Notes

Outline

PUAD 626, Information Resource Management First Lecture Government and Information Government Information Managers Information As a Resource History of Computing Platform Wars Toward 2000

Government and Information The Government’s Business Is Information Processing Government runs on information. Information is a major input to government programs; and a primary output of government activity. Imagine no telephones, no fax machines, no copiers. Imagine no computers. No payments to senior citizens, dependent children, veterans, Medicare and Medicaid. Loss of ability to access criminal justice information. These are mission-critical applications, not only important to the agency responsible for administering the programs, but for the State as a whole.  And, worst of all, NO PAYCHECKS.

Government Information Managers Require Skills From Disparate Knowledge Domains

Information as a Resource Characteristics of Information Value of information. A source of supply A source of wealth or revenue A source of expertise An ability to handle a situation Attributes of Information Non-Consumable Indivisible Reproducible Diffusive Leaky

History of Computing 1900 – 1959 1900 – 1949: Conception Computers as electronic “super” calculators, useful only for complex scientific and engineering calculations (WW II); scientific curiosities. 1950s: Birth of the “Electronic Brain” “Stored program” computers and programming languages. “Stored program” computers could carry out varied tasks; programming languages were introduced (e.g., COBOL and Fortran). But computers were essentially PCs — one user at a time, communicating with the machine through toggle switches, paper tapes, and punched cards. Large models involving millions of calculations Computers could be used for large models involving millions of calculations (planning factory schedules, forecasting the weather, predicting elections). Lack of permanent storage (no disks or tapes) precluded use in day-to-day business activities. During the entire 1950s, only a few dozen machine were sold, each costing hundreds of thousands of dollars (in 1950 money).

Slide 6

History of Computing, 1960s Infancy of the “Business Machine” Visions of “data-based” organizations Modern operating systems and permanent storage Modern operating systems were developed that allowed the computer to be a shared, constantly available resource. Permanent storage (particularly magnetic disks) allowed databases, which in turn promoted a vision of a “data-based” organization, in which the computer could act as central coordinator for company-wide activities. Indispensable for preserving large organizations By the end of the 1960s computers had become indispensable for large organizations — but, they didn’t change these organizations, they preserved them, keeping them functional as they grew. Large organizations that were previously drowning in paperwork suddenly found that computers allowed them to preserve their existing structures. Rather than changing existing business processes and practices, computers made them faster. The computer made large, complex, centrally-controlled bureaucracies possible — even though all they did was automate many of the inflexibilities and overheads of the past (and added a few of their own).

History of Computing, 1960s Infancy of the “Business Machine,” cont. IT costs money Modern operating systems and permanent storage By 1970, organizations were spending millions of dollars on IT, a previously non-existent budget category. Along with the big computers and systems, organizations invested in computer departments, programmers, professional and support staff, and custom applications software. Sigh of relief By 1970 it appeared that most companies had finally built most of the applications that made sense. Expensive, probably worth it, and finally containable in cost.

History of Computing, 1970s Adolescence: The “Databased” Corporation Technology breakthroughs — in on-line databases, terminals, networks, and large scale permanent storage — caused existing applications to become obsolete and created a need for a whole class of new applications (circa 1970). The concepts associated with on-line databases — databases connected to and fed by networked terminals — promised an end to management isolation if management would only:    * view information as a key resource;    * build a totally consistent, company-wide, database;    * ensure that information was entered into and       updated in that database as soon as it entered       the company. The vision — all of the organization’s information is instantly available in up-to-date, accurate, and consistent form at the touch of a keyboard, allowing faster operations and enabling everyone in the organization to stay in touch with the real world.

History of Computing, 1970s Adolescence: The “Databased” Corporation, cont. IT costs more money (1% or ½ – 10% of budget) The new class of technology cost even more than before, but by the end of the 1970s a rule of thumb was emerging about technology spending: organizations should target 1% of their expense budgets on IT. But the rule was rough, 1% held as an industry average for manufacturing; but 10% of non-interest expense was more the norm for banks, investment houses, and insurance companies. While ½% was closer to the average for retail organizations.

History of Computing, 1980s Transition to Maturity: Software Engineering for Mainframes In the 1980s, large computer technology stood still (comparatively). Although mainframe computers continued to get better, cheaper, and faster, with new features, large computer technology at the end of the 1980s was very similar to technology at the beginning of the decade. A mainframe application built in 1980 looked virtually identical to one built in 1990. Focus shifted to designing and building applications in some kind of disciplined fashion. In 1970, organizations had dreams of on-line corporations, wired desktops, totally consistent databases. By, 1980, reality set in — the vision was harder to achieve than anyone imagined. In the 1980s, the industry focused on developing the necessary tools, disciplines, and methodologies to make predictably possible the big systems required by the dream. Promised applications had to arrive on schedule and in good working order.

History of Computing, 1980s Transition to Maturity: Software Engineering, cont. By 1990, MIS organizations had learned (more or less) how to build predictable and reliable big systems — using methodologies that prescribe ways to build high-quality systems, and CASE (computer-aided software engineering) tools to support analysts, designers, and developers in building bigger systems better. A methodology is a formal prescription for describing how software should be built. They can range from loose frameworks describing basic design approaches to 60-book encyclopedias that define every step and function required to build an application. Fundamental insight: software design is a formal process. CASE tools offer ways to use computers to help automate the design process and the associated methodologies.

History of Computing, 1980s, ver. 2 Pesky PCs – Toys or Appliances? By 1985, PCs and their tools (WordStar, 1-2-3 (VisiCalc) and dBase) were starting to make users think in new ways. First, although they only addressed special needs, these applications also were dramatically easier to learn and use than mainframe applications. And, the tools could be used to build custom applications – Why wait for MIS to build a forecasting system, when your local hacker can build the same system in three days? Why wait, when the 3-day system will be easier to use, more flexible and run on cheaper hardware? And, why wait when the MIS system will take two years, cost tens of thousands of dollars, and then be too expensive to run anyway? In 1985 PCs were not applicable to business computing: they could not be used to build run-the-business applications. First (and most important), PC applications couldn’t share data. Second, they were not capable of handling large amounts of data.

Platform Wars In 1985, MIS folks looked at PCs seriously for the first time — and they didn’t like what the saw. To the MIS folks who were building mission critical apps, PCs were irrelevant. PCs lacked sophisticated time-sharing and transaction oriented operating systems, had no industrial strength databases, no batch-scheduling facilities, no CASE tools, no modern programming languages. AND, the little applications being built had none of the fail-safe mechanisms or design methodologies that MIS folks now knew were critical to creating bullet-proof applications. Thus arose the cultural split we all live with today. A new breed came out of the ‘80s: yesterday teenage hackers and dBase and 1-2-3 programmers have become today’s client/server gurus. Yesterday’s computer club president is today’s network administrator. AND, the two sides still don’t talk to each other, the PC professionals and large system professionals still don’t talk to each other — they have different value systems, different technical backgrounds, and different beliefs.

Here and Now, 1990s PCs Meet Individual Needs, Not Organizational Needs Mainframe and associated costs up from 1% to 1.5%. Mainframe systems run the business, but they are hugely expensive, hard to use, inflexible, and time consuming to develop. Still, the systems pay their own way — they run the business. Every transaction, every piece of work done, is justified based on ROI (return on investment). New mainframe systems don’t get built without a business justification. PCs, a previously nonexistent budget category, have also jumped to 1.5% of organization budgets in 10 years. The problem is that PCs can’t be used to run the business. PCs make individuals more productive. They’re easy to use, and application development is relatively quick and flexible.  Although individually inexpensive, in the aggregate, they cost as much as mainframes. They require support, network connections and servers, and a variety of other infrastructure and orgware costs. Worst of all, the costs can’t be justified on a ROI basis because PCs meet individual, not organizational, needs. Something has to give.

Here and Now, 1990s Information Costs

Here and Now, 1990s Productivity and Office Automation Mainframe and associated costs up from 1% to 1.5%. Are fancier (and longer) memos better memos? Do they contribute to the “bottom-line”? Are managers good typists? PCs spoil users (GUI). Something has to give. The solution doesn’t start with the technology, it starts with the organization of the business.

Toward 2000 Adolescence: The “Databased” Corporation The Business Front Business Process Re-engineering (and TQM) Moving decision making to the front lines The Technology Front Information access and sharing (LANs, WANs) Distributed computing Client/server computing