Speaking Notes
PADM 5500
November 4, 2009
Dr. Neubauer
WHERE WE ARE
Stair and Reynolds,
Chapter 8 – Systems Development
This is an important chapter to me. I think it gets at the heart of what general administrations need to understand about systems development.
There are two major models of DEPLOYMENT. One is ABRUPT CUTOVER. The other is some form of PHASED DEPLOYMENT. ABRUPT CUTOVER is high risk but "cleaner." PHASED DEPLOYMENT is safer but may be confusing during the transition and may require RUNNING PARALLEL SYSTEMS for a while, which may not be feasible.
Chapter 8 of the
Barrett and Greene book – Procurement
Efforts to get the fraud and waste out of purchasing (everything) can interfere with the IT because of the time required to take bids, etc.
TCO means total cost of ownership. In the long run, it is want matters. The initial purchase price is just the
beginning.
What is TCO of an iPhone?
CENTRALIZED PURCHASING may lead to departments not getting the hardware and software they really need.
DECENTRALIZED PURCHASING may lead to wasteful purchase decisions by people who do not specialize in IT purchases.
STANDARDIZATION (in this context) is not necessary as long as all the parts work with one another.
Creating an RFP (Request for Proposals) is difficult. It generally assumes a waterfall approach to the SDLC. In fact, it is difficult or impossible to separate out ANALYSIS from DESIGN. Ideally, the vendor should be involved in ANALYSIS, but that is difficult or impossible given the way government contracts are done.
Each vendor's response to an RFP should reflect evidence of an understanding of what the agency wants, even if the "specs" are at best, "fuzzy." The vendor may know a better way, but then the decision becomes, "apples and oranges."
Each project should be carefully scoped. There is often a need for the project specs to evolve, but can changes in scope and features be built into the contract?
Usually, there is a list of approved vendors. That means that agencies don't have to continually invest resources in learning about the qualification of new vendors.
Vendors invest A LOT of money in preparing a response to an RFP. They deserve and expect to be treated fairly.
Unrelated constraints probably should be avoided. Does it really matter if the vendor does any business in Burma?
Requiring small projects to be bid is probably counter productive.
Requiring acceptance of the lowest bid is probably counter productive.
The SERVICE LEVEL AGREEMENT is important!
PERFORMANCE INCENTIVES should be "two
sided." "All sticks" is
probably not a good incentive system.
Chapter 9 of the
Barrett and Greene book -- Outsourcing
Government agencies (and some nonprofits) have difficulty recruiting IT professionals because of higher pay offered by corporations. If an IT person was paid more than the county prosecutor, (for example) that would not be acceptable.
As a consequence, there is a lot of outsourcing of government IT functions.
A city or county should not outsource ALL of its IT expertise. CONTRACT MANAGEMENT requires that somebody who is actually employed by the organization "knows the score." Otherwise, the organization may not realize if a vendor is taking advantage of the organization.
There are probable issues involving unions and political bodies also. For example, can an agency trust outsiders to protect sensitive information, that the agency has legal responsibilities to protect?
Is it wiser to lock into a long-term contract with one vender or to "play the field" over time?
Ideally, the best strategy is probably limited outsourcing of some activities that require costly areas of expertise that the organization cannot afford to hire and cannot use to full potential if employed traditionally.
Chapter 9 of the
Barrett and Greene book -- Training (new material this week)
DOES INFORMATION TECHNOLOGY GENERALLY REQUIRE SMARTER EMPLOYESS OR LESS SMART EMPLOYEES?

"Old pilots" flew by the feel of the stick and rudders. Modern pilots are taught to fly by the instruments and not to trust what they feel in their bodies. Which is easier? Which requires smarter pilots? WHICH IS TACIT KNOWLEDGE AND WHICH IS EXPLICIT KNOWLEDGE?
At the time this (Barrett and Greene) book was written there was a shortage of IT professionals. Cities and counties were at a competitive disadvantage in terms of recruiting and retaining IT professionals.
It is difficult for offices that have operated with paper to suddenly make the transition to automated systems.
Systems can be designed to "empower" employees.
The sad fact is that systems can also be designed to produce the information age equivalent to industrial sweat shops.
BAD REASONS FOR NOT PROVIDING END USERS ADEQUATE TRAINING
WAYS TO REDUCE THE NEED FOR TRAINING