Digital obsolescence: Saving software from the march of time
Digital obsolescence isn’t slowing down. Many of the files from yesteryear are lost to time, and we have no way of recovering lost data. However, efforts are being made to preserve retro code and software. Meet some of the teams that are changing how we think about cultural archiving.
Have you ever found an old game from your youth and wanted desperately to relive some of that sweet nostalgia, only to encounter a problem running the program? It’s the dreaded issue that comes alongside aging technology: digital obsolescence. Just as the health of our joints slopes downwards as we age, older technology is facing a crisis. Old software often has impossible read formats, leaving behind broken files, and lost data.
Of course, this goes beyond wanting to play your favorite retro video game. It has a real world impact that will only grow as the march of time moves forward. We are starting to realize that digital files don’t last forever and we often lose the capability to open file extensions. Have you ever sent someone a file who wasn’t able to open it because they were either missing software or their software wasn’t compatible? Imagine that happening, but worldwide. The software and file are gone.
Physical media stability is also a real issue: CDs and DVDs are prone to disc rot, the batteries in our Gameboy Advance cartridges are dying, USBs last only 10,000 writes, and magnetic tape reaches it upper limit at thirty years. What happens to that data?
Museums host archaeological content from long-gone times, why are we not doing the same for software? Much of history has moved beyond the physical and now lives in the domain of code. It’s time to change how we think about preserving digital information. We must ensure that future generations can access the code that we have created.
What efforts are currently being made to ensure that the software of yesteryear doesn’t become a permanent error message?
University to the rescue
Carnegie Mellon University embarked on a project to preserve software and executable content, preventing digital obsolescence. Olive (Open Library of Images for Virtualized Execution) currently consists of 17 virtual machines that run different operating systems and executable files.
Games aren’t the only software entering the archive. Olive also emulates tax-prep software, word-processors, early web browsers, a historical census visualization, and perhaps most impressively, a Cancer, Heart and Soft Tissue Environment simulation. Saving medical software from the dark corners of obsolescence is a noble endeavour. Preservation efforts are securing the future of research and maintaining scientific reproducibility (the cornerstone of the scientific method).
There is of course, no great project without a few potholes. All of that old data is, believe it or not, entered manually into the Olive archive. This slow work hinders the amount that can be processed, so it is a slow (often error-prone) road.
In an article about his efforts, computer science professor Mahadev Satyanarayanan wrote that the problem of GPUs lack of input/output standardization is a problem. “For a long while now, the scientific community has been leveraging the parallel-processing power of GPUs to speed up many sorts of calculations. To archive executable versions of software that takes advantage of GPUs, Olive would need to re-create virtual versions of those chips, a thorny task.”
Accessing the archives
The Internet Archive Software Library currently has a collection in the six digits range and is growing. You may recognize archive.org’s popular contribution to archival efforts, the Wayback Machine. This internet archive provides endless entertainment digging up old sites from times gone by. However, it is also an important historical reference.
There is the usual treasure trove of retrocomputing gaming favorites: MS-DOS, Apple II, and the ZX Spectrum for example. Besides the flood of games, there are also retro text editors, financial inventory organizers, and programming languages from the 80’s.
Carnegie Mellon isn’t the only University on board fighting digital obsolescence. Yale is working on a new project that “seeks to recover more than 3,000 obsolete computer programs so that they can be accessed with modern technology”. The problem of standardization is mentioned by the project’s primary investigator Euan Cochrane. He stated that they hope to establish a standard for digital content. As the project is still in its infancy, there is not much known about it yet. The Software Preservation Network has some extra info on its duration and goals.
Save your files from digital obsolescence
Potential for digital obsolescence is something that everyone should consider, not just for developers but for users of digital media.
Consider some of these points:
- Are you backing up your cloud files elsewhere?
- Do you run any outdated programs with an at-risk format type? The Digital Preservation Coalition has a list of “digitally endangered species” to keep an eye on.
- Archiving online isn’t enough; despite the cautious warning, what’s up on the internet isn’t there forever.
- Physical media degrades even under the best of conditions.