Notes 10/21

bugenigmaΛογισμικό & κατασκευή λογ/κού

30 Οκτ 2013 (πριν από 3 χρόνια και 9 μήνες)

117 εμφανίσεις

Speaking Notes

PADM 5500

October 21, 2009

Dr. Neubauer


WHERE WE ARE




We have In the News reports scheduled this evening by Barbara West and
Tejuanda Whitehead. I believe the following students may also be prepared to
present In the News reports this even
ing: Jonica James, Cotyus Parks and Stacy
Jones.



NEXT WEEK we have In the News reports by Quanta Bell and Angela Carr



I returned
midterm examinations last week. If you were not here to get your exam
back, please ask me.



I would like receive the knowledge
management assignment this evening. This
was the knowledge management assignment. It was not lab 2 although I think I
accidentally identified it as such somewhere.



We have only covered the first four chapters of each book so far. According to
the cour
se calendar we are suppose to be covering chapter 8 of each textbook this
evening.



For your Arena modeling project, IDENTIFY and DESCRIBE the existing
process, produce an Arena model OF THE EXISTING PROCESS, and then
SUGGEST an improvement in the existing
system and explain what you think
would be the benefit of the improvement.


BARRETT AND GREENE, CHAPTER 5
--

Who's in Charge?


The CIO is not just the head of the IT Department. The IT Director is probably already
very overworked. There are limits to wha
t one person can do. Ideally, STRATEGIC
LEADERSHIP should be separated out from TACTICAL MANAGEMENT and the two
areas of responsibility reflected in two positions.


Information is now considered as valuable a resource as is money. Most large
organization
s have a CFO. Likewise, it is becoming increasingly common to have a CIO.


Granted, most small cities and counties cannot afford to employ a professional CIO.
Nevertheless, the recognition of the STRATEGIC IMPORTANCE of information
management should be e
vident.


SOMEONE needs to be qualified to make strategic decisions regarding INFORMATION
MANAGEMENT, and that is far more than how many computers (and what brand) to
buy.


Who should the CIO report to? If he or she reports to an ELECTED EXECUTIVE, then
it tends to become a political position. The old idea that there can be a clean separation
between ADMINISTRATION AND POLITICS has fallen. Anyone employed by
government is going to have to deal not only with "office politics" but real politics as
well.


Most CIO's came up through the technical ranks. Such persons may lack advanced
administrative and political skills. It is a rare person who can be a successful CIO.


In state and local governments the role of CIO is becoming more common and
increasingly
important. It is not an easy job. This is a POLICY LEVEL position and
perhaps should not be held responsible for everything that happens at the operational
level. But in a small organization people sometimes must "wear more than one hat."


STAIR AND RE
YNOLDS, CHAPTER 5


E
-
Commerce (E
-
Government) and
Transaction Processing Systems


When the Internet was invented for the military and a few lucky professors, they had no
idea what it could become.


Under "no child left behind" are we partly outsourcing the

education of our children to
India?


Will it come to largely replace radio, television and the telephone; and transform the
music and movie industries?


Will the internet evolve to change the structures of governments and perhaps even the
meaning of democ
racy?


The same technologies (hardware and software) are being applied to both e
-
commerce
and e
-
government.


E
-
commerce "models"



B2B


B2C


C2C


Basically in commerce you have supply chains among organizations and value chains
within organizations.


In go
vernment (in the US) you have relationships among agencies based upon federalism
and regional governance. Plus you have working relationships between agencies and
nonprofit organizations.


Paper
-
based B2B systems are slow and costly. They cannot support
lean "just in time"
production systems. They are not agile in response to changes in the market. Older B2B
electronic systems are based upon a technology called EDI (electronic data interchange).
This is costly and not flexible. Web services and SOA (s
ervice oriented architecture) is
probably the way of the future. This is kind of slow but it is much less costly and
potentially much more flexible.


B2B is also likely to become more portal
-
based, just as B2C e
-
commerce already is.
Basically we are talk
ing about automating business processes both within organizations
and between organizations that are strategic allies.


B2C commerce comes in three varieties . . .




bricks only


real stores in the real world



"bricks and clicks"


real stores supplemented

by an online portal (or a portal
supplemented by real stores)



online only


no real stores


C2C is a situation in which a company facilitates communications or transactions among
others (customers) and somehow has a business model that allows it to make

some
money without being directly involved selling things to customers.


B2B is primarily about saving costs. B2C is primarily about selling something.


E
-
government is primarily about saving money and making government services more
easily available t
o citizens (or facilitating citizens solving their own problems without
involving government employees.)


Government is largely about information.


I think we will see new kinds of payment systems on the Web, such as letting citizens of
Florida pay for cer
tain government services with Sunpass on the Web. We need a micro
payment system for the web that gets us away from the attitude that so many things "have
to be" free. Payments for "toll sites" could be made easy and most people would not
mind paying sma
ll change for premium sites if it was no hassle to make the payment.


Transaction processing systems support processes within organizations and between
strategic partners. They can also extend out to include customers.



Internet


"the big one"


Intranet



"the small one in our organization that works like the big one"


Extranet


"our letting selected others have access to parts of our intranet"


sales portal


the way retail customers can buy things from us


ERP's (Enterprise Resource Planning systems)


These are large, modular "do it all" information systems for organizations.

They generally one on one big enterprise
-
wide database.

They are very expensive and to my knowledge are still based on legacy technologies.

They are difficult and costly to custom
ize.

WHAT IS THE FUTURE OF E
-
COMMERCE, E
-
GOVERNMENT AND
TRANSACTION PROCESSING SYTEMS?




more flexible?



lower cost?



more modular?



easier to "mix and match?"



smaller, more specialized organizational units?



more empowered citizens, less dependent upon go
vernment employees to get
involved in their own "self government?"



CHAPTER 6 OF BARRETT AND GREENE


What they mean by "standards" here is not exactly the same as what technical people
mean by standards. To a technical person, a standard is a technical
standard, such as the
shape of a port or the voltage of an electronic part.


This chapter is more about "business architecture" and the degree to which some central
authority forces all the departments to buy the same kinds of hardware and software.


For e
xample, in Georgia all the state universities are required to use WebCT.


These are some of the benefits of "standardization."




minimize training costs




help insure systems work together




bulk purchase discounts



Here are some of the problems with "stan
dardization."




yet another source of central bureaucracy




possible failure to accommodate specialized needs and preferences




if a vender goes out of business, you may be sunk (for example, Wang)




possible lack of competitive pricing




possible inability to
adopt innovative new things



I suppose requiring every city or county to use the same ERP would be the ultimate form
of "standardization."


The best ground is almost certainly, "in all things moderation."


So what do you do when someone insists that he
or she must have Word Perfect? If the
"middleware" exists to translate the resulting files, it may not be a big deal. If diversity
interferes with the cooperation needed to get the work done efficiently, then it may well
be a big deal.


Macs and PC's c
an coexist, although most businesses run on PC's. Macs are stronger in
areas such as multimedia production.


STAIR AND REYNOLDS, CHAPTER 6


Structured and Unstructured Data (and related applications)


There are different kinds of Management Information S
ystems (MIS systems).


Systems using structured data are most useful for "street level bureaucrats."


Systems using unstructured data are most useful at the strategic (policy) level of
the organization.


Computers are really good with structured data. The
y can be programmed to
play chess because the rules are clearly known.


If the facts are not known, or the rules cannot be clearly stated, human judgment
and intuition become necessary.


The distinction between data and knowledge




data are facts



knowledge

is the aggregation of facts in a meaningful context



relational databases are great for storing and retrieving data



it is up to people to make the translation from information into knowledge



Knowledge Management (KM) is about moving information in an
org
anization into the minds of the people who need to know it. It is about
shared situational awareness, for example.


OLTP and OLAP applications




transaction processing systems support sequences of activities that
happen many times over



decision support sys
tems are designed differently because they have a
different purpose



most relational databases support transactions. They are volatile and
therefore must be designed for EASE OF UPDATE. For this reason, the
design of them is HIGHLY NORMALIZED. In other w
ords, there and
many "skinny" tables and very little redundant storage of data among the
tables.



a DATA WAREHOUSE stores historic data for strategic DECISION
MAKING purposes. Data warehouses tend to be VERY LARGE.
Because they are so large, PERFORMANCE i
s more of an issue than is
ease of update. To improve performance they are designed to be
STRATEGICALLY DENORMALIZED. In other words, there is some
redundant storage of data so that the computer does not have to work so
hard to answer a question.



OLTP ap
plications usually "run over" highly normalized relational
databases.



OLAP applications may "run over" strategically renormalized data
warehouses.



The data in a data warehouse comes from one or more OLTP systems
and is usually updated once a day. The proc
ess is called an ETL. ETL
stands for extraction, transformation and load. There is specialized
software available to do this.


Document management systems




this may just mean scanning paper documents into .pdf files for storage
and retrieval



this also ma
y mean use of XML markup to store the data in documents in
a way that computers can "understand"



a document (form) is really just a complex data structure. XML is a good
way to manage complex data structures using simple ASCII code. ASCII
code is all cha
racters.


Barrett and Greene, Chapter 7


The IT plan needs to fit into the business plan of the organization. If there is no business
plan, how can there be a good IT plan?


Available technologies become available so fast, that future projections become f
uzzy
beyond about two years.


New political leaders may have new goals that affect IT plans. Term limits can also be a
problem.


The strategic plan should answer, WHAT, HOW and WHY?


It should be easier to justify specific requests if ALIGNMENT with the s
trategic plan is
evident.


The long
-
term plan probably should be reviewed and updated at least annually. In should
be a "living document."


If the CIO is "just" a technical person, there is a serious problem.


Stair and Reynolds, Chapter 7


The two majo
r programming languages of artificial intelligence are Lisp and Prolog.


SWI
-
Prolog program for Windows from University of Amsterdam is available at this
URL.

http://www.swi
-
prolog.org/download.html


Here is a very small practice Prolog knowledgebase.


capital_of(atlanta,georgia).

capital_of(tallahassee,florida).

located_in(athens,georgia).

located_in(X,Y) :
-

capital_of(X,Y).


You can run the following queries against it.


capital_of(X,Y
).

capital_of
(X
,georgia).

capital_of(athens,georgia).

located_in(atlanta
,georgia).


The beauty of it is that given the FACTS and the RULES in the Prolog knowledgebase it
can produce the answer to the last query even though it does not have that particular fact.
How do
es it do this?


For more information on artificial intelligence, see the following URL.

http://www.kurzweilai.net/


voice recognition


command


discrete


continuous


rule
-
based expert system

artificial neural netw
ork

genetic algorithms

fuzzy logic (crane example)

knowledge acquisition from experts


tacit and explicit knowledge


system modeling and simulation

http://www.umar.biz/pdfs
/Simulation_Modeling_in_Arena_for_BU395.pdf



Stair and Reynolds, Chapter 8


Systems Development


It is often not possible for an organization to buy the software it needs OFF THE SHELF
and going with an ERP is either not desired or not possible.


There
are two major ways for an organization to obtain customized software for its needs.




Hire consultants to create the new software.



Have in
-
house programmers create the new software.


There are potential problems either way.




You may not want to share info
rmation and data with consultants.



You may not want to enter into a long
-
term working relationship with consultants.



You may not have the necessary programming expertise on payroll.



You may not have the budget or the long
-
term need to add additional
pro
grammers on payroll.


EITHER WAY, you and some of your fellow employees are likely to become involved in
the ANALYSIS necessary to build a new software application.


If this is a major project, be sure your have a qualified PROJECT MANAGER. This
person p
robably has a programming background and also has the skills to participate in
BOTH ANALYSIS AND DESIGN and to manage the project.


Get some "real" END USERS involved. There are at least two good reasons to include
end users.


Have a DEVELOPMENT METHODOLO
GY for the entire SOFTWARE
DEVELOPMENT LIFECYCLE (SDLC).


The SDLC lasts a long as the resulting software if being used. It may take 12 months or
more to do the ANALYSIS, DESIGN, IMPLEMENTATION, TESTING, AND
DELOYMENT.


One approach is called the WATERFAL
L METHODOLOGY. The "spirit" of this
approach is, "no going back." The major problem is that it is not agile enough and it
delays testing until all the code is written, which is VERY RISKY.



PROJECT INITIATION



ANALYSIS OF THE NEEDS




DESIGN OF THE SOL
UTION





IMPLEMENTATION (WRITING CODE)






TESTING







DEPLOYMENT








MAINTENANCE


MAINTENANCE includes "bug fixes" and addition of features as it becomes necessary
to add features. MAINTENANCE IS THE MOST COSTLY PART OF IT!


The newer approach is

called the ITERATIVE INCREMENTAL methodology.




Representation of the Iterative
-
Incremental Methodology

source: Rational Software Corporation (now part of IBM Corporation)


This is the MORE MODERN and BETTER approach and is possible in large part
becaus
e of the OBJECT
-
ORIENTED programming paradigm.


PROGRESS is measured by the completion of PHASES which are broken down into
multiple INTERATIONS. There is a MILESTONE at the completion of every PHRASE.


At the end of every iteration you should REFACTOR
all the artifacts of the project.


Your development team can be doing multiple kinds of work at the same time. It is not
necessary to finish one kind of work before going forward to the next.


You can accommodate (reasonable) CHANGE REQUESTS while sti
ll in development.


You start writing code and testing code quite early along the way.


You can begin to deploy in BETA while still completing the programming and testing.