Skip to Content
Computing

What Is the Oldest Computer Program Still in Use?

Keeping software alive for just a few years without constant updates and overhauls might seem nearly impossible. But some software systems remain in fine fettle decades after their launch.

It’s widely accepted that the first computer program was written by Ada King, Countess of Lovelace, in 1842, although a device for which it was intended wasn’t built in her lifetime. But a reader asks: what is the oldest computer program that was not only written and put into use, but remains active today?

The Pentagon

In 1958, the United States Department of Defense launched a computerized contract-management system that it dubbed Mechanization of Contract Administration Services, or MOCAS (pronounced “MOH-cass”). The system was designed to use the latest in computation and output technology to track contracts in progress and payments to vendors.

Fifty-seven years later, it’s still going.

MOCAS is written in COBOL, a language that wasn’t formally approved until a couple of years later, and it was likely initially created in a similar predecessor called FLOW-MATIC, which was developed by noted computer scientist and U.S. Navy rear admiral Grace Hopper. MOCAS’s original interface relied on punchcards or keycards, says Michael Graham, the information technologist responsible for managing MOCAS. In the following decades the program got upgraded to work with what’s often called “green screen” access: a terminal-style system that remained in broad use at airlines, travel agents, banks, and telecom companies until just a few years ago. “I’m not sure I would call it a graphical user interface,” says Graham.

You can still find antique green-screen systems if you look hard—in some cases, a pleasant Web interface just disguises the old guts. Indeed, the Defense Department has built newer interfaces that connect to MOCAS. This has kept it usable even as the world has passed it by. The current system is integrated with several other software packages, meaning that a user today can attach, say, Microsoft Word documents to records.

Trillions of dollars have passed through the computational records in MOCAS. In its current form the system is managing roughly $1.3 trillion in obligations and 340,000 contracts. Its current hardware configuration is an IBM 2098 model E-10 mainframe that can carry out 398 million instructions per second. It has a modest eight gigabytes of RAM, and all kinds of attached storage devices.

There have been efforts in the past to build a full replacement for MOCAS, and they’ve sputtered due to cost, complexity, and transition planning. Because the system handles so much that’s in progress and critical to the DoD, any new system has to overlap and perfectly hand off everything underway. The government is asking vendors once again to submit bids to shed this highly functional vestige of the past.

Punch cards

MOCAS was the oldest software we could verify, but it may not be the oldest still in routine use. Experts in banking and telecom suspect some funky outliers may remain clacking away in a back office somewhere, though it was impossible to find examples, partly because of changes in the late 1950s and 1960s that rendered some early systems obsolete in their industries.

If one stretches the definition of a software program, the oldest one in use is probably at Sparkler Filters, a maker of water filtration devices that was founded in 1927 and is based in Conroe, Texas. To this day it relies on a 1948 IBM 402 punchcard system for inventory and accounting along with an IBM 83 sorter and IBM 129 keypunch, according to the firm’s long-time repairman, Duwayne Lafley, who lives in New Mexico. The IBM 402 doesn’t have memory, though: it relies on programs physically wired onto plugboards, which are swapped out for the task at hand. In 2013, the firm said it was moving toward PCs, but it hasn’t switched over yet, Lafley says.

The government report that introduced the programming language COBOL.

Coding in the dark

In May 1972, the go-ahead was given for what have become the longest space exploration missions yet. Voyager 2 and then 1 were launched in 1977. Both probes continue to send data back to earth from the farthest visited reaches of space.

The two craft are nearly identical, down to their redundant sets of three computers: the flight data subsystem, the computer command subsystem, and the attitude and articulation control subsystem. Together, these keep the probes on track, pointing their high-gain antennas back to Earth, and managing scientific instruments. The onboard cameras haven’t been used since the probes finished their flybys of outer planets.

Thirty-eight years in, the Voyager software keeps on ticking. Suzy Dodd, the project manager for Voyager, says the software has been updated in flight, but NASA has never “shut it off or changed it completely.”

Because the probes have so little memory—the equivalent of about 70 kilobytes total—pieces of code were swapped in and out during different mission phases. This happened 18 times during the Jupiter flyby alone, the mission’s long-time project scientist, Ed Stone, told me in 2013. Once, in 2010, Voyager 2 began to send back garbled scientific data. Testing at the Jet Propulsion Lab, where Voyager was designed and is still managed, revealed that a single bit in a program had flipped from 0 to 1, according to Stone. The program was reloaded, and has worked since.

Its maximum lifespan, however, can be thought of as 48 years: 1977 to 2025. In 2013, Voyager 1 passed beyond the far reaches of the sun’s magnetic influence, putting it in the midst of particles that are more highly charged than the ones in our solar system. By 2020, mission scientists will have to shut down some remaining instruments, but the craft should continue to talk to earth until it pings home for the last time around 2025.

Thanks to Christopher Eddy for this week’s question. If you have one, send it to questionoftheweek@technologyreview.com.

Deep Dive

Computing

Inside the hunt for new physics at the world’s largest particle collider

The Large Hadron Collider hasn’t seen any new particles since the discovery of the Higgs boson in 2012. Here’s what researchers are trying to do about it.

Why China is betting big on chiplets

By connecting several less-advanced chips into one, Chinese companies could circumvent the sanctions set by the US government.

How Wi-Fi sensing became usable tech

After a decade of obscurity, the technology is being used to track people’s movements.

VR headsets can be hacked with an Inception-style attack

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.