A strip of plastic with a magnetic coating
Magnetic tape stores information as patterns of magnetization along a thin plastic strip coated with a magnetic material. A recording head writes the patterns, and a reading head senses them later. The same basic idea was used for sound recording, and it was adapted for computers in the late 1940s and early 1950s.
Several groups worked on tape for computing around the same time. The UNIVAC I, delivered in the early 1950s, used metal tape in its Uniservo drives, and IBM introduced its own tape units shortly afterward, associated with machines such as the IBM 701. Soon after that, plastic-based tape became the common choice, since it was lighter and cheaper than metal.
Why tape was a big step up
Compared with punched cards, the improvement was dramatic. A reel could hold what would have taken very large stacks of cards, and it could be read at a far higher speed. Tape could also be erased and reused, which cards could not. For organizations processing large files, such as payroll, billing and inventory, it was a natural fit.
In the era of mainframe computers, a typical data center had rows of tape drives with reels turning back and forth as programs ran. The image of spinning reels became shorthand for “computer” in films and advertising for decades, even after the real machines had moved on.
Sequential access
Tape’s defining trait is that it is sequential. Data is written one block after another along the strip, and to reach a record the drive must wind to that position. Reading something near the start is quick, while reading something near the end can take a long time. Random access, jumping directly to a spot, is not a strength of tape.
Early software was built around this fact. Batch jobs commonly read an old master file from one tape, merged it with new transactions from another, and wrote an updated master file to a third. The records had to be sorted in the same order beforehand so that each tape could be passed through once. Programmers also paid attention to block sizes and gaps between blocks, because the tape had to be stopped and started between them.
When hard disk drives arrived with direct access, tape did not vanish, but its role moved. Disks took over work that needed quick lookup, and tape became the place for large files that were read or written in bulk.
Tape at smaller scales
Tape was not limited to big machines. In the late 1970s and 1980s, many home computers offered cassette tape as a cheap way to save programs, using ordinary audio cassettes and a recorder. Loading a game could take several minutes, accompanied by a screeching sound, and failures were common if the volume or tape head alignment was off. For many families, a cassette was the first form of storage they used, and it cost much less than a disk drive in the early years of home computing.
Minicomputers and business systems used cartridge formats that were more robust than loose reels. Cartridges enclosed the tape in a protective shell, simplified loading and often held more data in less space. Manufacturers introduced many competing cartridge designs over the years, and a long list of them faded as the market settled.
Backup and archive
By the 1980s and 1990s, the main job of tape in most organizations was backup. The reasons were practical:
- Tape cartridges were inexpensive per unit of data compared with disks, at least for much of the period.
- Cartridges could be removed and stored elsewhere, away from the computers they protected.
- Sequential writing suited the nature of a full backup, which copies everything in order.
- Offline tapes could not be changed by a malfunctioning program or later software error.
Open industry formats such as Linear Tape-Open, which emerged around the turn of the twenty-first century, helped standardize cartridges across vendors. Large organizations, including scientific facilities and cloud providers, have continued to use tape for very large archives, in part because stored cartridges draw little power when they are idle.
Limits and care
Tape has weaknesses that archivists and administrators learned about the hard way:
- Heat, humidity and dust can damage the coating or the plastic base.
- Older tapes can shed oxide or become sticky, and some require careful treatment before they can be played.
- Tape drives are mechanical devices, so drives for obsolete formats can be hard to find or repair.
- A tape that is stored for years without being checked may fail without warning.
For that reason, organizations that depend on tape generally test restores, keep multiple copies, and move data to new cartridge generations periodically. The wider problem is described under digital decay, and the practical response is media migration, which means copying content to a current format before the old one becomes unreadable.
Why it lasted
Tape lasted because it filled a specific niche well. It is cheap for large amounts of data, it works well for long sequential writes, and it can be stored offline. It was slow for lookup and did not suit everyday computing, so it never competed directly with disks or flash for the tasks people see on their desks. That is a common pattern in computing history: a technology that is replaced in the spotlight may continue quietly behind the scenes for much longer than anyone expected.