of text. Headings stand apart from other text,
    as
    well, so a heading is a special (short) paragraph, or information node.

    The first sentence quoted is a postulate that probably won't stand the test
    of
    scrutiny. The second sentence turns the tables on the postulate; and that
    quite
    aside from the stated ideal in writing. Beaudelaire, Joyce, Conrad, Schlink
    are
    but some of the people whose celebrated works are what they are because
    their
    ideal is to stuff in a little extra. As do children and salesmen, and,
    well,
    don't we all? As for headings being nodes of information, may I invite you
    back
    to the smorgasbord of headings at the beginning of this piece. Which one
    best
    conveys what I am writing about?

    At this point, I ought to realize that I am reading Eric's post out of
    (his)
    context and putting it into mine. In other words, the meaning of text is
    subject
    to environmental influence. (Geez, I think I could expand that last
    sentence
    into a book.) I also understand that precisely because of this problem,
    language
    must contain something that is not just "purely informational." It must
    contain
    a funnel of words to guide the reader or listener coming in from the cold
    as
    quickly as possible to the point the emitter is trying to make. That funnel
    of
    words has been called redundancy. I understand from having read a couple of
    atoms from Shannon and Wheeler that English is about 30 percent redundant;
    redundant, that is, from the point of view of its central messages, but an
    essential redundancy to guide the innocent to the nectar of an attempted
    communication.

    How then, with these notions in my mind, may I feel compelled to keep on
    reading? But I continued anyway, forcing myself.

    Quoting: Node behaviors are defined in a class (object template). Every
    text
    node must contain an attribution -- a pointer to the author, or an
    identifying
    string. A copy of that node may be edited, which suggests the need for a
    split
    operation, for example. After node is split into one or more fragments, and
    edit
    operation could replace some fragments or insert new ones that have a
    different
    author. Some of the operations appropriate to a node might therefore
    include
    split, delete, replace, and insert.

    My immediate problem, after blindly sliding through the first sentence
    (because
    it lacks the, for me, prerequisite redundancy), is "a pointer to the
    author."
    Individual authorship, of course, is a concept that belongs to the class
    "culture." Ancient Greek culture did not recognize this kind of authorship.
    Homer rhapsodized, literally meaning that he stitched together. He stitched
    descriptions that were fragments of other tales to create his tale. He then
    added rhyme to reason for staunching his memory. (No art of poetry for him;
    just
    plain craftmanship.) A good thing that Eric added "or an identifying
    string."

    Luck has it that I stayed the course for at this point I find Eric
    introducing
    ideas that capture my attention. I might quickly add to his list of editing
    operations on a node of information: re-emote and recontextualize (ain't
    she
    sweet?). It is well to remember that re-emoting may change "objective
    meaning"
    (??) totally. A simple re-ordering of wordsmay efffect this. "Just so," you
    may
    think, whereas the editor conscienciously meant to be oh "so just."

    At this point, my mind fleetingly dwells on translation. How simple would
    it be
    to translate from one language into another if language could be clearly
    atomized. ("How simple would it be," I wrote. Not "How simple it would
    be.")

    At this point, I must realize, I think, that the kind of text Eric writes
    about
    (no, I didn't say wrote!) is not natural language, not even a transcript of
    natural thoughts. He is writing about formalized transcripts of some sort
    of
    culturally bridled thoughts. Among these are the, supposedly
    redundancy-free
    languages of mathematics and computer programming. And, perhaps, of
    zealots, who
    tend to consider their ethics so purified from redundancy as to justify an
    attitude of "my way or no way."

    Languages of scientific and technological cultures might be less redundant
    than
    natural language -- and easier to translate. The disciplined listener needs
    only
    half the words required by an undisciplined one, Which brings up a concern
    one
    should have with public DKRs.

    Quoting: Note that when the node is split, two objects exist where one did
    before. Every node must therefore be capable of being the root of a
    subtree.
    Although it may start out life as a simple node that contains or points to
    an
    item of text, it must also be capable of pointing to a list of text
    elements.
    (That list might also include markup elements, like HTML bold tags: <b>.)
    Since
    each item in that list may itself point to a list of subitems, the
    resulting
    structure is a tree.

    Interesting indeed. At one time, I dwelled on the use of adjectives as a
    means
    of splitting a node and on adverbs as means of further modulation. I
    mentally
    compared their use to the words we have for color. The little red engine
    that
    could. Blue moon. But primary colors don't suffice beyond childhood.
    Orangy,
    brown-gray become necessary additions. For the artist we have special
    chromatograms. For scientists we have that concretized abstraction of
    vibrational frequencies. Serving roles not unlike adjectives are hyphens,
    and I
    have wondered whether we might introduce for greater expressive precision a
    more
    potent hyphen by borrowing the equal sign, e.g. "a brown=gray-colored
    object"
    would make a somewhat more precise statement than "a brown-grayish object"
    once
    propagated by the discerning.

    And at this point, I must ask myself, what is the best use of my time and
    how
    much may I impose on others? And how just am I to the author who began this
    thread, unquestionably a man one cannot but hold in high esteem?

    I feel so frustrated.

    Better go outdours for some fresh air. With oxygen of a particularly
    refreshing
    molecular form.

    Ever so diatomic in vibrational embrace.

    Henry

    P.S. I wrote this yesterday morning and decided not to put it on the forum.
    Until I read Jack Park's piece a moment ago. Funny how different our two
    pieces
    are, and still so very much the same. That's language for you.

    Jack Park wrote about
    Knowledge Representation (wasRe: [unrev-II] Jack Park's "10 Step" Program):

    ________________________________________________________________________
    ________________________________________________________________________

    Message: 4
       Date: Thu, 27 Apr 2000 11:58:44 -0700
       From: Lee Iverson <leei@ai.sri.com>
    Subject: Use case scenarios for OSS development

    Slow as usual to get this stuff to everybody, but here it is:

         http://www.ai.sri.com/~leei/OHS/ossusecases.html

    -------------------------------------------------------------------------------

    Lee Iverson SRI International
    leei@ai.sri.com 333 Ravenswood Ave., Menlo Park CA 94025
    http://www.ai.sri.com/~leei/ (650) 859-3307

    ________________________________________________________________________
    ________________________________________________________________________

    Message: 5
       Date: Thu, 27 Apr 2000 13:32:38 -0700
       From: "Sandy Klausner" <klausner@cubicon.com>
    Subject: Re: Re: Towards an atomic data structure

    Henry van Eyken
    "I question the usefulness of a Simon-pure object-oriented approach. A
    top-down approach, it seems to me, is bridling the unbridable, a tool that
    communicates a hard-to-discipline melange of logical order and emotions, of
    what wells from the conscious and the levels of the less-than-conscious."
    "I think, that the kind of text Eric writes about is not natural language,
    not even a transcript of natural thoughts. He is writing about formalized
    transcripts of some sort of culturally bridled thoughts. Among these are
    the, supposedly redundancy-free languages of mathematics and computer
    programming."
    "Quoting: Note that when the node is split, two objects exist where one did
    before. Every node must therefore be capable of being the root of a
    subtree.
    Although it may start out life as a simple node that contains or points to
    an item of text, it must also be capable of pointing to a list of text
    elements. (That list might also include markup elements, like HTML bold
    tags: <b>.) Since each item in that list may itself point to a list of
    subitems, the resulting structure is a tree."
    The following is a repeat of my Sunday, April 24 posting:
    The DKR team has identified two distinct levels of information abstraction
    that require development to achieve the group's goals. The underlying
    abstraction appears to be based upon a general system cognitive model based
    upon deterministic behavior that can machine execute. This technology
    model
    could be used as a foundation to design and implement "An interactive tool
    for discussion and deliberation* that records decisions and their
    rationales in a way that allows the knowledge gained in the process to be
    applied to future projects."
    To fully achieve the interactive tool goal, the fundamental capabilities in
    the underlying technology model must include a robust way to traverse,
    edit,
    read, and write untyped text. In addition, there needs to be a way to
    intelligently analyze the text in interesting ways to determine fundamental
    semantics in the symbolic patterns and link these patterns to other
    passages
    to anywhere in the web. This text may be linked to typed atomic data that
    may itself be composed into typed molecular data representing pictures,
    sounds, and other rich multimedia information. All this information may
    itself exist as part of a data structure within a domain object within a
    system.
    The SGML community has a long history of developing ways to markup
    documents
    to capture semantic knowledge embedded in strings. As processing
    requirements become more sophisticated, new ways of managing this
    complexity
    need to be developed. One possible solution is to move to a "clear
    document
    model." This model separates concerns by parsing the clear text from the
    markup information. The clear text is parsed into a collection of linked
    character nodes, while one or more composite structure processors maintain
    position and range links into the clear text collection. Each processor
    may
    have specialized behavior to analyze and hold semantic information on
    format, organization, navigation, narrative, reference, graphic control,
    publication, and filters. The model must be able to allow clear text
    editing
    while automatically maintaining the processor links into the clear text
    collection. Such a model would be able to manage the requirements for a
    robust DKR environment.
    New Posting
    Conventional markup language technology cannot effectively cope with the
    demands required for a robust DKR environment. The clear document model
    that
    we have developed has been successful applied to several domain-specific
    applications by others. We have generalized the mechanisms behind this
    model
    into a novel Graphical Language technology that can effectively manage this
    level of system complexity. Your right Henry, the conventional "Simon-pure
    object-oriented approach" will not be effective. But, any technology that
    can augment human intelligence must be based upon a concrete and
    deterministic computational model. Perhaps unrev-II should take a good look
    at what Cubicon has to offer the community.
    Sandy Klausner
    klausner@cubicon.com

    [This message contained attachments]

    ________________________________________________________________________
    ________________________________________________________________________

    Message: 6
       Date: Thu, 27 Apr 2000 14:22:51 -0700
       From: Eric Armstrong <eric.armstrong@eng.sun.com>
    Subject: Re: GOOD: Traction, by Twisted Systems (on browser neutrality)

    Thanks, Chris.

    "We can use anything" probably makes sense as a prominent
    bullet point in the presentation. You never said anything
    to indicate that it *was* a one-browser system, but not
    hearing that wasn't led one to suspect...)

    ________________________________________________________________________
    ________________________________________________________________________

    Message: 7
       Date: Thu, 27 Apr 2000 14:32:34 -0700
       From: Eric Armstrong <eric.armstrong@eng.sun.com>
    Subject: Re: Re: Towards an atomic data structure

    Sandy Klausner wrote:
    >
    > ... As processing requirements become more sophisticated, new ways of
    > managing this complexity need to be developed. One possible solution
    > is to move to a "clear document model." This model separates concerns
    > by parsing the clear text from the markup information.
    > ^^^^^^^
    I think you mean "separating".

    > The clear text
    > is parsed into a collection of linked character nodes, while one or
    > more composite structure processors maintain position and range links
    > into the clear text collection. Each processor may have specialized
    > behavior to analyze and hold semantic information on format,
    > organization, navigation, narrative, reference, graphic control,
    > publication, and filters.
    >
    Can you give a short example that shows one or more of these, and how
    they would work together? What are narrative, reference, and graphic
    control semantic information, anyway? What happens when you change one
    set of external links? For example if you change the organization, what
    happens to to the others?

    > The model must be able to allow clear text editing while
    > automatically maintaining the processor links!! into
    > the clear text collection.
    >
    How can that be done? How does the system know when I add a new word
    whether it is part of a heading or the paragraph that follows it. Or
    whether it is inside or outside of a bolded section. There must be
    dozens of "end points" for which the proper location for an insertion is
    indeterminable.

    ________________________________________________________________________
    ________________________________________________________________________

    Message: 8
       Date: Thu, 27 Apr 2000 12:48:38 -0700
       From: "Sandy Klausner" <klausner@cubicon.com>
    Subject: Re: Towards an atomic data structure.

    Henry van Eyken
    "I question the usefulness of a Simon-pure object-oriented approach. A
    top-down approach, it seems to me, is bridling the unbridable, a tool that
    communicates a hard-to-discipline melange of logical order and emotions, of
    what wells from the conscious and the levels of the less-than-conscious."
    "I think, that the kind of text Eric writes about is not natural language,
    not even a transcript of natural thoughts. He is writing about formalized
    transcripts of some sort of culturally bridled thoughts. Among these are
    the, supposedly redundancy-free languages of mathematics and computer
    programming."
    "Quoting: Note that when the node is split, two objects exist where one did
    before. Every node must therefore be capable of being the root of a
    subtree.
    Although it may start out life as a simple node that contains or points to
    an item of text, it must also be capable of pointing to a list of text
    elements. (That list might also include markup elements, like HTML bold
    tags: <b>.) Since each item in that list may itself point to a list of
    subitems, the resulting structure is a tree."
    The following is a repeat of my Sunday, April 24 posting:
    The DKR team has identified two distinct levels of information abstraction
    that require development to achieve the group's goals. The underlying
    abstraction appears to be based upon a general system cognitive model based
    upon deterministic behavior that can machine execute. This technology
    model
    could be used as a foundation to design and implement "An interactive tool
    for discussion and deliberation* that records decisions and their
    rationales in a way that allows the knowledge gained in the process to be
    applied to future projects."
    To fully achieve the interactive tool goal, the fundamental capabilities in
    the underlying technology model must include a robust way to traverse,
    edit,
    read, and write untyped text. In addition, there needs to be a way to
    intelligently analyze the text in interesting ways to determine fundamental
    semantics in the symbolic patterns and link these patterns to other
    passages
    to anywhere in the web. This text may be linked to typed atomic data that
    may itself be composed into typed molecular data representing pictures,
    sounds, and other rich multimedia information. All this information may
    itself exist as part of a data structure within a domain object within a
    system.
    The SGML community has a long history of developing ways to markup
    documents
    to capture semantic knowledge embedded in strings. As processing
    requirements become more sophisticated, new ways of managing this
    complexity
    need to be developed. One possible solution is to move to a "clear
    document
    model." This model separates concerns by parsing the clear text from the
    markup information. The clear text is parsed into a collection of linked
    character nodes, while one or more composite structure processors maintain
    position and range links into the clear text collection. Each processor
    may
    have specialized behavior to analyze and hold semantic information on
    format, organization, navigation, narrative, reference, graphic control,
    publication, and filters. The model must be able to allow clear text
    editing
    while automatically maintaining the processor links into the clear text
    collection. Such a model would be able to manage the requirements for a
    robust DKR environment.
    New Posting
    Conventional markup language technology cannot effectively cope with the
    demands required for a robust DKR environment. The clear document model
    that
    we have developed has been successful applied to several domain-specific
    applications by others. We have generalized the mechanisms behind this
    model
    into a novel Graphical Language technology that can effectively manage this
    level of system complexity. Your right Henry, the conventional "Simon-pure
    object-oriented approach" will not be effective. But, any technology that
    can augment human intelligence must be based upon a concrete and
    deterministic computational model. Perhaps unrev-II should take a good look
    at what Cubicon has to offer the community.
    Sandy Klausner
    klausner@cubicon.com

    [This message contained attachments]

    ________________________________________________________________________
    ________________________________________________________________________

    Message: 9
       Date: Fri, 28 Apr 2000 09:51:28 +0200
       From: NABETH Thierry <thierry.nabeth@insead.fr>
    Subject: ResearchIndex http://www.researchindex.com/

    Hello,

    I wanted to raise to your attention what appear as a very interesting
    active search document system such as:

    Autonomous Citation Indexing
    (ACI) automates the construction of citation indexes

    Awareness and tracking
    ResearchIndex provides automatic notification of new citations to given
    papers, and new papers matching a user profile.

    Autonomous location of articles
    ResearchIndex uses search engines and crawling to efficiently locate
    papers on the Web.

    ResearchIndex (formerly CiteSeer)
    http://www.researchindex.com/

    ResearchIndex engine is Freely available
    (The full source code of ResearchIndex is available at no cost for
    non-commercial use.)

    Thierry Nabeth
    Research Fellow
    INSEAD CALT (the Centre for Advanced Learning Technologies)
    http://www.insead.fr/CALT/

    PS:
    For your information, I am currently working on an active
    "collaborative web referencing system" using the Zope technology.

    I want to transform my Encyclopedia of links
    http://www.insead.fr/CALT/Encyclopedia/
    into a more active, personalized and collaborative system.

    However, I am still at a very early stage of the project,
    and I do not have currently a lot of time to dedicate to this
    project.

    Found on the UMBC agentslist
    <http://www.cs.umbc.edu/agentslist>
    ------------------------------

    Date: Thu, 4 May 2000 09:00:26 -0700
    From: "Ahmad Abdollahzadeh Barforoush" <ahmad@ce.aku.ac.ir>
    Subject: Re: Paper available: Indexing and Retrieval of Scientific
    Literature

    Dear Karla,
    A good place that you can search for agent related papers is Research
    Index,CiteSeer, (www.researchindex.com). You can search for "agent" and
    find
    a good summary of published articles (about 8800).
    Best,
    R. Shirazi.

    >Hi,
    >I am making my PhD and I am needing recent bibliography in intelligent
    >agents (learning), if possible published in periodic of the type IEEE
    >Transactions.
    >Can anybody help?
    >Thanks,
    >Karla Figueiredo

     --
    See <http://www.cs.umbc.edu/agentslist> for list info & archives.

    ------------------------------

    [This message contained attachments]

    ________________________________________________________________________
    ________________________________________________________________________

    Message: 10
       Date: Thu, 27 Apr 2000 14:23:31 -0700
       From: Eric Armstrong <eric.armstrong@eng.sun.com>
    Subject: Re: Use case scenarios for OSS development

    Thanks, Lee. I think this really begins to get us where
    we need to go.

    Lee Iverson wrote:
    >
    > Slow as usual to get this stuff to everybody, but here it is:
    >
    > http://www.ai.sri.com/~leei/OHS/ossusecases.html
    >
    >
    -------------------------------------------------------------------------------

    > Lee Iverson SRI International
    > leei@ai.sri.com 333 Ravenswood Ave., Menlo Park CA
    > 94025
    > http://www.ai.sri.com/~leei/ (650) 859-3307
    >
    > ----------------------------------------------------------------------
    >
    > ----------------------------------------------------------------------
    > 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

    ________________________________________________________________________
    ________________________________________________________________________

    Message: 11
       Date: Thu, 27 Apr 2000 15:04:23 -0700
       From: "Sandy Klausner" <klausner@cubicon.com>
    Subject: Re: Re: Towards an atomic data structure

    > The clear text
    > is parsed into a collection of linked character nodes, while one or
    > more composite structure processors maintain position and range links
    > into the clear text collection. Each processor may have specialized
    > behavior to analyze and hold semantic information on format,
    > organization, navigation, narrative, reference, graphic control,
    > publication, and filters.
    >
    Can you give a short example that shows one or more of these, and how
    they would work together? What are narrative, reference, and graphic
    control semantic information, anyway? What happens when you change one
    set of external links? For example if you change the organization, what
    happens to to the others?

    Let me clarify this text processor classifcation a little further. Any
    particular unit of content can have one of a number of characteristics that
    we can group into several categories: format, organization, navigation,
    narrative, reference, and metadata:
    · Format obviously is the way the text looks.
    · Organization is the way content is grouped into coherent units.
    · Navigation is text that acts as finding aids -- tables of contents,
    indices, and the like.
    · Narrative is content-bearing structures; paragraphs, lists, tables, etc.
    · Reference is text that acts as a gateway to other, related content.
    · Metadata is like the Z-axis of your documents. It is the information in
    front
    of or behind the words on display that makes it possible for your to
    exploit
    the
    content in novel and sophisticated ways. For example, metadata about a
    publication can be used to automate assembly and publishing. Metadata about
    graphics can allow the print product to use full-page TIFF images while the
    Web
    site uses the thumbnail JPEGs linked to a full-sized images.
    The issues of changes to a set of external links and internal clear text
    edits are related. I will take more time to explain clearly (perhaps even
    by
    using some pictures) as to how the Cubicon architecture manages these type
    of dynamic changes to a composite data structure.
    Sandy Klausner
    klausner@cubicon.com

    [This message contained attachments]

    ________________________________________________________________________
    ________________________________________________________________________

    Message: 12
       Date: Fri, 28 Apr 2000 11:26:48 -0000
       From: "Henry van Eyken" <vaneyken@sympatico.ca>
    Subject: All Colloquium transcripts available

    All Colloquium transcripts are now available via

    http://www.bootstrap.org/colloquium

    Henry

    ________________________________________________________________________
    ________________________________________________________________________

    Message: 13
       Date: Fri, 28 Apr 2000 11:56:09 -0000
       From: "Henry van Eyken" <vaneyken@sympatico.ca>
    Subject: A small experiment to help students

    The "no-margins" version of the transcript for the Colloquium's
    Session 1 contains an experimental feature that permits a student to
    quickly step through all the graphics or, independently, through all
    the tables without losing his place in the document. It is found at:

    http://www.bootstrap.org/colloquium/session_01/session_01-.html

    Find underneath the heading the words "quick-step." A click on the
    asterisk leads to a footnote that explains what quick-step does. I
    perceive it as an attempt towards easing the studying of large
    documents. College texts, for example.

    I wonder whther there are other, better ways for achieving that
    objevtive.

    Henry

    ________________________________________________________________________
    ________________________________________________________________________

    ------------------------------------------------------------------------
    Was the salesman clueless? Productopia has the answers.
    http://click.egroups.com/1/3019/3/_/444287/_/957132508/
    ------------------------------------------------------------------------

    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



    This archive was generated by hypermail 2b29 : Sun Apr 30 2000 - 15:16:16 PDT