Long-term storage of digital information

Long-term preservation of digital information is challenging. There are durable storage media, like photographic film, but the problem of interpreting the stored information remains.

On film it is possible to combine analog descriptions – such as text and diagrams – with digitally coded information. However, after hundreds of years the formats used to represent and encode this information is very likely to be lost. In particular, storing computer programs to run in the future is not workable.

Storing and retrieving data

Together with our partners, NR has developed a solution based on three elements: a simple abstract machine; several technology-neutral and analog descriptions of the machine, stored in analog form on film to be read by future programmers and mathematicians; and a C compiler for the machine. Our toolbox currently support storing and retrieving data in the formats JPEG, TIFF, and PDF/A, but other formats can be added with little effort by adapting existing C-programs for processing these formats.

Executable binary files compiled for running on the abstract machine are stored together with data files and the analog machine descriptions. In the future, users can then decode and present information by implementing the machine and then run binary files from the film to process data files from the film.

Further reading

Past projects: Immortal virtual machine: funded by Eureka EUROSTARS 2019-2021 

Code on GitHub 

Key publication: Article published in journal IEEE Access