Clarification/Follow-up by arunkulfi on 08/30/03 6:37 am:
However the question is voiceguy HIGHWAYS AND TECHNOLOGY ARE NOT THE SAME IN COMPARISON if that was the case why did bill gates who thought that internet is a waste suddenly created the internet explorer to match netscape.
Or IBM for that matter APPLE loose out to others in computer hardware becoz they kept things which were obsolete or expensive .

I think any company which resists change in the silicon valley will die one day or the other.
why are so many US companies even after opposition from there own people chucked out people in the US and outsource them to india COST @ GETTING MORE WORK DONE.

so i don't necessary agree with ur comparision of highways witha technology .
TECHNOLOGY IS A TOOL OF MANAGEMENT IT KEEPS CHANGING .

THE BIGGEST RISK COMPANIES LIKE MY OWN ARE TAKING IS TAKING NO RISK.

lets look foreward to the future.
MAINFAMES WILL CHAGE HOPEFULLY.

expecting ur reply.
arun

Clarification/Follow-up by voiceguy2000 on 08/30/03 9:02 am:
I guess I made the mistake of thinking you were asking an actual, legitimate question, when in fact you were asking a rhetorical question and simply fishing for people to concur.

You did not want an actual explanation of why mainframes are still in use. You simply wanted somebody to agree with you that mainframes are obsolete.

OK. I agree with you. Mainframes are obsolete.

There. I have said it. I hope you are happy.

Now, let me try again on why something that is obsolete might still be in use. (That was the entire point of the answer I originally posted, which I spent some time on, and which deserved far better than your grudging rating.)

If I have a bank (the example you are using) with thousands of transactions flowing through it every day, worth tens of millions of dollars, I DON'T WANT TO PUT THOSE TRANSACTIONS AT RISK. My core IT capability is truly the heart of my business, much like the heart in a human being. Just as major, open-heart surgery in a human is risky, any change I try to make to the bank's current mainframe-based system is going to involve extraordinary risk. It will almost certainly involve downtime; it will probably not work 100% correctly at first after the change; it may lose a number of transactions, exposing the bank to significant liability and driving customers away; and there is a chance (like the patient dying on the operating table) that something will go terribly wrong and the bank will be left with no system on which to operate. I could literally put the bank out of business by making (or attempting to make) this change.

Major surgery is not normally undertaken on humans unless there is no less-intrusive, less-risky alternative, and the human is in such critical condition that there is no choice.

If I am a manager in charge of a bank, I am not going to allow its central computer system to be subjected to the equivalent of open-heart surgery unless I have no other choice. If the system right now is constantly crashing, constantly losing important transactions, or otherwise not functioning, then I will have to look into alternatives. But if the system is working-- even if it is far from state-of-the-art -- I am not going to let you rip it out and put something else in. That would be foolish from a business standpoint.

The analogy to the busy highway is, in fact, a perfect one. Everybody understands that this highway is obsolete. Everybody would like to change it. But the highway is so important on a day-to-day basis, and changing it would be so expensive and involve so much disruption, that no one is willing to do it. The price we would have to pay to change it far exceeds the price we pay for keeping it in its current state.

In evaluating this bank's decision to stay with the working mainframe system, it is a mistake to decide that the bank is simply "resisting change." That is a thoughtless epithet.

The bank, rather, is weighing the costs and risks of this kind of change against the actual downside of keeping things as they are, and concluding that any benefits of this particular change are outweighed by the costs and risks.

Customers do not choose one bank over another because one bank uses mainframe computers and another uses distributed minis or some other alternative. They choose the bank because they get the service they need, reliably and conveniently. They don' care if the bank is using quill pens in the back room, as long as transactions are processed reliably and at reasonable speed.

If a bank the size of Bank of America or Wells Fargo decided to change out its central computer system, the task would be a staggering one, involving hundreds of branches and back-office locations (including their inbound telephone customer service centers), and tens of thousands of terminals, printers, ATMs, and other dependent devices. It would take years to plan and prepare for the switch (during which time, of course, even the chosen new system would become somewhat obsolete), and even with all the planning and testing in the world, experience tells us that on actual cutover some things would not work under load. With computers that did not work, the bank might lose a lot of customers, and be exposed to legal liability if transactions were lost or corrupted. The business impact could be massive.

It would not be a prudent business decision to put the lifeblood of the bank's business at risk simply to "keep up with technology." Just as with my highway example, the cost, disruption, and risk are simply too high.

I am sorry that you did not understand my example. I told you that it was a parable. A parable is a story that makes a point through an illustration. The point was this: When you have something vitally important that is working, even if it is obsolete and outmoded, there may be reasons to keep things the way they are because it is too costly and too disruptive to modernize them. I am confident that is the reason the bank mentioned in your question continues to depend on mainframe computers for its business.

I answered your question originally because I thought you might find value from a business perspective on the issue you posed, and because you expressly asked for views from "all experts." Because you seem to be one of those people who penalize any expert who does not jump up and agree with you, regardless of the actual quality of the answer or the effort that went into it, I regret now that I even bothered to say anything. Perhaps some other expert will be willing to agree with you (completely ignoring the business issues) and you can reward that person with 5 stars.

Clarification/Follow-up by voiceguy2000 on 08/30/03 5:32 pm:
By the way, I do not want to be heard to say that all mainframe systems are obsolete. Clearly that is not true; mainframes still form the backbone of a great number of large-scale, enterprise-class systems.

IBM still sells a bunch of them, currently offering both the more traditional S/390 and the newer zSeries models, usable with a variety of operating systems.

Thus, my statement would be: Some mainframe systems are obsolete.