Map of/Excerpts from June 11, 2014    Interview of Kenton Andersen and Bruce Miller


Link to complete unedited audio recording of interview

(There seem to be problems with moving to a different place in the recording under Chrome; IE and Firefox are OK.)



0:12:20    Bruce and Kenton marketing NOTIS, 1981-83

0:28:00    Library’s budgetary powerlessness; desire to get income from NOTIS

1:36:50    …“Every school should bring in money.”  “Early days of universities doing technology transfer”

1:48:20    Three main things Jane did to make the system more attractive to the broader market.

2:14:40    Kenton: “I know of no other software company in which the salesperson had carte blanche”

2:20:05    Bruce: “Support of full MARC record”; tight integration (“in early ‘70’s they were all silo systems”)

2:34:15    Kenton:  “I do mean that Geac [hardware] was “Brand X”….  ]


Complete map/excerpts


0:00:00    Introductions

0:00:00    Jerry

0:02:10    Kenton

0:07:00    Bruce

0:12:20    Bruce and Kenton marketing NOTIS, 1981-83

0:14:50    Kenton   Circ, Venezuela Project, 1978-81; Spanish-language support

0:22:10    New Circ; self-service kiosks

·       1/8th of recording

0:24:20    Punched cards à barcodes; barcoding project 

0:28:00    Library’s budgetary powerlessness; desire to get income from NOTIS

0:31:25    GTO

0:31:25    Visit to Stanford/RLIN; genesis of GTO

0:35:10    Catalogers/Catalog assistants

0:37:00    US MARC; Kenton’s work on tag-table module

0:39:50    IBM 4331 processor; moved from VSE, in library, to MVS, at UMS, 1988

                  IBM 9370 installed at UMS for NOTIS Office VSE support

0:41:15    Kenton’s NOTIS projects, 1984-88, while at UMS (besides GTO)

0:41:50    Customer data conversion 

0:44:00    Growth from 8 customers (1983) to 100 (1988)

·      ¼  of recording

0:46:00    Kenton’s involvement with z39.50; OSI; NOTIS VSE àMVS conversion (1981-2)

0:49:00    U. Florida / FCLA

0:52:00    Kenton installs:  Clemson, Auburn, Washington University (1981-4)

0:53:15    Jane’s hiring

0:53:50    Kenton:  work on LUIS (OPAC) module; origin of “Guide Screen” concept; “epiphany”;

                  Author-title-subject index; compression

0:59:30    OPAC transaction log

1:00:50    NUL computer room; 4331; quick response compared to modem-connection

1:06:15    Lee Ellis; library budget

·       3/8 of recording

1:08:10    IBM macro-level / command-level; CWA (Common Work Area);

                 Telex 476 (special library terminal); “no more than 24K of RAM”; even under CICS, only 64K; Small footprint

1:15:50   Most proud of having done … 

               1:15:50  Kenton

     1:16:30   Bruce

1:17:50    Why no NOTIS ILL module?

1:22:10    Velma, NUGM 1993 speech … frustration with course of events; “dollar signs in eyes of University Librarian and University Administration”; “marketing people want commitments that suit their marketing purposes”.

1:27:00    “marketing operation” began in 1983 with Jane’s hiring

1:28:00    John as driving force?   1980 EDUCOM report;    

1:30:00    Library as “profit center”

·       ½ of recording

1:32:00    Bruce:  Velma instilling “systems analysis” mentality in library staff

1:33:40    Bruce:  “Behind the scenes was the strong unrelenting support of John McGowan”;  “Change from old European ‘bookman’ model….”

1:36:50    John not factoring Jane into the equation? 

                 “Every school should bring in money.”  “Early days of universities doing technology transfer”

1:41:40     “Certainly the [marketing] model changed with Jane.”  NU’s problem with “unrelated business income” solved by creating NOTIS Systems Inc.  Endowed positions in library created by sale; $10+ million in 2014 dollars.

1:48:20    Three main things Jane did to make the system more attractive to the broader market.

1:48:20    1)  Real-time bibliographic interface

1:49:20    Bruce on GTO

1:53:05    Central State’s OCLC Transfer

·       5/8 of recording

1:54:35    GTO improvements: error handling; robustness; handling of RLIN (and other vendors); COMARC, Cooperative MARC Cataloging

2:01:20    2) Keyword/Boolean

2:01:20    Importance of

2:04:45    Difficulties implementing

2:06:10    “Wave in the marketplace of people wanting keyword/Boolean”

2:09:45    3)  Packaging:  training, documentation, and support

2:13:00    “Jane committed to things without telling anyone”; prioritization of fixes

2:14:40    “I know of no other software company in which the salesperson had carte blanche”

2:15:20    To noble-minded people, the greatest satisfaction comes from the creation of a really good system which serves their clientele (Northwestern University Library) … really well. 

·       ¾ of recording

2:17:10    “in terms of actual profit”

2:17:30    Jerry:  Keyword/Boolean is a very interesting and challenging function which I think Jim and Kenton could have done very well.

2:18:30    J:  The indexes are 10 times the size of the bib data; speed/efficiency really important.

2:20:05    Reasons for NOTIS’ success….

        “Support of full MARC record”

        Bruce:  Tight integration (“in early ‘70’s they were all silo systems”)

2:22:50   K:  “I don’t recall many other vendors having strong authority-handling capabilities”

2:25:25   K:  “It’s about 50/50 [between IBM hardware/software & NOTIS programming efficiency]”.  By early ‘90’s minis had caught up; but efficiency of Assembler code was still a real advantage.

2:28:45    K:  Marketing; 100% of ARLs could have benefitted from NOTIS; overseas sales: “missed opportunity”?

2:32:00    “What would have happened if Jane hadn’t come into the picture”

2:32:35    How important was running on IBM?

2:33:30    K:  “I’m not aware of too many institutions that were trying to do administrative computing on anything other than IBMs.”

2:34:15    K:  “I do mean that Geac was “Brand X”.  No university bought Geac hardware --except for the library.”

2:36:00   J:  2/3 of revenue from MVS sites (running on shared computers)?

2:39:00   K:   Another advantage:  “The fact that NOTIS did not start out as a commercial product.” Jim and Velma were important factors in success.


·       7/8 of recording


2:41:20   Natural proselytizing:  “NUL staff moved to other libraries.”

2:42:15    How about NOTIS giving customers the source code?

2:43:30   K:  Distribution of object code: “You need a pretty good-sized client base to take that on.”

2:44:30    K:  “For products at a certain price point it [distributing source] is definitely the way to go.”

2:45:30    K:  “It was licensed source code [different than open source code].  That was, at the time, IBM’s business model.”

2:46:40    K:  [By getting away from source code distribution] “IBM shot themselves in the foot.”

2:47:10    Use of Assembler.

2:50:00    K: “People who’ve worked with other computers’ assembly language don’t appreciate that IBM Assembler was actually pretty easy to code.”

2:50:40    J: Use of comments on most lines made the code quite intelligible – even for someone who knows nothing about Assembler.

2:51:55   B:  “Jim was actually far-sighted in picking the mainframe Assembler. “

2:54:15    J:  Good programming (in any language) involves the creation and extensive use of efficient subroutines. “And I think NOTIS had some very efficient subroutines.”

2:58:15    K:  “Every system that’s been really successful has had a certain ‘sweet spot’.”

3:02:10   “NOTIS Horizon” became Voyager.

                 “There are a lot of NOTIS-like features in Aleph.”

                 “Good features become the standard; they become known….”