Prince of Persia’s Apple II source survived on a 22-year-old floppy
Maintained vintage hardware and specialist knowledge made an extraction possible long after the creator had forgotten implementation details.
A 22-year-old floppy still held Prince of Persia’s source. Getting it off the disk took most of a day and night, plus people who knew the old hardware.
In his repository, Jordan Mechner credits Jason Scott and Tony Diaz with recovering the archive. Diaz’s maintained vintage Apple equipment and expertise helped extract code that might otherwise have disappeared.
The archive contains the 6502 assembly Mechner wrote between 1985 and 1989. He also points readers to a technical document prepared for teams porting the game to other systems. That combination gives a reader both the source and some of the original explanation for moving it elsewhere.
Mechner welcomes study and experimentation, but explicitly says this does not grant Prince of Persia franchise rights. He also admits the limit on asking him to explain the old code: “I don’t remember!”
The next archive worth rescuing may look equally ordinary from the outside. Its readable drive could be harder to replace than its plastic case suggests.
The rescue depended on keeping more than the disk. The people involved needed a way to read it, knowledge of the old equipment and enough patience to finish the extraction. A label on a floppy can identify a treasure while leaving the next owner no practical way to open it. That is a rather fragile filing system for a game people still recognize decades later.
The source and the porting notes also answer different questions. Code records the implementation; a document prepared for another team can explain assumptions that the original author no longer remembers. Keeping those materials together gives a later reader a better chance of understanding what the machine was expected to do without demanding that the creator reconstruct every decision from memory.
A useful first visit to the repository can therefore begin with its explanatory links before trying to change the game. That separates historical study from the work of setting up a build or emulator. Somebody interested in animation or control logic may want to follow a small part of the program, while somebody interested in preservation may care first about how the recovered material was organized.
The rights statement puts a separate boundary around those possibilities. Access to source makes inspection possible. It does not by itself settle what a new distribution may contain. A reader should keep the archive’s explicit statement alongside any plans that go beyond private study, rather than assume a familiar hosting site supplies a different permission.
Mechner’s admission about forgotten details is useful precisely because it comes from the person who made the game. Documentation can outlast even the creator’s ability to explain the work on demand. The next recovery project worth watching would preserve the surrounding notes while the people who understand the disk still have working equipment.