Mr. Hite.
I have had some time to review the packet you were kind enough to send to me. I have some questions and comments which perhaps you can clear up for me.
You refer to the Year 2000 problem as a virus. Since a computer virus is a very specific class of intentional computer bugs which affect personal computers, and they are designed, through the indidious cleverness of their authors, to interfere with a PC's operating system, causing damage and replicating themselves to other PC's, why do you consider a design flaw a virus?
The statement "The Analyzer reads all source programs used in a given business system, applying artificial intelligence algorithims to determine and capture all information related to a systems's flow input'output"........this appears to be beyond the ability of modern computing. Would you care to elaborate?
You state " Systems include the option of adopting an entirely new computer software language" and yet the next column says "the AIM system does not generate program ccode nor does it require a compiler"
Comment:
The term "program code" has two usages. The first is object code which is the numeric instructions that the machine can execute. The second is the source code which is the computer language, intelligible to the programmer, that is translated into abject code by the compiler. To say that IVXR has a language but no code is an oxymoron. It would appear to be the same as saying you have developed an odor-free perfume.
You also state "conceptually, all business software is functionally the same" Please elaborate.
In the purple box on page two:
"The specification wqill then operate the same whether it is on a mainframe, mini, or micro." Really! Each computer design and each operating system on the same computer will require extensive programming investment befor any software will run on it. Saying that the AIM system has already been implemented on all cumputer platforms does not seem credible.
Your October 15 internet response saing that "There are several object oriented systems that have been talking to all operating systems for years" Huh? In some sense it is the intention of object oriented programs as well as all computer languages to bridge the gap between hardware platforms so that an investment in software, made on one machine, can be ported to another machine. It is true that many programs have been talking to various operating systems for years. I don't know of any one program that runs on all systems. The closest exception may be the COBOL compiler itself. Some of them are "object oriented systems" which is a different way of writing programs. But each such implementation requires the same major investment as I mentioned above.
It also states "Allows for development, debugging, user testing, and documentation on personal computers, which is less costly. Designeers may be located anywhere there is communications capability. Comment: How will I test on a PC without the data base that was installed on the mainframe? How do I test the response of an automated teller machine which has been installed on the mainframes and is not available to the PC? Has IVXR converted the communication drivers to AIM as well? Two programs can run simultaneously on a mainframe and may interfere with another. This is part of testing. How do I test these on a PC which runs only one program at a time? This is changing as we speak, but the way the programs interact will be different oon each computer platform.
Another statement "AIM software applications incorporate phone messaging, electronic mail" Please tell me how does the telephone line get connected to the computer terminal? Does voice now transmit down data lines. Is a digital signal now converted to voice by the old analog telephone? And who wants this?
In the Baker Report it states "Eliminating code redundancy" An executive cited an expamle of a 14-line instruction that was repeated 87,000 times and required the services of five full time people to maintain. Really! I don't think anyone has seen one person assigned full time to maintain one single program, let alone 14 lines of code, in the past 20 years. If the code was repeated 87,000 times, which seems to be an exaggerated number, it would be placed in a system library and changed once as needed. Each program referencing the code would then get the new version on demand or would be "re-linked" ahead of time, as is said in the IBM terminology.
Please answer these at your convenience. |
