CONTACTS 
0201 - Bootstrap Institute                                                                                                                                                O-00000735 0101
020101 - Mr. Douglas C. Engelbart, Ph.D.

SUBJECTS
OHS Program Code Architecture Under Production
OHS Architecture Eric Armstrong Creating Code for OHS to be Released So  , 001017
Architecture Snag OHS Eric Armstrong, 001018
Architecture Makes Sense Contemporaneous Engineering Record with Code
Architecture Problems KM Linking Versioning, Eric Armstrong, 001018
Linking Architectural Snag KM, Eric Armstrong

2208 -
2208 -    ..
2209 - Summary/Objective
2210 -
221001 - Follow up ref SDS 76 M4W7, ref SDS 74 0001.
221002 -
221003 - Paul Fernhout indicates there are no complete solutions for problems
221004 - Eric encountered in OHS design. ref SDS 0 M4W7  Paul sees an evolving
221005 - architecture, and notes that ad hoc programming errors require a range
221006 - of ways for users to make adjustments. ref SDS 0 R53H  Gil Regev
221007 - requests a summary of the meeting at SRI on 001017.  Jack Park
221008 - submitted Henry Liu's report, and Gil advised this provided helpful
221009 - context.  Jack, also, proposed cross-postings to improve productivity
221010 - on the project.  This indicates possible customer demand for Knowledge
221011 - Management to guide system architecture. ref SDS 0 TF5H
221012 -
221013 -     [On 001122 Eric reports continuing difficulties desiging KM
221014 -     program. ref SDS 80 0001
221016 -      ..
221017 -     [On 001130 IBM reports delays releasing its KM program, Raven,
221018 -     based on Lotus Notes. ref SDS 82 F26K
221019 -
221020 -
221022 -  ..
2211 -
2212 -
2213 - Progress
221401 -  ..
221402 - Snag in OHS Architecture on Linking and Versioning
221403 -
221404 - Received ref DRT 1 0001 from Paul Fernhout offering ideas to address
221405 - the problem Eric reported in his letter, ref DRP 5 0001, submitted to
221406 - the SRI team yesterday. ref SDS 76 TP6O
221407 -
221408 -     [On 001024 Paul cites problem of complexity. ref SDS 77 M4W7
221410 -  ..
221411 - Paul cites Eric's explanation of design problems...
221412 -
221413 -     Some possiblities for resolving the dilemma include:
221414 -
221415 -        •  Drop the content/structure distinction That way, the node
221416 -           classes are uniform. (But making that division is helpful
221417 -           for performance, conceptual clarity, and compile-time
221418 -           type-checking. Otherwise, any node can sit anywhere, even
221419 -           when it doesn't make sense. A simple error in naming can
221420 -           then corrupt the repository in highly damaging ways -- and
221421 -           the error could be on a seldom-used path so it goes
221422 -           undiscovered for years.), ref DRP 5 0149
221423 -
221424 - ..
221425 - ...and then comments that he is...
221426 -
221427 -      Moving in that direction would be my first choice.
221428 -
221429 -      But that's somewhat like the way the Pointrel Data Repository
221430 -      System is set up, so I am biased. For example, the Pointrel Data
221431 -      Repository System has one type of object -- a node which can link
221432 -      up to three things, and which for efficiency can also include a
221433 -      binary string. ref DRT 1 007M
221435 -       ..
221436 -      Then one builds more complex systems on top of this sort of
221437 -      infrastructure.
221438 -      ..
221439 -      Versioning is handled in various ways -- one is simply a
221440 -      list of versions attached to a node representing the abstraction.
221441 -      The sample included with the Squeak version of the Pointrel Data
221442 -      Repository System includes a versioning system for hierarchical
221443 -      tree data. ref DRT 1 YF7N
221444 -      ..
221445 -      One has to accept that the architecture will change over
221446 -      time. It is better to have a flexible substrate. ref DRT 1 OF8I
221448 -       ..
221449 -      Yes adhoc programming errors will happen. They can be also be
221450 -      fixed in an adhoc way, or by rolling back the repository, or by
221451 -      cloning it. ref DRT 1 005M
221452 -
221454 -  ..
221455 - How does all of this contribute to Knowledge Management?
221456 -
221457 -
221458 -
221459 -
221460 -
2215 -

SUBJECTS
Context Management SDS Organization Saves Money Avoid Mistakes
Context Time Chronology Organizes Complexity Faster Better Cheaper
Understanding Time Chronology Experience Causation
Summary Context Increase Understanding of Meetings Intelligence Adds

5606 -
560701 -  ..
560702 - Context, Understanding, Summary -- Customer Demand for KM
560703 -
560704 - Received ref DRT 2 0001 from Gil Regev commenting on a letter from
560705 - Eric congratulating the SRI team, ref DRT 2 0002, for gaining interest
560706 - by NIH to provide funds for sponsoring a KM development program to
560707 - produce an OHS software program, ref SDS 75 4877, using a core team,
560708 - ref SDS 75 01EY, reported in the record on 001017. ref SDS 75 0001
560709 - Eric further says the world needs the project, which aligns with his
560710 - letter on 000927 citing huge needs. ref SDS 65 VX5J
560712 -  ..
560713 - Gil says a summary would help people who cannot attend meetings at SRI
560714 - understand what kind of focus change was decided. ref DRT 2 0001
560715 -
560716 -
560718 -  ..
5608 -
5609 -
5610 - 2107
5611 -
561101 - Another letter from Gil, ref DRT 3 0001, says thanks to Jack Park for
561102 - taking time, ref DRT 3 0002, to submit a letter, ref DRP 4 0001, from
561103 - Eugene Kim on 001018, ref SDS 76 1410, that includes a summary of the
561104 - meeting at SRI on 001017, ref SDS 75 0001, prepared by Howard Liu,
561105 - ref DRP 4 0002,
561107 -  ..
561108 - Jack recommends cross-postings in all cases. ref DRT 3 0002  This is
561109 - another way of saying adding connections that make sense of
561110 - information by adding context.  Jack's proposal for "cross-postings"
561111 - can be accomplished with links that construct trails that Vannevar
561112 - Bush proposed in 1945 to use technology for augmenting intelligence,
561113 - reviewed on 960304. ref SDS 2 L47F  Doug Engelbart recognized this
561114 - idea in a proposal sponsored by SRI in 1962, reviewed on 991222.
561115 - ref SDS 9 7H5H
561117 -  ..
561118 - Gil says it helps a lot to understand the context, ref DRT 3 0001,
561119 - which is another way of saying to get the story of cause and effect,
561120 - i.e., we are doing x because of y, and expect to get such and such
561121 - result.
561123 -  ..
561124 - This record indicates people care about....
561125 -
561126 -      a.  understanding
561127 -      b.  context
561128 -      c.  summary
561129 -      d.  cross-postings
561131 -  ..
561132 - The fervor of Gil's letter suggests an emotional desire, possibly
561133 - rooted in a drive to succeed, to be included in a group, and to save
561134 - time, fundamentally reflecting fear of loss, of being left out, of not
561135 - understanding context that connects cause and effect, essential for
561136 - taking effective action to control the future.  Gil is associating
561137 - context with the power of knowledge.
561139 -  ..
561140 - On 000824 Eric Armstrong feels context is useful, ref SDS 58 G92D; on
561141 - 000926 Henry van Eykan, also, commented on the value of context that
561142 - aids understanding. ref SDS 64 375K
561143 -
561144 -      [On 001024 Jack Park applies Gil's suggestion that context is
561145 -      useful component of knowledge management. ref SDS 78 I58G
561147 -       ..
561148 -      [On 001025 OHS launch plan calls for multiple views, does not
561149 -      require context. ref SDS 78 D6H8
561151 -       ..
561152 -      [On 001121 offered to sell SDS that makes it fast, easy and fun
561153 -      to capture a greater share of organizational memory that provides
561154 -      context of information in relation to objectives, requirements
561155 -      and commitments. ref SDS 79 XU8I
561157 -       ..
561158 -      [On 001126 Eugene Kim explained procedures for understanding
561159 -      context using IT rather than SDS for Knowledge Management.
561160 -      ref SDS 81 BY4K
561162 -       ..
561163 -      [On 011003 Eric Armstrong reports IT doesn't work for doing KM,
561164 -      causes hopeless quagmire of information overload because people
561165 -      cannot understand context. ref SDS 83 EC5N
561167 -       ..
561168 -      [On 011006 letter to John Deneen recommends providing context of
561169 -      links, ref SDS 84 MZ7N, giving feedback to illustrate how context
561170 -      can be presented to aid understanding. ref SDS 84 FF5K
561172 -       ..
561173 -      [On 040312 power of knowledge to control the future derived from
561174 -      locality principle that connects cause and effect, defines
561175 -      "meaning," "context," and "making sense" of complexity.
561176 -      ref SDS 85 YH4G
561177 -
561179 -  ..
561180 - It seems clear that on this occassion, when the deliverable is free,
561181 - Gil "cares about" understanding context, and having a summary to
561182 - acquire it quickly enough to take effective action by understanding
561183 - cause and effect.
561184 -
561185 -     Innate biological drive to understand context reveals benefits of
561186 -     knowledge management presented in a recent letter on 001004,
561187 -     ref DIP 3 TK4H, that proposes making "intelligence" fun.
561188 -     ref SDS 70 6N9G
561190 -      ..
561191 -     On 000709 Bill DeHart explained why people care about better
561192 -     organization that SDS supports. ref SDS 50 1029
561194 -      ..
561195 -     However, if people only care about "intelligence" if it is free,
561196 -     then there is no viable marke, per meeting with Morris on 000706.
561197 -     ref SDS 49 RX3K
561199 -  ..
561200 - Jack's call for cross-postings may be useful for the ontology problem
561201 - he cited on 000221, ref SDS 30 7455, and which seems to address the
561202 - architecture snag Eric reported yesterday. ref SDS 76 M4W7  This
561203 - aligns with Doug's explanation of knowledge management reported on
561204 - 000327, which suggests, by inference, a demand for knowledge
561205 - management. ref SDS 38 3971  These ideas may help the team solve
561206 - collaboration issues which the program sponsor, NIH, wants to improve,
561207 - reported on 001017. ref SDS 75 014M
561209 -      ..
561210 -     Gil, however, does not elaborate on what "understanding" and
561211 -     "context" mean.
561212 -
561213 -     "Understanding" is defined in NWO, ref OF 2 5361, does it relate
561214 -     to Gil's objective expressed today?
561216 -  ..
561217 - It would help to get an explanation of how "context" helps, as Gil
561218 - suggests today.  For example...
561219 -
561220 -     The record for the meeting on 001017, ref SDS 75 0001, provides a
561221 -     summary with links to related details; is this the "context" Gil
561222 -     has in mind that helps understanding?
561224 -      ..
561225 -     The record also provides a summary of headings, ref SDS 75 0000,
561226 -     and action items. ref SDS 75 AI  Does this help people understand
561227 -     what transpired at the meeting to guide follow up action?
561228 -
561229 -
561230 -
561231 -
561232 -
561233 -
561234 -
561235 -
561236 -
5613 -