THE WELCH COMPANY
440 Davis Court #1602
San Francisco, CA 94111-2496
415 781 5700
rodwelch@pacbell.net


S U M M A R Y


DIARY: November 30, 2000 07:36 PM Thursday; Rod Welch

Jack Park submits ideas for improving SDS with XML for KM.

1...Summary/Objective
2...Ontology "Engine" May Enhance SDS Combines Elements for KM
3...SDS Interface Turns KM Structure into Effective Tool for Users
4...Jack says SDS needs...
....1...Navigation engine
....2...Ontology for itself by which users can do the navigating.
...........Collaboratory Results from Linking SDS with Ontology
....3...IBIS mechanism by which collaborative SDS implementations can
5...3-layer Architecture Adds Features to SDS that Lead to OHS/DKR
.....upper level,
.......•..topic maps
.......•..IBIS
.......•..import/edit facilities for generation of major "papers" to
.....middle
..........ontology engine builds nodes and links, a vast array...
..........engine maintains "knowledge structures" (KS).
.....bottom
.......•..Raw information, which includes: papers, scanned in books,
.......•..IBIS argumentation records,
..........Different Doc Types Can be Opened
..........XML Grove Architecture Makes Everything Addressable
.......•..SDS duplicates raw data rather than link to original sources.

ACTION ITEMS.................. Click here to comment!

1...Why can't people interact with the record now, demonstrated
2...Need explanation of semantic interoperoperability.
3...How does Jack solve Pandora's Box of complexity, which he
4..."IBIS mechanism" is vague.
5...Need constructive explanation of "knowledge," on all three
6...How do these three elements relate to the new proposal today?
7...Why aren't proposals, plans, budgets,... etc., simply more
8...Is "import" here similar to the scope reported on 001121,
9...Has this been done by anyone, anywhere? What record of
10...Does the discussion about "lens" technology relate to
11...How is linking accomplished? How does the "engine," or
12...Does this improve handling of daily working information,
13...Getting confused here about the layers. Need diagram,
14...This is a very good point about redundant storage.

CONTACTS 

SUBJECTS
Architecture Makes Sense Contemporaneous Engineering Record with Code
SDS Design Represents Underlying Structure of KM Combining Elements T
User Interface SDS Turns Structure Needed for KM into Useful Tool for
SDS Proof of Concept DKR Knowledge Space, Jack Park, 000426
SDS Interface Turns Structure Knowledge Management into Useful Tool
Text Structure GUI Wordprocessing Knowledge Management Dilemma Balan

8708 -
8708 -    ..
8709 - Summary/Objective
8710 -
871001 - Follow up ref SDS 63 0000, ref SDS 61 0000.
871002 -
871003 - Jack submits plans to enhance SDS with XML, ref SDS 0 MU7H, that would
871004 - automate subjects and links, ref SDS 0 MN3W, using a 3-layer
871005 - architecture. ref SDS 0 WG5G  He seems to use "ontology" broadly for
871006 - both subjects, and links, which are further related to Topic Maps and
871007 - Knowledge Structures. ref SDS 0 GK8O Jack wants structured analysis
871008 - exemplified by IBIS, ref SDS 0 AZ7J He, also, wants to "interact" with
871009 - SDS records on the web, ref SDS 0 WG5G, using addressability.  These
871010 - objectives are already supported by the SDS program. ref SDS 0 OW3F
871011 - Recently, ideas were proposed for making SDS available to the team so
871012 - that others can realize these benefits.
871014 -  ..
871015 - Submitted ref DIT 1 0001 thanking Jack for ideas to improve SDS, as
871016 - set out in the letter, ref DIP 1 02WQ, on 001102, ref SDS 55 XR8Y, and
871017 - more recently in another letter, ref DIP 4 009O, on 001116.
871018 - ref SDS 58 009O  Recommend focusing on doing KM to gain experience
871019 - needed for building and using KM tools, ref DIT 1 XC7K, per the
871020 - letter, ref DIP 6 6W6L, earlier today explaining need for culture of
871021 - knowledge, ref SDS 64 6C8H, based on telecon with Doug on 001128.
871022 - ref SDS 63 1O6J
871023 -
871024 -     [On 001202 letter notifies DKR team about the course offered by
871025 -     KMCI to develop a theory of knowledge, and suggest participating
871026 -     in dialog to foster a culture of knowledge. ref SDS 65 ME7O
871028 -      ..
871029 -     [On 001214 Jack submits use case analysis to develop ontology
871030 -     based on Qualitative Process Theory. ref SDS 66 0001
871032 -      ..
871033 -     [On 001231 engineer plans to automate subject management, wants to
871034 -     study SDS for ideas. ref SDS 69 NW8R
871036 -      ..
871037 -     [On 010114 example of time and cost savings using "intelligence"
871038 -     from SDS records in Knowledge Space. ref SDS 70 0001
871040 -      ..
871041 -     [On 010316 Jack reports progress developing OHS code. ref SDS 76
871042 -     Q57N
871043 -
871044 -
871045 -
871047 -  ..
8711 -
8712 -
8713 - Progress
8714 -
871401 - Ontology "Engine" May Enhance SDS Combines Elements for KM
871402 - SDS Interface Turns KM Structure into Effective Tool for Users
871403 -
871404 - Received ref DRT 1 0001 from Jack Park responding to the letter about
871405 - summarizing project documents, ref DIP 5 0001, which was submitted on
871406 - 001128, ref SDS 63 0001, and addressed Jack's ideas in a letter
871407 - received on 001123. ref SDS 61 3F9O
871409 -  ..
871410 - Today, Jack provides some analysis of SDS which makes progress on the
871411 - action item he set set on 000403. ref SDS 25 E53F
871413 -  ..
871414 - Jack explains SDS uses "time" and "subjects" to manage daily work,
871415 - ref DRT 1 645I; he feels tracking time and context helps "make sense
871416 - of the record." ref DRT 1 ZW5I  Making sense of complex events occurs
871417 - through diary narrative that comprehensively connects cause and effect
871418 - in related "history" -- also called "experience," "intelligence,"
871419 - "personal and organizational memory," and more simply "knowledge" used
871420 - to guide current work, and plan for the future, explained in POIMS,
871421 - ref OF 3 SE4F, sometimes called "good management." ref OF 3 1X6G
871423 -  ..
871424 - Citing time and context (subjects) today, aligns with Jack's call on
871425 - 000331, when this issue was discussed and he planned to review POIMS.
871426 - ref SDS 24 TQ68  Making sense of the record by adding connections
871427 - across time and distance, also, noted by Jack today fits the 3-layer
871428 - architecture for complementarity presented in POIMS. ref OF 3 8555
871429 - Vannevar Bush anticiapted this design in his 1945 article on
871430 - requirements for efficiencies of electronic records management,
871431 - reviewed on 960304. ref SDS 7 GX59
871432 -
871433 -     [On 040622 Jack proposes a primary purpose of Knowledge Management
871434 -     is "sensemaking" from complexity theory, ref SDS 92 FM5W; which
871435 -     was analysed at that time. ref SDS 92 T66N
871437 -  ..
871438 - Jack reports today that the SDS design represents the structure
871439 - necessary for Knowledge Management. ref DRT 1 645I  This aligns with
871440 - his letter on 000426 saying SDS work product demonstrates proof of
871441 - concept for Knowledge Management. ref SDS 28 3315  Jack further
871442 - reports today the SDS interface turns the structure for representing
871443 - knowledge into a useful tool for users, ref DRT 1 AV6I, as called out
871444 - in POIMS. ref OF 3 6M5H
871446 -  ..
871447 - Making the structure of knowledge useful requires developing a user
871448 - interface that gives people command and control of the record,
871449 - reported on 990530. ref SDS 13 EE6R  Earlier this year, on 000824 the
871450 - list of improvements continued to grow on integrating commands to
871451 - provide multi-tasking based on continually evolving situations and
871452 - scenarios that leverages the power of knowledge. ref SDS 43 0001
871453 -
871454 -     [On 020531 Mike Poremba creating Java outline program notes SDS
871455 -     optimized for efficient usability of program functions,
871456 -     ref SDS 90 5M3G; dates make sense of complex information so links
871457 -     are contextually comprehensible, ref SDS 90 OP7M, SDS support for
871458 -     timely analysis to aid human thinking is hard to achieve; DOS
871459 -     enables working at the speed of human thought. ref SDS 90 5W9J
871461 -      ..
871462 -     [On 040720 ASB file notes SDS user interface makes working
871463 -     intelligently fast, easy and fun; user interface makes command and
871464 -     control of daily work like conducting a symphony. ref SDS 93 U46J
871466 -  ..
871467 - Favorable comments today align with the record on 000504 where Jack
871468 - recommended that people ponder POIMS to learn how SDS supports
871469 - knowledge management, ref SDS 30 6082, to solve problems like John
871470 - Maloney, who leads a professional group for knowledge work, and
871471 - reported difficulty finding critical information with popular tools.
871472 - ref SDS 15 XO3G
871474 -  ..
871475 - More recently, on 001030 Jack noticed that talented engineers
871476 - struggling to understand KM in order to build an OHS/DKR have
871477 - overlooked that someone has been doing KM for "umpteen years,"
871478 - ref SDS 54 0S8L, and two (2) weeks ago on 001116 Jack reported SDS is
871479 - amazing. ref SDS 58 SN4N  However, a few days ago, the team decided to
871480 - use IT methods to develop KM, rather than SDS, as reported on 001126.
871481 - ref SDS 62 T63N
871482 -
871483 -        [On 010411 Jack commends SDS subject index support for "finding
871484 -        every little thing."
871486 -         ..
871487 -        [On 010425 Morris Jones reports using SDS is a utopia compared
871488 -        using other methods because everything is in the right place at
871489 -        the right time. ref SDS 80 EP7F
871491 -         ..
871492 -        [On 010916 Eric Armstrong says SDS enables amazing memory
871493 -        because everything is in the right place at the right time.
871494 -        ref SDS 87 0001
871496 -         ..
871497 -        [On 011003 Eric Armstrong reports the user interface everybody
871498 -        likes requires hard work to perform KM, so nobody is doing it,
871499 -        causing an innovation loop that prevents discovery of how to
871500 -        design KM. ref SDS 89 O73F
871502 -         ..
871503 -        [On 011003 Jack Park reports there no clues about how to use
871504 -        other technologies for improving information technology that
871505 -        Eric says paralyzes productivity. ref SDS 89 O76L
871507 -         ..
871508 -        [On 011003 long history shows SDS improves productivity by
871509 -        enabling knowledge management. ref SDS 89 R18J
871511 -         ..
871512 -        [On 040622 Jack Park describes Knowledge Management using tools
871513 -        for analysis that makes sense of complex information,
871514 -        summarized by the notion of "sensemaking. ref SDS 92 V47N
871516 -         ..
871517 -        [On 040622 Cliff Joslyn at LANL discussed SDS support for
871518 -        mathemaical knowledge structures. ref SDS 92 SY9I
871520 -         ..
871521 -        [On 041213 Edison popularized cometemporaneous notes that make
871522 -        scientific method effective for research and development.
871523 -        ref SDS 94 DK4N
871525 -         ..
871526 -        [On 050219 SDS improvement makes Context Management more
871527 -        intuitive, complementary, and fun like playing a piano.
871528 -        ref SDS 95 MF5N
871530 -         ..
871531 -        [On 070729 report Jack's letter on 070501 describes SDS subject
871532 -        index as constructing and maintaining a graph structure, seems
871533 -        to indicate Topic Maps may not add value. ref SDS 97 HF56
871535 -         ..
871536 -        [On 070729 Jack recognizes SDS records support relational and
871537 -        hierarchial data structures. ref SDS 97 HF56
871539 -       ..
871540 -      On 910810 Morris reported SDS has a breakthrough design that
871541 -      integrates hierarchial and relational structure. ref SDS 1 7793
871543 -       ..
871544 -      On 940603 Gio Wiederhold's article reports KM efforts at Stanford
871545 -      to create a "mediator" methodology that sounds similar to the
871546 -      "engine" Jack proposed on 000623 for processing information.
871547 -      ref SDS 37 2915  Widerhold maintains that user interface should
871548 -      be machine-and-communication-friendly, rather than "user
871549 -      friendly," ref SDS 5 OV7N, in order to improve productivity and
871550 -      earnings.
871552 -       ..
871553 -      On 950710 user interface is explained in Tom Landauer's book "The
871554 -      Trouble with Computers." ref SDS 6 3P6G
871556 -       ..
871557 -      On 000425 SDS seems to have a unique design for KM. ref SDS 27
871558 -      0480
871560 -       ..
871561 -      On 000921 Eric Armstrong explained that respect for customers
871562 -      requires interface that balances ease of use with usefulness.
871563 -      ref SDS 45 0R9O
871565 -       ..
871566 -      On 000925 Cliff Joslyn related that SDS has effective design for
871567 -      a wide variety of KM tasks. ref SDS 47 XW5O
871569 -       ..
871570 -      Jack's report today aligns with recommendations to the team by
871571 -      Dick Karpinski on 000227, ref SDS 19 0897, and more recently by
871572 -      Henry van Eykan on 001126. ref SDS 62 0001  On 911121 people have
871573 -      difficulty at first learning a new way of working using
871574 -      technology for knowledge rather than for information and games.
871575 -      ref SDS 2 5699
871577 -         ..
871578 -        [...in another record today, Paul Fernhout submits an article
871579 -        reporting that IBM has given up trying to convert Lotus Notes
871580 -        into a Knowledge Management program, they were hoping to
871581 -        release under the name of "Raven." ref SDS 64 F26K
871583 -         ..
871584 -        [On 010411 Jack reports SDS enables remembering everything
871585 -        little thing, which is effective organizational memory.
871586 -        ref SDS 79 X34F  This contrasts sharply with Eric Armstrong's
871587 -        report on 010916 that nobody can find anything using other
871588 -        methods. ref SDS 88 KA6H
871590 -         ..
871591 -        [On 010916 Eric Armstrong reports SDS enables organizational
871592 -        memory; wants a better user Interface. ref SDS 87 0001  On
871593 -        011003 Eric reports that productivity is paralyzed by interface
871594 -        people like. ref SDS 89 EC5N
871596 -         ..
871597 -        [On 011003 two years of research have not yielded a clue about
871598 -        how to create KM for improving productivity. ref SDS 89 O74L
871600 -  ..
871601 - Jack proposes adding an Ontology as a central core of an OHS that ties
871602 - everything together.  He asks whether the team should develop a
871603 - summary of documents as originally suggested?  see 001123. ref SDS 61
871604 - 3F9O  Or, should we focus on developing navigational tools necessary
871605 - to get a better handle on what we have been saying?  Jack feels both
871606 - should be done. ref DRT 1 0246
871607 -
871608 -      Need typical day scenario showing how "ontology" is created: who
871609 -      does it, using what tools, what structure, for how much time, how
871610 -      is it maintained, and how it is applied minute by minute, hour by
871611 -      hour, day by day, etc. see for example SDS. ref OF 4 0001
871613 -       ..
871614 -      Scenarios can illustrate use of "navigation engine" discussed
871615 -      below. ref SDS 0 WG5G
871616 -
871617 -
871618 -        [On 010201 Jack submit ideas for SDS to support OHS core system
871619 -        model. ref SDS 72 0001
871621 -         ..
871622 -        [On 010211 Jack's 3-layer architecture for OHS incorporated
871623 -        into planning for facilities management business. ref SDS 73
871624 -        1V6L
871626 -  ..
871627 - On 000427 Jack explained a 3-tier architecture based on WBI
871628 - transcoding. ref SDS 29 4674  On 000623 he proposed 3-layer ontology
871629 - architecture, ref SDS 37 EK3W, based on the presentation he made with
871630 - Marry Keeler at SRI on 000518. ref SDS 31 C53G
871631 -
871632 -     SDS strives to integrate what we say with what we write, and also
871633 -     with what we do and think.  This is a more robust perspective of
871634 -     KM that is necessary for effective enterprise management.
871636 -         ..
871637 -        [On 001214 Jack submits use case analysis to develop ontology
871638 -        based on Qualitative Process Theory. ref SDS 66 0001
871640 -         ..
871641 -        [On 010114 example of time and cost savings using
871642 -        "intelligence" from SDS records in Knowledge Space. ref SDS 70
871643 -        0001
871645 -  ..
871646 - Jack says initially in relation to the letter on 001128, ref DIP 5
871647 - 0001, that "Your question hits the target I have been painting for
871648 - some time now." ref DRT 1 SS6J
871650 -      ..
871651 -     Jack, however, does not identify any questions from sources,
871652 -     nor provide links to show context and orientation on the answers
871653 -     he submits.
871655 -      ..
871656 -     This deficit underscores another dimension of KM that needs help
871657 -     along with better tools.  Eugene Kim submitted a 6-point program
871658 -     on 001126 for KM, but does not include the need to specify who,
871659 -     what, when, where, and why that comprises requirements of good
871660 -     reporting and engineering.
871662 -      ..
871663 -     In fact, the letter on 001128 does not have any express questions,
871664 -     ref DIP 5 0001, so Jack is evidently referring to the SDS record
871665 -     on 001123, which has several.
871667 -      ..
871668 -     One question was...
871670 -         ..
871671 -        What are the various aspects of summarization that Jack and the
871672 -        team feel should be performed? ref SDS 61 3F9O
871673 -
871674 -            Jack does not mention this, possibly because in an earlier
871675 -            letter he indicated that the summary in the SDS record is
871676 -            adequate.  Today, he concludes by proposing the team work
871677 -            on both a document summary project, and to create an
871678 -            ontology "engine" for SDS. ref DRT 1 0246
871679 -
871681 -      ..
871682 -     Another question was...
871683 -
871684 -        Is document summary part of managing communications, e.g.,
871685 -        meetings and correspondence, which Doug requested on 000427,
871686 -        as a Guide to Running NICs, ref SDS 29 2867, based on his 1972
871687 -        paper which he cited on 000327? ref SDS 22 3971, as shown on
871688 -        001123. ref SDS 61 J55I
871690 -             ..
871691 -            Jack does not mention the Guide to Running NICs.
871693 -             ..
871694 -            My sense is that going forward on a Guide to Running NICs,
871695 -            is an important point because it provides continuity
871696 -            between Doug's historical experience in formulating
871697 -            objectives, and current effort to produce a new tool.
871699 -             ..
871700 -            Why not just say that summarizing the record is a key
871701 -            aspect of developing useful intelligence set out in POIMS,
871702 -            ref OF 2 0367, and that documents are an important part of
871703 -            the record that need to be summarized along with everything
871704 -            else.
871705 -
871707 -      ..
871708 -     Another question was...
871710 -         ..
871711 -        Do the summaries, also, need organization?  Do we need
871712 -        subjects, or topics (categories) for cross-referencing
871713 -        documents? ref SDS 61 465N
871715 -  ..
871716 - Jack addresses this question indirectly.
871717 -
871718 -
871719 -
871720 -
8718 -

SUBJECTS
Ontology Definition
Ontology Evolves from Use Cases under Qualitative Process Theory, Jac
Ontology SDS Improvement
Ontology formed by Use Cases under Qualitative Process Theory, 001214
Topic Maps, Roy Roebuck, 000125
Engine Create Topic Map Organizes Information for Retrieval
Topic Maps Graphical Maps Organize Record Subjects

A509 -
A51001 -  ..
A51002 - Jack says SDS needs...
A51003 -
A51004 -    1.  Navigation engine
A51005 -
A51006 -           Jack wants a navigation tool so people can interact with the
A51007 -           SDS record on the web, and feels this would comprise a more
A51008 -           complete KM tool. ref DRT 1 0232
A51010 -            ..
A51011 -           Below, Jack describes a 3-layer architecture enables users
A51012 -           to manipulate the archive (solely) by adding new commentary,
A51013 -           etc. ref SDS 0 NX66
A51015 -            ..
A51016 -           Why can't people interact with the record now, demonstrated
A51017 -           by this record, and typical day scenario for SDS, cited
A51018 -           above on "ontology." ref SDS 0 0246  For example, today Jack
A51019 -           wrote an important letter, and this record "interacts" by
A51020 -           commenting based on context of related history and
A51021 -           objectives.  On 001121 the team was notified they could get
A51022 -           SDS to similarly facilitate communication, if this method
A51023 -           seems useful. ref SDS 59 XU8I
A51025 -            ..
A51026 -           Jack needs to prepare a few scenarios showing a navigation
A51027 -           engine would add to the current SDS experience, per above.
A51028 -           ref SDS 0 0246
A51029 -
A51030 -              [On 010114 example of time and cost savings using
A51031 -              "intelligence" from SDS records in Knowledge Space where
A51032 -              someone "interacted" in some degree. ref SDS 70 0001
A51034 -               ..
A51035 -              [On 010620 SRI proposes engine to Automatically Manage
A51036 -              Distributed Multi-user Change Tracking and version
A51037 -              control, performing an RDMS function. ref SDS 84 0157
A51039 -            ..
A51040 -           How does this navigation engine relate to the engine Jack
A51041 -           proposed on 000623. ref SDS 37 2915  On 990424 Morris, also,
A51042 -           wanted an "engine" of some kind.
A51044 -            ..
A51045 -           Is this navigation engine the Topic Map, or does it create
A51046 -           the Topic Map proposed for Upper Level below. ref SDS 0 HM7J
A51047 -           How does it relate to the "engine" in the middle layer?
A51048 -           ref SDS 0 CC7K  Does this engine create the "ontology"
A51049 -           discussed in the next section?
A51050 -
A51052 -         ..
A51053 -    2.  Ontology for itself by which users can do the navigating.
A51054 -        ref DRT 1 ED7K  (on 000221 Jack provided source for definition
A51055 -        of ontology. ref SDS 18 6W6N)
A51056 -
A51057 -           Jack says there are humans doing all this, but the engines
A51058 -           can, and should automate some of the "backoffice"
A51059 -           bookkeeping.  By automating that, we get uniformity in the
A51060 -           ontology generation. Thus, we get (or, at least, strive for)
A51061 -           semantic interoperability. ref DRT 1 5G9I
A51062 -
A51063 -              Need explanation of semantic interoperoperability.
A51064 -
A51065 -              "Backoffice bookkeeping" needs an explanation, per below.
A51066 -              ref SDS 0 HM5J
A51067 -
A51068 -                 [On 001214 Jack mentions interoperoperability.
A51069 -                 ref SDS 66 VQ3N
A51071 -                  ..
A51072 -                 [On 001230 engineer plans to automate subject
A51073 -                 management, wants to study SDS for ideas. ref SDS 69
A51074 -                 NW8R
A51076 -                  ..
A51077 -                 [On 010223 Jack makes similar proposal. ref SDS 75
A51078 -                 I56H
A51080 -            ..
A51081 -           On 001105 Henry observed that SDS users can navigate quickly
A51082 -           and easily. ref SDS 56 WG6F  He confirmed this in a letter
A51083 -           on 001126. ref SDS 62 1V6O
A51085 -            ..
A51086 -           On 000405 Jack explained ontology in relation to signs and
A51087 -           symbols, ref SDS 26 1548, which was cited later on 000713
A51088 -           when Cliff Joslyn reviewed Semiotics. ref SDS 39 4078
A51089 -
A51090 -              [On 001214 Jack correlates ontology to use cases and
A51091 -              Qualitative Process Theory (QPT). ref SDS 66 0001
A51093 -            ..
A51094 -           How does Jack solve Pandora's Box of complexity, which he
A51095 -           cited on 000221? ref SDS 18 7455  On 000405 Paul Fernhout
A51096 -           pointed out that AI theory to formulate useful knowledge
A51097 -           representation has not been effective. ref SDS 26 B8R4
A51099 -            ..
A51100 -           Earlier, on 970116 complexity of subject management was
A51101 -           discussed at Intel. ref SDS 10 3109
A51102 -
A51103 -              [On 001220 Jack demonstrated initial work developing and
A51104 -              applying an ontology. ref SDS 68 BV4E
A51106 -            ..
A51107 -           On 001011 research was proposed to study this problem.
A51108 -           ref SDS 50 M4R8
A51110 -            ..
A51111 -           Ontology in Jack's design scheme today seems to cover a
A51112 -           number of things, beyond the definition set out on 000221.
A51113 -           ref SDS 18 2622 and on 000606. ref SDS 32 J13M  Does this
A51114 -           require 4 separate "engines" or a single "integrated" engine
A51115 -           for the following....
A51116 -
A51117 -             •  navigation engine. ref SDS 0 WG5G
A51119 -                 ..
A51120 -             •  topic maps. ref SDS 0 HM7J
A51122 -                 ..
A51123 -             •  knowledge structures, ref SDS 0 CC7K, and
A51125 -                 ..
A51126 -             •  linking. ref SDS 0 MN3W
A51128 -            ..
A51129 -           On 001025 Jack's ideas for automatic construction of
A51130 -           knowledge representation was reviewed by Paul Fernhout.
A51131 -           ref SDS 52 LR6N
A51132 -
A51133 -               [On 001214 Jack proposes "firing" process rules that
A51134 -               sound similar to the engine discussed today. ref SDS 66
A51135 -               NO6I
A51137 -                ..
A51138 -               [On 010223 Jack proposed engine for SDS. ref SDS 75 3RKS
A51139 -
A51140 -
A51141 -
A51142 -
A512 -

SUBJECTS
Collaboration Goal Enterprise/Knowledge Management Does Not Improve P
Collaboratory - Link SDS with Ontology
Collaboration Customer Requirement NIH, Others, Pat Lincoln

AM05 -
AM0601 -            ..
AM0602 -           Collaboratory Results from Linking SDS with Ontology
AM0603 -
AM0604 -           Jack expects ontology he has in mind will allow SDS to link
AM0605 -           up with other SDS engines for a web-based collaboratory.
AM0606 -           ref DRT 1 FF8O  This reflects KM industry focus on improving
AM0607 -           collaboration reported on 991217.  On 000608 Sandy Klausner
AM0608 -           plans a Collaborium for e-business. ref SDS 34 8446  On
AM0609 -           000623 Jack proposed a collaborative evolutionary
AM0610 -           epistomology. ref SDS 37 4752
AM0611 -
AM0612 -              [On 010114 example of time and cost savings using
AM0613 -              "intelligence" from SDS records in Knowledge Space where
AM0614 -              someone "interacted" in some degree. ref SDS 70 0001
AM0616 -               ..
AM0617 -              [On 010412 Jack wants a collaboratory. ref SDS 79 NV5I
AM0619 -               ..
AM0620 -              [On 010614 inquiry about collaborating on a
AM0621 -              collaborarium. ref SDS 83 LH3O
AM0622 -
AM0623 -
AM0624 -
AM0625 -
AM0626 -
AM0627 -
AM07 -

SUBJECTS
IBIS Processing SDS Record Supports Lexus Research
IBIS Support to Compliment SDS in OHS
IBIS Catagories for Question, Answer, etc Tracked by OHS, Eugene 0006
IBIS Decision Support
IBIS Rigid Ineffective, 000218

AV07 -
AV0801 -  ..
AV0802 -    3.  IBIS mechanism by which collaborative SDS implementations can
AV0803 -        communicate with each other. ref DRT 1 5G9I
AV0804 -
AV0805 -           "IBIS mechanism" is vague.
AV0806 -
AV0807 -               [On 010114 "deliberately managed conversation" proposed
AV0808 -               as foundation of KM. ref SDS 71 1550
AV0810 -                ..
AV0811 -               [On 010223 Jack proposes IBIS for OHS, ref SDS 75 0S9F,
AV0812 -               and for improving education by aiding critical thinking
AV0813 -               in combination with topic map technology. ref SDS 75
AV0814 -               SL3O
AV0816 -                ..
AV0817 -               [On 010329 Jack explains SDS he proposes for OHS is for
AV0818 -               "Structured Dialog System," which sounds like an IBIS
AV0819 -               objective. ref SDS 77 XZ4K
AV0821 -                ..
AV0822 -               [On 010426 Pat Lincoln lists IBIS for providing
AV0823 -               structure to support KM. ref SDS 82 QY7N
AV0825 -            ..
AV0826 -           On 000614 Eric incorporates IBIS in requirements for CDS,
AV0827 -           ref SDS 35 4939, which the next day was converted to OHS, so
AV0828 -           Jack's suggestion aligns with part of the project record.
AV0829 -           ref SDS 36 2808  IBIS is not mentioned in the OHS Launch Pan
AV0830 -           received on 001025. ref SDS 52 G3W8
AV0832 -            ..
AV0833 -           On 000218 concerns were raised about value added of IBIS.
AV0834 -           ref SDS 17 7050
AV0835 -
AV0836 -              Need use case example showing how IBIS can help design
AV0837 -              the DKR, go to the dentist, fix the car, analyse a
AV0838 -              letter, as in this record.
AV0840 -               ..
AV0841 -              Provide an example using IBIS this past week, or this
AV0842 -              past year?
AV0844 -            ..
AV0845 -           SDS balances structure and creativity by enabling people to
AV0846 -           use rigid IBIS forms when desirable, and conventional forms
AV0847 -           of analysis when convenient or desirable.
AV0848 -
AV0849 -              [On 001220 Gary Johnson submits reference to paper by
AV0850 -              Frank Shipman who cautions against requiring rigid
AV0851 -              structure based on evidence indicating it reduces
AV0852 -              productivity. ref SDS 67 RD9N
AV0854 -  ..
AV0855 - These three enhancements can be accomplished by incorporating Jack's
AV0856 - 3-layer architecture into SDS, discussed below....
AV0857 -
AV0858 -
AV0859 -
AV0860 -
AV0861 -
AV09 -

SUBJECTS
DKR Architecture 3-layers Supported by SDS
SDS Standard SDS Time Organize Information Produce Knowledge
Topic Maps Graphical Maps Organize Record Subjects
IBIS Categories for Question, Answer, etc Tracked by OHS, Eugene 0006
Import/Export Legacy Document Support

BJ07 -
BJ0801 -  ..
BJ0802 - 3-layer Architecture Adds Features to SDS that Lead to OHS/DKR
BJ0803 -
BJ0804 - Jack explains 3-layer architecture to accomplish objectives of CODIAK
BJ0805 - set out by Doug Engelbart at the following location...
BJ0806 -
BJ0807 -              http://www.bootstrap.org/augment-132811.htm
BJ0809 -  ..
BJ0810 - Jack says that adding navigation, ontology and IBIS, per above,
BJ0811 - ref SDS 0 WG5G, would give SDS the 3-layer architecture for OHS/DKR he
BJ0812 - proposes. ref DRT 1 PR4F, that enables the collaborative construction
BJ0813 - of knowledge, ref DRT 1 0153, based on following assumptions...
BJ0814 -
BJ0815 -       [On 020730 Gary Johnson proposes intermediate work product
BJ0816 -       format to use SDS Knowledge Repository that is less onerous and
BJ0817 -       so easier for people to transform work practice from information
BJ0818 -       to a culture of knowledge. ref SDS 91 YR6F
BJ0819 -
BJ0820 -     a.  knowledge, itself, is a human-centric notion and cannot be
BJ0821 -         "contained" anywhere else but in meat-space.
BJ0823 -          ..
BJ0824 -     b.  knowledge can be *represented* in the form of statements,
BJ0825 -         icons, links, which evoke responses when meat-space is exposed
BJ0826 -         to those representations. ref DRT 1 6X8I
BJ0828 -             ..
BJ0829 -            Jack proposes an "engine" to create knowledge structures,
BJ0830 -            per analysis below. ref SDS 0 CC7K
BJ0832 -          ..
BJ0833 -     c.  knowledge representations are never the *territory*, they are
BJ0834 -         only the *maps* to the territory. ref DRT 1 0152
BJ0835 -
BJ0836 -            Need constructive explanation of "knowledge," on all three
BJ0837 -            (3) points, based on the architecture of human thought, see
BJ0838 -            POIMS. ref OF 2 0561
BJ0840 -          ..
BJ0841 -     d.  constructing a consensus ontology is:
BJ0842 -
BJ0843 -         a.  very hard (some would say, impossible), ref DRT 1 K39F
BJ0844 -
BJ0845 -               See analysis below. ref SDS 0 MN3W and ref SDS 0 CC7K
BJ0846 -
BJ0847 -               On 000405 Jack explained ontology in relation to signs
BJ0848 -               and symbols, ref SDS 26 1548, which was cited later on
BJ0849 -               000713 when Cliff Joslyn reviewed Semiotics. ref SDS 39
BJ0850 -               4078
BJ0851 -
BJ0852 -                  [On 001214 Jack correlates ontology to use cases and
BJ0853 -                  Qualitative Process Theory (QPT). ref SDS 66 0001
BJ0855 -              ..
BJ0856 -         b.  worth doing if semantic interoperability is desired
BJ0858 -          ..
BJ0859 -     e.  consensus ontologies are most useful in specific (often called
BJ0860 -         *vertical*) domains, ref DRT 1 0161
BJ0862 -          ..
BJ0863 -     f.  a common *upper* ontology to which all vertical domains are
BJ0864 -         anchored allows for eventual discovery of links between
BJ0865 -         domains (analogies, etc). ref DRT 1 P44G
BJ0867 -  ..
BJ0868 - Jack says in a "note" that a favorite illustration of analogical links
BJ0869 - between domains, as often used in the AI community, is that of the
BJ0870 - mapping of the *divide and conquer* military notion into the radio
BJ0871 - therapy domain, ref DRT 1 J64K, and that hanging domains on a common
BJ0872 - *upper* ontology enables a background discovery process to run on a
BJ0873 - continuous basis. ref DRT 1 0181
BJ0874 -
BJ0875 -     Background processing helps accomplish high volume of transactions
BJ0876 -     for creating subjects and links, which Jack seems to include for
BJ0877 -     the scope of ontology. ref SDS 0 S29N
BJ0879 -  ..
BJ0880 - On 000710 received diagram of 3-layer architecture, ref SDS 38 0001,
BJ0881 - with the title "Categories Diagram," which illustrates the 3-layered
BJ0882 - architecture Jack presented on 000623, ref SDS 37 EK3W, based on Mary
BJ0883 - Keeler's remarks during a meeting at SRI on 000518, which used a
BJ0884 - 3-layer diagram. ref SDS 31 8439
BJ0886 -  ..
BJ0887 - Jack's diagram showed three elements...
BJ0888 -
BJ0889 -     g. Topic Maps............................. Probability
BJ0890 -     h. Knowledge Structures................... Actuality
BJ0891 -     i. Documents.............................. Possibility
BJ0892 -
BJ0893 - ...submitted with his letter on 000710. ref SDS 38 V3YP
BJ0894 -
BJ0895 -     How do these three elements relate to the new proposal today?
BJ0896 -
BJ0898 -  ..
BJ0899 - Today, Jack says that Doug, ref DRT 1 0194, speaks of....
BJ0900 -
BJ0901 -     g.  Intelligence Collection (IC), ref OF 1 0175
BJ0902 -
BJ0903 -     h.  Dialog Records (DR), ref OF 1 0176
BJ0904 -
BJ0905 -     i.  Knowledge Product (KP), ref OF 1 0179
BJ0906 -
BJ0908 -  ..
BJ0909 - Jack provides a general reference to Doug's 1992 paper that explains
BJ0910 - the CODIAK system...
BJ0911 -
BJ0912 -              http://www.bootstrap.org/augment-132811.htm
BJ0914 -  ..
BJ0915 - On 001025 Doug called for references to specific information that
BJ0916 - expedites review and aids alignment. ref SDS 52 00VU
BJ0918 -  ..
BJ0919 - Review on 991222 shows SDS implements CODIAK objectives for knowledge
BJ0920 - and intelligence capability, ref SDS 14 8064  More recently on 001121
BJ0921 - the DKR team was notified that KM can be accomplished using SDS.
BJ0922 - ref SDS 59 XU8I, per telecon with Doug on 001027. ref SDS 53 XV6F
BJ0924 -  ..
BJ0925 - Jack explains the three-layer architecture provides an archive and
BJ0926 - interface for information flowing in from some environment (IC). That
BJ0927 - environment includes users manipulating the archive (solely) by adding
BJ0928 - new commentary and participating in IBIS-based dialogs (DR).
BJ0929 - ref DRT 1 0194
BJ0930 -
BJ0931 -       Above, Jack proposes users interacting with the record on the
BJ0932 -       Internet. ref SDS 0 WG5G
BJ0934 -        ..
BJ0935 -       This explanation of Intelligence Collections aligns with the
BJ0936 -       record on 991222, ref SDS 14 8064, citing Doug's 1992 paper,
BJ0937 -       which seems to include the scope for "Dialog Records (DR),
BJ0938 -       ref OF 1 0175, and all other working information, discussed
BJ0939 -       below in connection with the "bottom" layer. ref SDS 0 0J8G
BJ0941 -        ..
BJ0942 -       IBIS-based dialog is reviewed above. ref SDS 0 AZ7J
BJ0943 -
BJ0944 -          [On 010411 Jack proposes re-implementation of SDS to support
BJ0945 -          a Collaborarium that enables online editing of SDS by
BJ0946 -          multiple parties. ref SDS 79 NV5I
BJ0947 -
BJ0949 -  ..
BJ0950 - Jack explains "knowledge product" (KP) covers proposals, plans,
BJ0951 - budgets, contracts, and so forth as constructed by the participants
BJ0952 - and users of a particular OHS.  KM is a process that enables KP.
BJ0953 - ref DRT 1 HV6N
BJ0954 -
BJ0955 -       Why aren't proposals, plans, budgets,... etc., simply more
BJ0956 -       information, part of "intelligence" gathering?  If someone draws
BJ0957 -       up a proposal, reviews a plan, posts a budget, isn't this part
BJ0958 -       of daily working information. ref SDS 0 0J8G  What is the
BJ0959 -       difference?
BJ0961 -  ..
BJ0962 - Jack recommends following additional scope for SDS....
BJ0963 -
BJ0964 -     upper level,
BJ0965 -
BJ0966 -       •  topic maps
BJ0967 -
BJ0968 -           On 000125 Topic Maps were proposed, ref SDS 16 9246, but the
BJ0969 -           application to KM is unclear relative to requirements for
BJ0970 -           improving handling of daily working information across an
BJ0971 -           enterprise. ref SDS 16 Y68L
BJ0973 -            ..
BJ0974 -           On 000403 Jack seems to correlate Topic Map with ontology.
BJ0975 -           ref SDS 25 1144
BJ0976 -
BJ0977 -              [On 010114 example of time and cost savings using
BJ0978 -              "intelligence" from SDS records in Knowledge Space where
BJ0979 -              someone "interacted" in some degree. ref SDS 70 0001
BJ0981 -            ..
BJ0982 -           How does this align with the scheme today that has Topic
BJ0983 -           Maps in upper level, and ontology in the middle level?
BJ0984 -           ref SDS 0 MN3W
BJ0986 -            ..
BJ0987 -           Is this the "navigation" engine discussed above? ref SDS 0
BJ0988 -           WG5G
BJ0990 -            ..
BJ0991 -           How does it relate to "ontology" which seems to be both
BJ0992 -           topic maps and linking.
BJ0994 -            ..
BJ0995 -           Does "literate programming" discussed on 000403 support
BJ0996 -           topic map formulation? ref SDS 25 0076
BJ0997 -
BJ0998 -
BJ1000 -           ..
BJ1001 -       •  IBIS
BJ1002 -
BJ1003 -             See comments above. ref SDS 0 AZ7J
BJ1004 -
BJ1005 -             Does this element show up, also, in the bottom level, per
BJ1006 -             below. ref SDS 0 SO5L  Are these different applications of
BJ1007 -             a common function?
BJ1009 -                ..
BJ1010 -               [On 001220 Gary Johnson submits reference to paper by
BJ1011 -               Frank Shipman who cautions against requiring rigid
BJ1012 -               structure based on evidence indicating it reduces
BJ1013 -               productivity. ref SDS 67 RD9N
BJ1015 -                ..
BJ1016 -               [On 010114 example of time and cost savings using
BJ1017 -               "intelligence" from SDS records in Knowledge Space that
BJ1018 -               provide flexible structure, but not rigidity of IBIS.
BJ1019 -               ref SDS 70 0001
BJ1021 -                ..
BJ1022 -               [On 010329 Jack explains SDS he proposes for OHS is for
BJ1023 -               "Structured Dialog System," which sounds like an IBIS
BJ1024 -               objective. ref SDS 77 XZ4K
BJ1025 -
BJ1027 -           ..
BJ1028 -       •  import/edit facilities for generation of major "papers" to
BJ1029 -          submit.
BJ1030 -
BJ1031 -             Is "import" here similar to the scope reported on 001121,
BJ1032 -             ref SDS 60 BO6N, and previously on 000812? ref SDS 42 E9R2
BJ1033 -
BJ1034 -             Right now people can submit a paper of any length, and it
BJ1035 -             is incorporated into the record, for example Drucker's
BJ1036 -             2,000 page book on 931130, ref SDS 3 0001, Landauer's 40
BJ1037 -             page paper on Plato's Problem reported on 960321,
BJ1038 -             ref SDS 8 0001, Jack's letter today.  Don't we have this
BJ1039 -             already?
BJ1040 -
BJ1041 -
BJ1042 -
BJ1043 -
BJ1044 -
BJ1045 -
BJ11 -

SUBJECTS
Epistomoloty Collaborative Evolutionary, DKR Architecture, 000623
Evolutionary Epistomology Middle of 3-layered Architecture for DKR
Engine Graphical Maps Organize Record for Reporting
Lens Filter Graphical Maps Organize Record for Reporting
Lens Filter Topic Maps Graphical Maps Organize Record Subjects Report

BO07 -
BO0801 -      ..
BO0802 -     middle
BO0803 -
BO0804 -          ontology engine builds nodes and links, a vast array...
BO0805 -
BO0806 -             An evolving ontology performs some of the temporal and
BO0807 -             conceptual coupling automatically, and, even notification
BO0808 -             to subscribers of the links found such that those users
BO0809 -             now can come in and add their personal commentary --
BO0810 -             perhaps through an IBIS interface. ref DRT 1 SX5K
BO0812 -              ..
BO0813 -             Automatic action should be real-time, always active.  This
BO0814 -             means that, as new information is deposited into a
BO0815 -             repository (bottom layer), it is automatically scanned,
BO0816 -             reduced to its ontological elements (some will exist,
BO0817 -             others may be new to the system [note] this is the hard
BO0818 -             part [/note]), and a linking engine started. [note] doing
BO0819 -             the ontology thing is saved for another post[/note].
BO0820 -             ref DRT 1 0130
BO0822 -              ..
BO0823 -             Jack says none of this exists in SDS at this time, so far
BO0824 -             as he can tell.  He says his comments to Rod have always
BO0825 -             been that SDS needs this capability, but it cannot exist,
BO0826 -             IMHO, until some commitment to an ontology engine is made.
BO0827 -             ref DRT 1 PR6M
BO0828 -
BO0829 -                "Ontology engine" seems to refer to topic map engine
BO0830 -                Jack discussed on 000623. ref SDS 37 2915
BO0831 -
BO0832 -                   [On 010916 Eric Armstrong reports Eugene Kim has
BO0833 -                   developed Dialog Mapping that does FAQ-indexing
BO0834 -                   which takes a lot of time, ref SDS 88 4J4J, and
BO0835 -                   seems to be monitored by ontology engine Jack has in
BO0836 -                   mind. ref SDS 88 8M6K
BO0838 -                    ..
BO0839 -                   [On 070511 created SDS support to automatically
BO0840 -                   construct new accounts in the Subject Index based on
BO0841 -                   people building a single template account, and
BO0842 -                   automatically assign to Control Fields in SDS
BO0843 -                   records. ref SDS 96 NU3J
BO0845 -                 ..
BO0846 -                Contending that an effective "ontology" capability does
BO0847 -                not exist in SDS overlooks Jack's report to the OHS DKR
BO0848 -                team on 001116 that SDS is "amazing." ref SDS 58 SN4N
BO0849 -                At that time he claimed anyone could produce what he
BO0850 -                sees in SDS records, but did not cite anyone who has
BO0851 -                done it, nor provide examples showing other capability
BO0852 -                can produce an effective Knowledge Space.  The same
BO0853 -                day, Paul Fernhout reported SDS is "inspirational."
BO0854 -                ref SDS 58 H23M  On 000727 Cliff Joslyn reported in a
BO0855 -                presentation to the OHS/DKR team that SDS has a unique,
BO0856 -                powerful KM design, ref SDS 41 3960, from having
BO0857 -                observed SDS support knowledge queries in a meeting on
BO0858 -                000723. ref SDS 40 RN7J  More recently, on 001126 Henry
BO0859 -                van Eykan reported to the OHS/DKR team that he saw SDS
BO0860 -                demonstrated on 001105, ref SDS 56 WG6F, and it showed
BO0861 -                SDS improves human memory. ref SDS 62 0001  On 000324
BO0862 -                Lee Iverson cited research at SRI showing nobody else
BO0863 -                has developed KM capability. ref SDS 21 4877  This was
BO0864 -                supported by Marcello Hoffman on 000330. ref SDS 23
BO0865 -                2627
BO0866 -
BO0867 -                   [On 010223 Jack says creating automated links and
BO0868 -                   subjects is not going to be easy. ref SDS 75 N5RW
BO0870 -                 ..
BO0871 -                On 911121 Morris explained AI system for engineering
BO0872 -                management that was better than SDS, but it turned out
BO0873 -                not to be useful. ref SDS 2 5609
BO0874 -
BO0875 -                   [On 010114 example of time and cost savings using
BO0876 -                   "intelligence" from SDS records in Knowledge Space.
BO0877 -                   ref SDS 70 0001
BO0879 -                 ..
BO0880 -                On 000405 Jack explained ontology in relation to signs
BO0881 -                and symbols, ref SDS 26 1548, which was cited later on
BO0882 -                000713 when Cliff Joslyn reviewed Semiotics.
BO0883 -                ref SDS 39 4078
BO0884 -
BO0885 -                   [On 001214 Jack correlates ontology to use cases and
BO0886 -                   Qualitative Process Theory (QPT). ref SDS 66 0001
BO0888 -                    ..
BO0889 -                   [...he expects OHS will be invented like bingo using
BO0890 -                   QPT. ref SDS 66 VQ3N
BO0892 -                    ..
BO0893 -                   [On 001230 engineer plans to automate subject
BO0894 -                   management, wants to study SDS for ideas.
BO0895 -                   ref SDS 69 NW8R
BO0897 -                    ..
BO0898 -                   [On 010217 KM researcher associates SDS with AI-like
BO0899 -                   search engine, Autonomy, explained in a Wired
BO0900 -                   Magazine article which sounds similar to Jack's idea
BO0901 -                   today. ref SDS 74 YL5F
BO0903 -                    ..
BO0904 -                   [On 010223 Jack explands ideas for "engine" to
BO0905 -                   improve SDS. ref SDS 75 I56H
BO0907 -                    ..
BO0908 -                   [On 010730 DCMA expects KM will be an "engine" to
BO0909 -                   process email. ref SDS 85 WK9K
BO0911 -                 ..
BO0912 -                An ontology "engine" that does "backoffice
BO0913 -                bookkeeping," would generate commitment to use it.
BO0915 -                 ..
BO0916 -                Has this been done by anyone, anywhere?  What record of
BO0917 -                experience shows benefits of this idea?
BO0918 -
BO0919 -                  [On 010425 Opencyc plans release on 010701 of an
BO0920 -                  "inference engine" to support common sense knowledge
BO0921 -                  base ontology. ref SDS 81 079K
BO0923 -                 ..
BO0924 -                The human mind adds subjects in parallel, immediately,
BO0925 -                and continually, subconsciously processing thousands of
BO0926 -                connections related to human needs and desires. see
BO0927 -                POIMS. ref OF 2 0561  How can that be programmed into a
BO0928 -                machine that has no life experience? On 000602 Jack
BO0929 -                reported Ontologos is doing this; on 000607 Ontologos
BO0930 -                said their claim was a mistake. ref SDS 33 0001  On
BO0931 -                000929 explained to Doug Englebart that engineers need
BO0932 -                experience working with SDS to understand scope of KM
BO0933 -                issues, including subject management. ref SDS 48 4F6J
BO0934 -                On 001027 explained this again, ref SDS 53 7B8F, again
BO0935 -                on 001102, ref SDS 55 5V5J   On 001121 explained terms
BO0936 -                for using SDS to gain this experience. ref SDS 59 XU8I
BO0938 -                 ..
BO0939 -                How does this relate to Topic Maps in the upper level?
BO0940 -                ref SDS 0 HM7J
BO0942 -                 ..
BO0943 -                What algorithm can automate formulation of subjects?
BO0944 -                How does it determine form and extent of constructions,
BO0945 -                and what and where alternate constructions are needed?
BO0947 -                 ..
BO0948 -                See Ontology above for comments. ref SDS 0 2X7N
BO0949 -
BO0950 -                   [On 001214 Jack proposes autolinks, and presents
BO0951 -                   scenarios that lead to an OHS. ref SDS 66 01UR
BO0953 -                 ..
BO0954 -                Does the discussion about "lens" technology relate to
BO0955 -                the scope under this section, relative to the record on
BO0956 -                000518, ref SDS 31 2812, and to Jack's letter on 000623
BO0957 -                laying out the 3-layer architecture? ref SDS 37 2915
BO0959 -              ..
BO0960 -             Jack proposes that when linking is done, subscribers are
BO0961 -             notified (e.g. by email), complete with appropriate
BO0962 -             response URLs which, presumably open an IBIS interface
BO0963 -             ready with the new information to which a response is
BO0964 -             solicited. ref DRT 1 TR7G
BO0965 -
BO0966 -                How is linking accomplished?  How does the "engine," or
BO0967 -                algorithm, determine what "chunk" of information to
BO0968 -                link, and where to place anchors and links in Knowledge
BO0969 -                Space?  See the record on 990303 explaining "chunk."
BO0970 -                ref SDS 11 2838
BO0972 -                 ..
BO0973 -                Why do we need "subscribers"?  Is this a way to obtain
BO0974 -                a revenue stream?  SDS let's people create the DKR of
BO0975 -                their choice, i.e., about anything, anytime, anywhere,
BO0976 -                as shown at....
BO0977 -
BO0978 -
BO0979 -                    http://www.welchco.com
BO0980 -
BO0982 -                 ..
BO0983 -                Creating links requires an assessment of alignment that
BO0984 -                shows cause and effect or conflict.  How does the
BO0985 -                "engine" evaluate alignment?
BO0986 -
BO0987 -                    [On 010822 Jack proposes AI technology can form
BO0988 -                    opinions that improve human performance.
BO0989 -                    ref SDS 86 MK4J
BO0991 -                 ..
BO0992 -                Assuming an "engine" can be created to create links,
BO0993 -                what prevents someone from changing the algorithm by
BO0994 -                accident or design that improperly creates links,
BO0995 -                causing huge error and mischief.
BO0996 -
BO0997 -                    A fall back position might be to create an engine
BO0998 -                    that could propose connections that are approved or
BO0999 -                    rejected by a party in interest, i.e., the user.
BO1001 -                 ..
BO1002 -                Since SDS grows constantly, notifying subscribers when
BO1003 -                linking is done would overwhelm any system, and all
BO1004 -                subscribers, per Jack's explanation on 000221 that
BO1005 -                ontology is a Pandora's Box of complexity. ref SDS 18
BO1006 -                7455
BO1008 -                 ..
BO1009 -                How would system infrastructure, and how would people
BO1010 -                avoid being overwhelmed?  How would systems catch up
BO1011 -                when hardware fails, or is otherwise taken out of
BO1012 -                service for maintenance or upgrading?   In the span of
BO1013 -                an hour, much less a day, billions of connections would
BO1014 -                be lost, if this idea is applied widely on a grand
BO1015 -                scale, as seems to be contemplated here.
BO1017 -                 ..
BO1018 -                Jack says in a note continuous background processing
BO1019 -                would be used. ref DRT 1 0181  What happens when the
BO1020 -                continuity is disrupted?  What level of redundancy is
BO1021 -                needed to overcome bumbling?
BO1022 -
BO1023 -
BO11 -

SUBJECTS
Engine Creates Links and Subjects to Improve SDS
AI Cannot Think No Biological Drives Cannot Map Human Experience
Engine AI Black Box Engine Mechanized Knowledge False Dangerous Idea
Engine Process Record
Engine Creates Links and Subjects to Improve SDS
IBIS SDS Improvement
IBIS SDS Needs Capability Proposed by Eric Armstrong
XML Grove Architecture Makes Everything Addressable

BZ10 -
BZ1101 -           ..
BZ1102 -          engine maintains "knowledge structures" (KS).
BZ1103 -
BZ1104 -             KS are explained as nodes and arcs.  It's not going to be
BZ1105 -             that simple, but that's a place to start.  KS must be
BZ1106 -             capable of existing in an evolutionary environment, one
BZ1107 -             which evolves as knowledge representation (KR) needs
BZ1108 -             change. The engine must be, itself, evolvable. This
BZ1109 -             implies extensibility, a plug-in architecture, and so
BZ1110 -             forth. ref DRT 1 XJ6J
BZ1111 -
BZ1112 -                   [On 001230 engineer plans to automate subject
BZ1113 -                   management, wants to study SDS for ideas.
BZ1114 -                   ref SDS 69 NW8R
BZ1116 -                    ..
BZ1117 -                   [On 010223 engine is proposed again. ref SDS 75 3RKS
BZ1119 -                    ..
BZ1120 -                   [On 010405 Grove engine shown in Nexist OHS project
BZ1121 -                   architecture. ref SDS 78 485N
BZ1123 -                    ..
BZ1124 -                   [On 010614 analysis shows SDS organic subject
BZ1125 -                   structure may support KM better than an engine,
BZ1126 -                   until experience is gained that enables an engine to
BZ1127 -                   be designed that supports the architecture of human
BZ1128 -                   thought. ref SDS 83 HY6M
BZ1130 -                    ..
BZ1131 -                   [On 010620 SRI proposes engine called NODAL to
BZ1132 -                   Automatically Manage Distributed Multi-user Change
BZ1133 -                   Tracking and version control, performing an RDMS
BZ1134 -                   function. ref SDS 84 0157
BZ1136 -                 ..
BZ1137 -                Is this the navigation step cited above? ref SDS 0 WG5G
BZ1139 -                 ..
BZ1140 -                See also discussion above for "Ontology." ref SDS 0
BZ1141 -                S29N
BZ1143 -                 ..
BZ1144 -                Jack notes caution about viability to accomplish this
BZ1145 -                objective, per above. ref SDS 0 U49H
BZ1146 -
BZ1147 -                  [On 010425 Opencyc plans release on 010701 of an
BZ1148 -                  "inference engine" to support common sense knowledge
BZ1149 -                  base ontology. ref SDS 81 079K
BZ1151 -                 ..
BZ1152 -                On 960324 Landauer indicates that subjects are complex
BZ1153 -                hierarchial structures, like SDS subject management;
BZ1154 -                would be nice to have an "engine" to help with this
BZ1155 -                difficult task. ref SDS 9 8844
BZ1157 -                 ..
BZ1158 -                On 000405 Paul Fernhout reviewed challenges creating
BZ1159 -                knowledge representations. ref SDS 26 5977
BZ1161 -                 ..
BZ1162 -                On 000623 Jack proposed an "engine" to process
BZ1163 -                information, ref SDS 37 2915, and described an
BZ1164 -                evolutionary expistomology. ref SDS 37 4752  Action
BZ1165 -                item is still pending. ref SDS 37 NE6M
BZ1167 -                 ..
BZ1168 -                On 001025 Jack and Paul discussed this again.
BZ1169 -                ref SDS 52 KY5K
BZ1170 -
BZ1172 -      ..
BZ1173 -     bottom
BZ1174 -
BZ1175 -       •  Raw information, which includes: papers, scanned in books,
BZ1176 -          email, personal commentary
BZ1177 -
BZ1178 -             Does this improve handling of daily working information,
BZ1179 -             cited by Doug Engelbart in the record on 000327,
BZ1180 -             ref SDS 22 3971; is it "intelligence collection" discussed
BZ1181 -             in Doug's 1992 paper, reviewed on 991222? ref SDS 14 8064
BZ1183 -              ..
BZ1184 -             If so, how is handling of "intelligence" improved, i.e.,
BZ1185 -             what needs to be done t convert information into useful
BZ1186 -             "intelligence"?  Provide an example or two.
BZ1187 -
BZ1189 -           ..
BZ1190 -       •  IBIS argumentation records,
BZ1191 -
BZ1192 -             Comments on IBIS are above. ref SDS 0 AZ7J
BZ1193 -
BZ1194 -             How is this different from IBIS in Upper level?
BZ1195 -             ref SDS 0 1Z4I
BZ1197 -                ..
BZ1198 -               [On 010329 Jack explains SDS he proposes for OHS is for
BZ1199 -               "Structured Dialog System," which sounds like an IBIS
BZ1200 -               objective. ref SDS 77 XZ4K
BZ1201 -
BZ1203 -           ..
BZ1204 -       •  Key concept:
BZ1205 -
BZ1206 -          everything in the bottom layer is an instance of an
BZ1207 -          "addressable thing" according to the OHS ontology that Howard
BZ1208 -          is developing.  Being an "addressable thing" means that all
BZ1209 -          elements of the documents in the bottom layer are somehow
BZ1210 -          addressable. ref DRT 1 PY5G
BZ1211 -
BZ1212 -              SDS has a lot of addressability.
BZ1213 -
BZ1215 -           ..
BZ1216 -          Different Doc Types Can be Opened
BZ1217 -          XML Grove Architecture Makes Everything Addressable
BZ1218 -
BZ1219 -          It turns out that the Grove architecture (an SGML notion, now
BZ1220 -          doable in XML) makes everything into an "addressable thing".
BZ1221 -          In fact, a Grove gives you a uniform API for addressing a
BZ1222 -          heterogenious pool of things, each of which might have a
BZ1223 -          different API requirement (e.g. Word documents, XML
BZ1224 -          documents, RDF documents, LaTex documents, Excel
BZ1225 -          spreadsheets, etc).  This all suggests that a part of the
BZ1226 -          middle layer "engine" will be something akin to a Grove
BZ1227 -          engine. ref DRT 1 435L
BZ1228 -
BZ1229 -              If addressability is a key concept for the bottom layer,
BZ1230 -              why is it an engine for the middle layer?
BZ1232 -               ..
BZ1233 -              Getting confused here about the layers.  Need diagram,
BZ1234 -              and some discussion.
BZ1235 -
BZ1237 -           ..
BZ1238 -       •  SDS duplicates raw data rather than link to original sources.
BZ1239 -          This, to me, makes perfect sense if you exist, or plan to
BZ1240 -          exist in an environment that should remain isolated and
BZ1241 -          unreliant on the availability of source information.  In some
BZ1242 -          OHS applications, particularly those where, say, a large
BZ1243 -          enterprise hosts a private OHS with lots of OHSettes
BZ1244 -          scattered about its intranet, then a central repository makes
BZ1245 -          some sense.  For a major web-based OHS community, perhaps a
BZ1246 -          central&mirrored repository makes sense. ref DRT 1 CN3I
BZ1247 -
BZ1248 -              This is a very good point about redundant storage.
BZ1250 -               ..
BZ1251 -              SDS actually strives to do both.  It links to an original
BZ1252 -              source, and also puts a copy where there is direct
BZ1253 -              control to maintain it as an independent record of what
BZ1254 -              was seen, heard and done, just as each human mind has an
BZ1255 -              independent recollection; so that in the event somebody
BZ1256 -              decides to eliminate portions of the record, or loses
BZ1257 -              information due to bumbling, or mischief, the basis for
BZ1258 -              taking action is preserved for the person who took the
BZ1259 -              action.
BZ1261 -               ..
BZ1262 -              The idea of a DKR as a big mainframe where everything is
BZ1263 -              stored, proposed, for example, by Xanadu, has a lot of
BZ1264 -              risks, see review on 000305, ref SDS 20 7733,  If we are
BZ1265 -              going to use the computer to aid human thinking, then we
BZ1266 -              need very high reliability for maintaining our memory,
BZ1267 -              which is what the stuff in the record is.  The Internet
BZ1268 -              and anything maintained by anyone else has near 0
BZ1269 -              reliability.  Someone will get up someday, and "Oh, I
BZ1270 -              don't see why we need this."  Or, "Oh, Corporate says
BZ1271 -              were supposed to do thus and so."  Bamm, a bunch of stuff
BZ1272 -              is wiped out.
BZ1273 -
BZ1274 -
BZ1275 -
BZ1276 -
BZ1277 -
BZ1278 -
BZ1279 -
BZ1280 -
BZ13 -