Volume 9 No 10 October 1979
November 1, 1979
Director, Department of Fisheries
"The Department of Fisheriesí Future"
Depending on whom you ask, fishing is a relaxing hobby, a fiercely competitive industry, or a controversial state issue, Gordon Sandison, as Director of the Department of Fisheries, determines the policies which regulate fishing both as a past-time and as a business,
Policy making results in headlines when federal statutes conflict with State law, At this monthís meeting of the DP Managers, Gordon Sandison will discuss recent actions of the Department of Fisheries to resolve the dilema of State vs. federal regulations, Fish may not be on the menu, but it will be first on the agenda for Thursdayís lunch,
This monthís personality corner features a person in State Government whom we have all dealt with in our day-by-day activities.
Bobbi Giovannini has been a Data Processing Coordinator for the State Data Processing Authority since August 1973. Her primary responsibility at the DPA is to serve as coordinator for the four year colleges and universities, community colleges and tie Office of the Superintendent of Public Instruction.
Bobbi's responsibilities also include special projects such as the development of various data processing policies and standards, review of agency acquisitions, development of statewide contracts and other management tasks such as staff scheduling and budget development
Bobbiís interest in data processing began while she was an undergraduate student at Washington State University. After graduation from Omak High School, she decided to attend WSU. While there, Bobbi received her Bachelor of Arts degree with a major in Business Administration - Statistics, and a minor in Economics. After graduation she stayed at WSU to work on a masterís degree in Marketing Research and Economics. To support her postgraduate studies, she worked as a teaching assistant. Later she worked for WSU as a computer analyst/programmer. As an analyst/programmer, Bobbi planned, designed, and implemented a Grade Prediction and Placement System used for students entering WSU. She also worked with instructional and research staff in setting up survey analysis and test scoring services.
The Giovanninis then moved to Ellensburg where Bobbi implemented a similar Grade Prediction and Placement System and a Test Scoring System for Central Washington University. The Giovanninisí moved to Olympia in 1967 and Bobbi went to work for the State as an analyst/programmer for the Department of Institutions. For three years she managed the departmentís institutional resident-related information systems.
In 1970, Bobbi transferred to the Office of Financial Management where she was involved with the development of a statewide accounting and budgeting system; later she was responsible for the natural resources program area of Information Services (Pocket Data Book, etc). This assignment involved the planning of information collection, data analysis, and presentations for the executive and legislative branches of State Government.
Bobbi is married to Jim who is a sergeant for the Olympia. Police Department. Jim and Bobbi have two children--Carrie, age 10, and Carmen, age 6. The Giovanninisí share many hobbies which include camping, gardening, swimming, WSU football, youth soccer and other family activities. Bobbi also enjoys racquetball and reading.
In addition to job and family responsibilities, Bobbi is a member of the Washington State School District Data Processing Advisory Committee, has served as past Secretary/Treasurer and Program Chairman for the ADPM, and has completed the Competent Toastmasterís Program while serving as Educational Vice President and Secretary/Treasurer of the Lacey Toastmasterís Club,
Following are some specific questions which this reporter asked Bobbi,
New DPA Member
For those of you who hadnít already heard, Jim Wilson, Division Staff Manager - Planning for Pacific Northwest Bell was appointed to the Authority effective September 12, 1979 by Governor Ray.
DPA Staff Office Move
The DPA staff moved to its new location on the third floor of the p9th and Columbia! building near the Capital Campus (the same building which formerly housed the Conestoga Restaurant). This facility will have essentially the same amount of space as the former location.
The DPA has initiated the process of filling the two vacant Senior Data Processing Coordinator positions (recently renamed Data Processing Coordinator IT). The minimum qualifications of this position have been upgraded. The bulletin for these positions will be opening on October 24, 1979 and a copy will be sent to all data processing managers.
Data Processing Authority
DP Personnel Moving
Ron Pierce, formally of DSHS is now the System Design Manager responsible for implementation of the Integrated Retirement Information System (IRIS) at the Department of Retirement Systems.
Joe Nogel, formally employed by the Office of Fiscal Management, has left State Government.
The Ideal System Design Technology
During the last several years there has been considerabl activity toward developing better methodologies to design computer systems. Some of these methodologies were discussed in the last article, and we concluded by saying that we had chosen the structure chart approach proposed by J. D. Warnier and modified by Ken Orr.
As we have been refining the structure charting technique, we have been asked numerous times what would be the ideal system design technology. If we could develop a method within the next few years that would significantly improve the price performance of people, what would that method be?
To begin, we have to define the ideal system design as an ideal to be reached within a few years. In that respect it may be more a goal than an ideal, however the basic intent is to achieve better price performance of people within a few years. We also must accept some of the basic premise ēof the structural design technology such as total system development, direct user involvement, and the fact that data processing has been its own worst enemy.
We also have to accept the fact that each successive generation of software is very dependent upon the previous generation of software. This development pattern is very apparent in hardware where we would not have fourth and fifth generation hardware without having had the second and third generation hardware to help us design and build the later generations. Third and fourth generation software is going to be heavily dependent on the previous generations for design and development assistance.
As we refined the structured methodology it became apparent that there were only four basic operations in designing systems: single processes, alternatives, repetitions, and parallels. Everything could be explained by one or more of these basic operations. We now realize there are only 10 or 12 basic functions to be performed by any source language, not the 40 or 50 that are apparent by the different types of statements in a source language.
We have also learned to use the basic building blocks to build programs, and programs to build systems, and systems to build applications for the user. The leading system designers today, and those of the future, will carry a variety of models to be pieced together to build an application to meet the users needs. It is very similar to the children today that carry their box of Legoís with them, and, are able to build anything that they and their playmates decide to build. Their only limits are the limits of imagination. We should watch, and learn, from their actions.
Given these basic premises, the ideal design process as we see it today is shown in the figure below. The user and the designer meet to discuss a need or a problem; usually both. Out of this meeting comes a series of requirements. In our methodology these requirements are in the form of objectives, output definitions, and in/out definitions.
U These are all placed on the computer via terminal. The key element of this process is the data structures and the data definitions. The two are tied together by the data element name. The input data definitions specify all of the edit criteria for the element.
Once these requirements are established the user and designer can meet and select the appropriate models to define the processing desired. This process can involve using a fully defined standard model, as an edit model against an input data chart and data definitions, by modifying a standard model to suit a particular situation, or by developing an entirely new model.
This process is currently being performed by computer via terminal with direct user involvement. The number of models is growing steadily, and the speed with which processing charts are developed is phenominal compared to traditional methods. The improvements that need to be made are computer combination of selected models to data structures and data definitions, to produce at least the basic processing chart.
The production of computer logic and people logic charts is the definition of the total system. It is conceivable to develop generators to convert these charts into instructions. The people logic charts are essentially in their final form and the only generation as such is the production of a users manual. This could be produced via computer with interaction of a designer and some predefined set of instruction standards.
The development of source code from computer logic charts offers the largest improvement in productivity. It is possible to define sets of hardware and language specifications and then to build code generators that take the logic chart and the hardware and language specs and produce source code in the specified language for the specified hardware.
We are in the process now of defining the basic set of functions that need to be performed in a source language. These will become the basic functions of a code generator. It appears now that development of code generators are within easy reach.
The advantages of code generators are improved productivity of people, uniform code and documentation development, easier hardware conversions, easier maintenance and upgrades, and improved professionalism on the data processing staff. Code generators offer us the ability to productionalize the process of developing source code. This has been one of the major stumbling blocks in the traditional approach. We have spent a great amount of time writing code, writing errors into the code, coding our way out of design errors, then debugging the lousy code. This process must be changed, but it must be changed as part of an overall revision of the methodology.
We have made great progress in the last few years, and we will continue to make progress. We must learn to use the computer to benefit ourselves and aid our own design and development process. We must improve our design methodology and we must improve the price performance of people. We must cease to be the cobblerís children,
Department of Fisheries
DATA PROCESSING PERSONNEL DEVELOPMENT
FY 1980 SUMMER QUARTER REPORT
During the summer quarter (July-September), the Data Processing Personnel Development Program offered 4 courses for a total of 145 participants from 21 state agencies and two academic Institutions,
A first for the DPPD program was co-sponsorship with the DP lanagers Association and coordination of an Information Forum. Held at Capital High School in July, the two-day forum focused on Distributive Processing with talks by recognized authorities (secured by Patti Palmer, DOT) and presentations by the principal vendors (arranged by Joe Coogan, DPA).
To expand training opportunities, two Computer Assisted Instruction courses, Introduction to Computer Concepts and Introduction to Statistics, were stored at WDPSC for availability through the DHRD teminal and agency terminals linked to WDPSC.
Colleen Bray served effectively as program assistant during summer quarter, Additional assistance in registration was provided by Patricia Richards,
The attached statistic sheets outline the program details. A comparison with DPPD figures from summer, 1978 indicate the following:
|Number of courses offered||6||4|
|Number of participants||93||145|
|Number of class hours||192||56|
|Number of training hours||2352||2192|
Summer quarter disbursements included the following:
|WDPSC (com link modems)||107|
|Total disbursements for spring quarter||$2,435|
The DPPD Advisory Committee for FY 1980* met on 6 September 1979 to finalize the budget and curriculum for thecoming fiscal year. The planned Curriculum includes 62 courses, including some for academic credit. This amount of needed training offerings is approximately equal to the total courses provided in FY 1979. A reduced budget, however, requires that the following three courses will be offered on a chargeback basis: Management of Systems Maintenance (BSI), Structured Systems Design and Structured Systems Update (Orr).
The DPPD staff consulted with staff from five agencies on individual and agency data processing personnel and training needs during summer quarter.
*DPPD Advisory Committee for 1980:
Cliff Cotey, WDPSC - Chairperson
Joe Coogan, Data Processing Authority
Gary Foote, DSHS
John Lawson, Service Center #3
Bill Lundberg, WDPSC Users Assoc.
Don Smith, DP Managers Association
NINE DATA PRO PUBLICATIONS ó IN FIVE LOCATIONS
Several state organizations subscribe to nine Datapro publications. This is a valuable resource to anyone researching the current offerings of data processing, data communications, and word processing hardware/software vendors. To share the cost of monthly updates supplied from Datapro, two State Service Centers and three agencies participate, as follows:
|I. At WDPSC:||ANNUAL COST|
|1. Datapro 70 3 volumes||$410|
|2. Directory of Software 2 volumes||315|
|3. Communications Solutions 2 volumes||280|
|CONTACT: Anne Jewell (3-0946)|
|II. At SC/3:|
|1. Data Communications 3 volumes||$440|
|2. Communications Solutions 2 volumes||280|
|CONTACT: John Lawson (3-1741)|
|III. At the DPA:|
|1. Datapro 70 3 volumes||$410|
|2. EDP Solutions 2 volumes||210|
|3. Minicomputers 2 volumes||400|
|4. Word Processing 1 volume||325|
|CONTACT: Joe Coogan (3-5465)|
|IV. At the Department of Licensing:|
|1. Banking Automation 1 volume||$310|
|CONTACT: Bob Woods(3-2315)|
|V. At the Liquor Control Board:|
|1. Retail Automation 1 volume||$310|
|CONTACT: Mike Phillips (4-1619)|
|STATE TOTAL $3,690|
Association Minutes - October 1979
The October Association of Data Processing Managers meeting was called to order by Chairman, Mike Brackett. There were 28 persons in attendance.
Mary Jo Lavin introduced Gary Reid from the Code Reviserís Office as guest speaker. Gary described the complexities of entering, proofing, and approving rules submitted by state agencies.
Each month all the rules that have been adopted by agencies are published in a register. The register method was created in 1977 to provide the public a review of each rule prior to its becoming law. The total time lag from agency submission to law is 94-124 days. Emergency rules may be adopted earlier but are subject to a 90-day temporary period.
Gary notes that prior to a commitment to data processing, the Code Revisorís Office required up to 18 months to publish an agencyís rules; they used cut and paste methods. Now an agency types their proposal; the Code Revisor enters it into the system; it is proofed by 2 persons; the output is routed to a line printer; and it is photo composed. All re-writes use text processing.
The largest document to date was from the Department of Revenue--250 pages. Labor and Industries rules make up 20 percent of the total volume.
Treasurer, Patti Palmer, reported a balance of $762.08 which includes $5.95 interest earned.
Jim Michal reported on the September 5 DPA meeting:
- Employment Security wrote to the DPA requesting users be contacted during the next Service Center #1 audit.
- Treasurerís Department announced IBM with a 4331 as the apparent successful bidder pending contract negotiations.
- Department of Transportation requested the Deloitte Haskins & Sells survey be made available for review--Terry Wold responded that it was a working paper and the private property of the vendor. Perhaps a meeting between DOT and Deloitte Haskins & Sells could be arranged.
- Higher Education requested approval to use WSU for their data processing.
- Don Tierney reported on the results of a sub-committee investigation into the Administrator of the Courts computer resource requirements. The tour included Pierce, Thurston and Snohomish counties. Recommendation: develop a long range plan to sell the legislature on a full DP implementation.
- Agency submission of their security plans are overlapping with long range plan due dates.
- Acquisitions requested:
|SC #2||Disk Storage||$200K||Approved|
|U of W||CPU Upgrade (CDC)||$65K/mo||Approved|
|Eastern Wa U||Software Pkg.||$103||Tabled unti1 Nov.|
- Next DPA meeting is scheduled for November 7, 1979.
Dick Applestone reported on Job Matrx--A letter from Leonard Nord reiterated DOP's position, i.e., that Tim Seth be removed from the task committee to do his own research. Dick and Mike Brackett will meet with Leonard on Friday, October 19 to discuss several items and determine DOPís official position in working with the ADPM.
Mary Jo Lavin reported on a special training meeting to be held on October 15 to present three consultants' plans for fiscal year 1980. She emphasized that the new programs will be explained by the experts.
The Programmer Entry Program (PEP) has been delayed until January. Agency personnel officers were left out of the communication loop plus agency response was limited.
Don Dahl made a motion to adjourn.
Association of Data Processing Managers Meeting Agenda Golden Carriage November 1, 1979 12:00 Noon
Gordon Sandison will discuss the Department of Fisheries and its plans for the future in light of existing, conflicting regulations.
REMINDER: Please send your notification of attendance to Patti Palmer, Mail Stop KF-0l.