Last week, the Library of Congress hosted the Preserving.exe conference to discuss the preservation of software and its history. There are quite a few organizations that have made some sort of attempt to collect and preserve old software, but this conference highlighted some of the big questions that these organizations should address, such as whether or not the original source code should be preserved, or just executable code, or additional data such as bug-tracking reports. Metadata standards seem to be a problem, too. It is beginning to look like the two organizations most likely to host successful collections of historic software might be the Internet Archive and the National Software Reference Library (NSRL), a project of the National Institute of Standards and Technology (NIST).
- The Internet Archive aggressively expands its software collection, now the largest of its kind (The Next Web/Harrison Weber) “According to the Internet Archive’s Jason Scott, the organization now hosts ‘the largest collection of historical software online in the world.’ Scott challenges you to find something bigger. This rapid expansion came in part through partnerships with many independent archives, including the Shareware CD Archive, the TOSEC archive, the FTP site boneyard, and the Disk Drives collection.”
- Change computer history forever: Well, here we are (ASCII/Jason Scott) “So what’s the problem? Well, our metadata is shit, I can tell you that. We’re not good at having all the careful twee metadata entry that most archives and libraries demand. If you look at, say, the Apple I manual we have online, it’s kind of just that – an Apple I manual. Not much detail, page listing, context. It’s just there. Preserved, easily accessed, easily read – but not described all that much.”
- Stanford, federal government partner to preserve historic software (Stanford Review/Salil Dudani) “Stanford Libraries owns the [Stephen M. Cabrinety] collection, but in its original media, the software is at risk of becoming lost as time passes. ‘(The collection) completely covers the gamut,’ Michael Olson, technical lead of the Cabrinety project, explained. ‘About 70 percent of it is games, from many different platforms. It also includes early office applications, like spreadsheet applications…We’ve got Commodores, there’s Atari cartridges in there…VIC-20s, tapes, things like that.’ With the aid of a grant from NIST to the tune of hundreds of millions, Stanford is cataloguing and shipping the Cabrinety collection to NIST, who will perform a bit-by-bit imaging of the data on the disks, as well as document all instruction manuals, covers, etc.”
- Life-saving: The National Software Reference Library (The Signal/Trevor Owens) “At the inception of the project, in 2000, organizations were creating and sharing metadata describing computer files on a very ad hoc basis. If the metadata were questioned, it was highly unlikely that the original media were available to resolve the issue. The NSRL operates in the same fashion as an evidentiary locker, with the original media available in the event of a question.”
Moving old software from old media (like floppy disks) to new media for preservation is complicated by the fact that the electrical charge of a digital bit can seem to disappear over time, a problem known as bit rot.