Carnegie Mellon is Saving Old Software from Oblivion


A prototype archiving system called Olive lets vintage code run on today’s computers.

Researchers’ growing dependence on computers and the difficulty they encounter when attempting to run old software are hampering their ability to check published results. The problem of obsolescent software is thus eroding the very premise of reproducibility—which is, after all, the bedrock of science. …

We created a system called Olive—an acronym for Open Library of Images for Virtualized Execution. Olive delivers over the Internet an experience that in every way matches what you would have obtained by running an application, operating system, and computer from the past. So once you install Olive, you can interact with some very old software as if it were brand new. Think of it as a Wayback Machine for executable content.

To understand how Olive can bring old computing environments back to life, you have to dig through quite a few layers of software abstraction. At the very bottom is the common base of much of today’s computer technology: a standard desktop or laptop endowed with one or more x86 microprocessors. On that computer, we run the Linux operating system, which forms the second layer in Olive’s stack of technology.

Read more at IEEE Spectrum