On these seven pages, I have answered over 100 questions submitted by MBA students in the introductory information systems course. Some of the questions are quite straightforward, calling for little more than a definition. Others raise deeply philosophical issues. I have done the best I could here within the limits of time and reasonable space (I’m hoping people will actually read these). Much of this writing was off of the top of my head, so if you feel that an answer needs further elaboration or that one or another of them is simply incorrect, send me a note or post your alternative answer on Blackboard. And, of course, I hope you will continue to raise new questions.
Fred
How do I get to know the basics of computer hardware (i.e. various components of the computer and their basic functions vis-a-vis network application)? [Noah Gyimah]
There are many good introductions to computers. One that I particularly like is A Cartoon Guide to the Computer, by Larry Gornick. It doesn’t really cover networks, but does provide a very sound introduction to the elements of a computer, what software is, and even manages to slip in a bit of information theory. I think that there are copies available in the library and I have one that I can loan out.
For how long can technology continue to be cheaper? [Noah Gyimah]
This is a tough one. Many factors influence the cost of computing—fabrication technologies, the cost of financing, scientific and engineering developments, engineering talent, the market for computers. What is most amazing is that all of these have been acting in concert to produce a remarkably regular rate of increase in the amount of computing that can be purchased at a given price.
What is binary coding?
It is the practice of coding everything in terms of two (bi-) states. We usually use 0 and 1 to represent them. Any code can be transformed into a binary code. Binary coded decimal (BCD) is a way of transforming decimal numbers into binary codes.
What is the difference between a bit and a byte?
A bit represents the smallest unit of information. It allows for distinguishing two states, such as true and false, off and on, up and down, in and out, 0 and 1. A byte is a standard size collection of bits; historically this has generally been 8 bits. It can be used to represent 256 different states (2^8).
When we talk about computers with infinite memory and infinite speed can we talk about computers that feel, in terms of subjective emotions?
It is important to appreciate that thinking about computers with infinite memory and speed is a mental exercise. Of course, nothing infinite can actually be built. But yes, as we build computers with larger memories and faster processing capabilities, interest in the possibilities for artificial intelligence (AI) builds. Even more compelling is the work that is being done on the design of more sophisticated architectures, some of them based more closely on the architecture of the human brain and nervous system. Will these produce something that we'll ultimately recognize as “subjective emotions?” Very intelligent people have taken positions on both sides of that question. I asked the futurist and inventor Bucky Fuller what he thought of the prospects for artificial intelligence. He replied that if it exhibited intelligence, there would be nothing artificial about it.
Why does RAM work more slowly after a period of time?
I am not sure I understand the question, as I have not had the experience of RAM working more slowly. How, indeed, would you know it is RAM. Perhaps what you are experiencing is the result of fragmenting memory (either main memory or more likely your hard disk’s), though I am not sure. If it is the latter, defragmenting your disk should restore its performance. The former I am unfamiliar with but rebooting should solve it. Perhaps more detail will help me sharpen the answer.
What is Ackoff’s warning?
In the December 1967 issue of the journal Management Science, Russell L. Ackoff published an article titled Management Misinformation Systems. In it he challenged the following assumptions, which he thought were implicit in the design of most information systems:
He argued that each of these is false, that a management information systems must be imbedded in a management control system. To do this he proposed a procedure for designing an MIS:
Ackoff warned that failure to question the implicit assumptions and adopt these design practices would produce systems that do not meet the expectations managers have of them.
What are all the laws mentioned at the beginning of the chapter and where did they come from?
Moore’s law is the 1965 observation made by Intel cofounder Gordon Moore that transistor densities would double every 18-24 months. That has been happening, more or less, since. The primary effect has been to produce a halving of the cost of computing on about that 18-month period.
Bob Metcalfe was one of the founders of 3-Com, a leading supplier of computer networking hardware and software. He observed that the value of a network increases as the square of the number of computers and people connected to it. So a network with four people on it is four times as valuable as one with two people on it (4^2=16 vs 2^2=4). This is related to what in economics is known as the law of network externalities. More people using a product makes the product more valuable.
Both of these are more on the order of observations than laws. That is, they have none of the physical force of laws of nature. Indeed, why they should be so is not clear. And in the case of Metcalfe’s law, it is not clear how value might even be assessed.
Would like to have gone into more detail about limitations involved with Moore’s law. Obviously, processors can not double every 18 months infinitely. [Joseph Guillory]
It is nearly impossible to understand the complex combination of forces that together have produced a doubling of chip density every 18 months. In addition to technical improvements in fabrication processes, there has been the invention of new techniques, the availability of capital to fund the required research and to build the plants needed to produce the new chips, and demand for the functionality they provide. If any of these elements were to fall off, the rate of doubling would diminish or halt altogether. Researchers have identified physical limits several times. But often the limits are a function of the cost-effectiveness of the particular technology being used. Normal operating temperatures impose limits that might be overcome by using very cool circuits, electronics imposes limits that could be overcome using optical circuits, and biological circuits offer a promise that is barely understood, much less exploited. In addition, new architectures could provide added levels of functionality at a given density. On the other hand, such breakthroughs are not guaranteed and as your question suggests, improvements cannot continue indefinitely.
Digital has replaced analog, but what is digital? How does it work?
Digitization is the method by which continuous (analog) signals are coded as discrete (digital) signals. It works by laying a grid over the information that we wish to code. The continuous signal is sampled at a regular frequency. Its value is represented by the nearest digital value. This is a function of the resolution. If many bits are used it will be very close to the value measured. If fewer bits are used, less close. Once this new representation, which can be stored using only 1s and 0s is available, all kinds of operations can be done to it, including error correction. This is one reason digital coding is gaining such wide acceptance.
I would like to understand in greater detail why digital technology is replacing analog.
Digital technology has several clear advantages. The first is that it is easy to correct errors (noise) introduced through transmission and amplification of the original signal. Because the noise tends to be small relative to the signal, it is possible to detect it, and then when the signal is retransmitted, to send what was originally intended. Checksums and other clever calculation techniques can further improve on this error correcting. Once signals are digitized (turned into 1s and 0s) they are all basically alike. So a transmission systems that has been designed to carry digitized audio data can as readily carry video or text or even odors once they are digitized. Because all information ends up in more or less the same form, the components that are devised to manipulate it can be mass-produced. A circuit that is used in cash registers, consumer amplifiers, and space shuttles gains from the economy of scale of making standardized components.
At some point won't bytes or the future divisions, be able to be so large that digital will essentially be analog? [John McCabe]
Yes, this is one way to put it. There is a trade-off between word size (bytes are always 8 bits; but words are collections of bytes) and precision. With a larger number of bits, you can discriminate in a finer way. If you have only one byte, you can divide a scale into 256 parts. So, if you want to use a byte to represent parts of a mile you can only code things down to a precision of about 20 feet. If you use two bytes you can code them to a precision of .089 feet.
How do analog computers work?
Digital computers store information using only two values (such as -5 volts and +5 volts) in each storage device. Analog computers use a continuous range of values (such as between -5 and +5) so that a value stored as 3.275 in an analog computer would map onto a different value than 3.5 does. In a digital computer, both of these would be taken to represent a 1, which is what +5 is mapped onto. Different kinds of circuits are used as adders, subtractors, and logic gates.
Is the world really analog? I mean, from a microscopic view, that the world is composed of atoms (or small particles). It is not continuous, more like digital. [Yi-Pin Wang]
MIT physicist Edward Fredkin has posited the view that at the most basic level the universe is digital. He argues that it is the patterns that determine the structure of our world, rather than the particular substrate in which the patterns are embedded. This all gets quite philosophical pretty quickly, but if you have an interest in theoretical physics and computer science you may find his thinking stimulating. An interesting introduction to his thinking is provided by the journalist Robert Wright in his book Three Scientists and their Gods.
I don't understand the whole process of what happens when you turn a computer on.
Things are pretty complicated in those first few seconds. The first thing that happens is that the power supply sends a signal to one of the pins on the CPU (central processing unit). That pin resets the CPU, causing some initial values to be set in the registers. One of the values that is set is the address for the first instruction to be read from RAM (random access memory). That instruction is typically located in the bootstrap ROM (read only memory). It contains a program that will set the computer up for basic operation. Usually it reads additional programming in from the hard disk drive. This additional programming is the operating system (OS). The operating system provides the basic services that are used by application programs. Of course, this is a very abbreviated account. Thousands of operations take place to achieve this result.
From that time on the computer repeatedly goes through a fetch-interpret-execute-increment cycle. It fetches the next program instruction from memory, interprets it, carries out the activity associated with the code in that instruction, and increments the program counter to point to the next instruction in memory. At any given time, the computer is executing only one instruction. Because it can do this so quickly the machine can be made to appear as if it is doing multiple things at one time.
I don't understand what a cache it.
A cache is a high-speed storage mechanism. There are two principal types. One is a memory cache. It is used to store frequently used information in an area of memory (usually static RAM) that is faster than the dynamic RAM that makes up most of the main memory. The other type is a disk cache. Data is moved from the hard disk, where reading it takes longer, to main memory. Accessing data in RAM can be thousands of times faster than on disk, so that cacheing can dramatically improve performance.
I am having trouble understanding the difference between IT and IS. I know the book addresses this briefly in chapter 2, but I nonetheless find myself confused as to the precise difference between the two. Could you please explain this concept? [Jessica Levenberg]
Well, the distinction is not a very precise one. IT, which stands for information technology is usually used by those who wish to focus attention on the role of technology in the discussion. IS, or information systems, is the more broadly used term. It tends to be used when a view of the whole systems—technology and people—is desired. The point of view of the authors of chapter 2 is that the capabilities provided by technology (the IT) determine what is possible in an IS. Therefore, if you wish to understand an IS, you must understand IT.
Exactly how much should a layman know about hardware and software?
As with other fields of knowledge, the more you know the better able you are to deal with important subtleties and details. But, we believe that managers need not become programmers or circuit designers any more than they need to become machinists or mechanics. You should know enough about software to use applications effectively. The line between programming and non-programming is becoming blurred. It is therefore useful to understand some basics of programming, so that you can use some of the more powerful features of sophisticated applications.
I liked the knowledge about how computers work. If possible I would like to learn about the newer technologies (e.g. CDMA).
We offer a course in advances in information technologies, which is a good place to learn about the newset technologies, as well as methods for assessing and evaluating them.
Code-division multiple access (CDMA) is a multiplexing scheme (multiplexing is discussed on the networks page) for cellular phones.
I’m not familiar with computers. Therefore, it is very difficult to understand this chapter [2].
It is difficult material. Just as accounting, economics, statistics, finance, and other disciplines are. You can only get it down by pursisting, peeling back the layers, digging deeper. Then things become clearer. I hope these notes help.
Please provide a diagram of how this all fits together (i.e. hardware, software, networks, servers, browsers,...).
I hope that these notes will help with this. And, I’ll attempt to illustrate in class.
This document was last updated on