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: October 17, 2000 02:43 PM Tuesday; Rod Welch

Eric Armstrong developing OHS for DKR project.

1...Summary/Objective


..............
Click here to comment!

CONTACTS 
0201 - Armstrong Consulting                                                                                                                                               O-00000767 0101
020101 - Mr. Eric Armstrong                                                                                                                                               O-00000767 0101

SUBJECTS
OHS Program Code Under Production
OHS Development Vector, Doug, 000419
Architecture Finalize, Doug Meeting with Key People, Eugene Kim
Architecture KM Design Not Established, Eugene Kim
OHS Design Breakthrough Announced, Eric Armstrong

3607 -
3607 -    ..
3608 - Summary/Objective
3609 -
360901 - Follow up ref SDS 78 0000, ref SDS 77 0000.
360902 -
360903 - Received ref DRT 1 0001 from Eric Armstrong who concurs with Eugene's
360904 - summary of Eric's concerns about OHS architecture in Eugene's letter
360905 - on 001015. ref SDS 78 RI9J
360907 -  ..
360908 - Eric says he has constructed some code that can form the basis for a
360909 - distributed, versioned, categorized, link-typable, linked, node-graph
360910 - repository.  He has been developing the design document side by side
360911 - with the code, so the architecture makes sense, and expects to
360912 - announce release of initial code soon (hopefully this week.)
360913 - ref DRT 1 CW5J
360914 -
360915 -        [On 001018 Eric reports a snag in architecture on linking.
360916 -        ref SDS 80 M4W7
360918 -         ..
360919 -        [On 001106 release of code delayed pending resolution of design
360920 -        problems. ref SDS 81 0001
360922 -         ..
360923 -        [On 001122 Eric disappointed about difficulty creating KM, sent
360924 -        letter suggesting he start with narrow objectives, of simply
360925 -        to link. ref SDS 82 0001
360927 -         ..
360928 -        [On 001130 IBM delays release of Raven, its KM program based on
360929 -        Lotus Notes, due to lack of understanding about KM. ref SDS 83
360930 -        F26K
360932 -         ..
360933 -        [On 010723 Eric Armstrong reports NODAL developed by Lee
360934 -        Iverson at SRI solves problems with simplified design for
360935 -        OHS/DKR using Groves and abstract data modeling. ref SDS 84
360936 -        3B7L
360938 -         ..
360939 -        [On 010912 Eric proposes planning for maleable archives to
360940 -        implement email improvements proposed by Jack Park. ref SDS 85
360941 -        406H
360943 -      ..
360944 -     Contemperaneous explanation of code seems to follow the practice
360945 -     Eric used on a project in Texas, reported on 000517. ref SDS 45
360946 -     2064  It aligns with article on good engineering practice reviewed
360947 -     on 951221, ref SDS 2 0966, which Eric supported on 000424.
360948 -     ref SDS 41 0786
360949 -
360950 -         [On 051130 Leonardo da Vinci credited for contemporaneous
360951 -         notes to support creativity, invention, and development of new
360952 -         tools and systems. ref SDS 86 PR4F  AIA Architect recommends
360953 -         diligence exercising the habit of document is critical to
360954 -         design and project management. ref SDS 86 G46F
360956 -      ..
360957 -     How does Eric's planning align with SRI's announcement, at the
360958 -     meeting today, that OHS architecture is not established, and will
360959 -     be determined by the new team being formed to perform NIH funding
360960 -     requirements. ref SDS 79 TJ3G
360962 -      ..
360963 -     On 000424 Eric noted that committing to a methodology before
360964 -     explicitly detailing the goals is risky. ref SDS 41 P95I
360966 -      ..
360967 -     On 000426 Jack Park explained the role of architecture that covers
360968 -     vision, marketing and engineering. ref SDS 42 0304
360970 -      ..
360971 -     The catagory feature cited by Eric today, ref DRT 1 CW5J, supports
360972 -     his OHS/CDS requirements on 000601 for catagories based on the
360973 -     Traction program. ref SDS 48 5933  At that time, he did not
360974 -     explicitly address analysis on 000505 citing Jack Park's work
360975 -     researching ontology, ref SDS 48 593M, and he does not mention,
360976 -     today, how the complexity issue is resolved, as cited on 000221.
360977 -     ref SDS 31 7455
360979 -      ..
360980 -     On 000630 Jack has a lot of questions about Eric's OHS/CDS specs,
360981 -     and hopes coding can commence. ref SDS 53 Q26H
360983 -      ..
360984 -     On 990525 there was a report that managers don't give engineers
360985 -     enough time for engineering management, and that engineers just
360986 -     want to do technology, and so won't do engineering management.
360987 -     ref SDS 3 0966  Eric's report today is a big advance for KM.
360989 -      ..
360990 -     On 000926 there was concern that adding intelligence to daily work
360991 -     required too much diligence. ref SDS 68 004M  On 000824 Eric
360992 -     explained disadvantages of a connected environment. ref SDS 61
360993 -     FJ5H
360994 -
360996 -  ..
360997 - Eric expects to make an announcement possibly this week.
360998 -
360999 -     This seems to address Eric's letter on 000927 asking if quiet
361000 -     progress was occurring on the project. ref SDS 69 0001
361001 -
361002 -     This makes progress called out by Eugene on 000928. ref SDS 70
361003 -     1Y9I
361004 -
361005 -        [On 001016 release of code delayed pending resolution of design
361006 -        problems. ref SDS 81 0001
361008 -      ..
361009 -     Source code available possibly by next week helps respond to the
361010 -     action item on 001004 to develop a schedule on implementing his
361011 -     specs. ref SDS 74 BV6F
361013 -      ..
361014 -     Eric does not advise if this code implements requirements he
361015 -     published on 000614 v0.9. ref SDS 50 0782  He does not indicate if
361016 -     the requirements have been reviewed by the project team, proposed
361017 -     on 000505. ref SDS 44 4864  In a telecon last night Eugene Kim
361018 -     advised that the team review proposed requirements and Eugene
361019 -     plans to submit comments at the end of this month. on 001030.
361021 -      ..
361022 -     Eric does not indicate if this implements Doug's architecture,
361023 -     which he reported on 000824. ref SDS 61 PU5N  On 001015 Eugene Kim
361024 -     explained need for initial architecture to guide production of
361025 -     source code, ref SDS 78 B4P2, and this is based on Augment.
361026 -     ref SDS 78 D4K2
361028 -      ..
361029 -     Eric does not indicate today if this system will boggle the mind
361030 -     as he worried on 000125, reviewed on 000505, ref SDS 44 4524, and
361031 -     which may have been a factor on 000824 objecting to a connected
361032 -     record of project archives. ref SDS 61 FJ5H
361033 -
361035 -  ..
361036 - The code will have a lot of sections still skeletal, and some well-
361037 - understood patterns are missing that could be applied to improve the
361038 - solutions. ref DRT 1 V36J
361040 -  ..
361041 - Eric likes the design, which energizes effort as reported by Grant
361042 - Bowman on 001012. ref SDS 77 B4P2
361044 -  ..
361045 - He developed a design for versioning in a node graph as part of a
361046 - solution for the "competing versions" problem. ref DRT 1 N66O
361048 -  ..
361049 - This work may provide a foundation that can be used to develop the
361050 - functionality.
361051 -
361052 -     Eric may be trying to address the issue of competing versions of
361053 -     the record, which he thought might be a problem using SDS, as set
361054 -     out in his letter on 000405. ref SDS 39 0784
361056 -  ..
361057 - Having defined a core architecture capable of supporting distributed,
361058 - peer-to-peer operation, Eric is a lot less concerned if the initial
361059 - implementations use a simple client and put most of the burden on the
361060 - server.  He feels that if that capability isn't built in at the
361061 - outset, it will be harder to retrofit later. ref DRT 1 A77M
361063 -  ..
361064 - Ensuring the basic functionality opens the door to a whole world of
361065 - implementations -- one of which *should* be a browser-based interface,
361066 - both for making it easy for others to take a look at it, and for
361067 - convenience in accessing the system from some unclean system that is
361068 - not enlightened enough to act as a peer-to-peer server. ref DRT 1 K88K
361070 -  ..
361071 - Submitted ref DIT 1 0001 congratulating Eric for plans to soon release
361072 - code for an OHS, per his letter received above. ref SDS 0 IO6J
361073 -
361074 -        [On 001018 Eric expresses appreciation for encouragement with
361075 -        his project. ref SDS 80 M4W7
361076 -
361077 -        [...Eric reports encountering a snag in architecture on
361078 -        linking. ref SDS 80 M4W7
361080 -      ..
361081 -     Jack's point on 000221 that catagories and ontology is a pandora's
361082 -     box, per above. ref SDS 0 O94F
361084 -      ..
361085 -     Congratulate Eric for following good engineering management
361086 -     practice, so the architecture makes sense, per above. ref SDS 0
361087 -     ID8J
361089 -      ..
361090 -     Ask if code applies CDS/OHS requiremnets v0.9 per above.
361091 -     ref SDS 0 BH3J
361092 -
361093 -
361094 -
361095 -
361096 -
361097 -
361098 -
361099 -
361100 -
3612 -
Distribution. . . . See "CONTACTS"