SusoSight SusoSight

This page explores what program might be the oldest computer program in the world that is still in active use. This question came up on Slashdot the other day and I wasn't really satisfied with the answers being given. I complained about people giving anecdotal evidence like "banks and government institutions have old computers". One commenter thought I was being unreasonable, but in order to answer the person's question, you need more accuracy than "I think that X company probably has an old computer". You at least need people who have worked at the place to give unofficial testimony that a location is running a certain program that has been in use for a while.

To qualify for this title, the following requirements must be met.

• Running the program on an emulator doesn't count.
• A complete rewrite doesn't count.
• Code changes/patches are ok, as long as some parts of the original code are there.
• It must be a whole program or algorithm.
• It must be in active use. Not archived or being run in a museum for display purposes.

Examples

To explore the answer to this question, let's name some computers with long lifespans

• 1977 - Voyager I and II onboard computer and software
• 1972 - Larry Yeager wrote a program called 'Quick', which is a three-dimensional geometry modeling system that is still in use today
• 1971 - There were probably several programs that were written for Bell Labs Unix and mostly unaltered from this year and still being used.
• 1969 - Nuclear Reactor management software in use at Oyster Creek, NJ

Why older code isn't replaced or rewritten

One person thought it was strange that older code was not replaced. I had a potential explaination for this.

Computers have become comoditized more and more. But the first computers and programs where created for the most important things in the world that needed them. And because they run such important things (Nuclear Power plants, Air Traffic Control Code, Banks, institutions, etc.), the managers and agencies that are in charge of them keep an attitude of "If it ain't broke, don't fix it." Or at least nobody wants to tackle that problem

When that code was created, you probably had to be at the top of your game in order to create such software. But nowadays, any Tom, Dick or Jane can be a programmer, but those aren't necessarily the people that you'd want to have rewriting cuclear power plant control software. And having been a programmer professionally and a system administrator for a while, I think that the people who are capable, don't want to bother with it because there isn't much glory for it.

I actually work at a company that has been very resilient to moving away from their antiquated platform because of the amount of code that would need to be ported, some going back as far as 1982.

But what will happen when it all fails and nobody knows how to fix it or the fix is incorrect? (See, STTNG: When the Bough Breaks).

Other thoughts

• If you think about it, its probably going to be extremely hard to find out who is running older programs simply because those who are, are probably not following the computer industry very well by definition. So they probably would not be reading Slashdot or this page.