Have you ever thought about how the world would be like after a cataclysmic event? How would we rebuild when we learned to rely on the internet so much, and all of a sudden there are no shops, tutorials, how-to videos with step by step instructions and so forth. Internet was supposed to be a tool to help us, but since we use it too much it has gradually become a crutch. Well, GitHub is more than aware of the situation and believe it or not they have a plan for it. Storing its open-source codes in an Arctic vault so it can survive almost anything.  

The announcement

Not too long ago GitHub was purchased by Microsoft and in November 2019, they are creating their Arctic Code Vault that will serve as a data repository in the AWA (Arctic World Archive). The facility has been around for quite some time and it is protected by 0.16 miles layer of permafrost in icy mountains of Arctic. The vault is used as one of the preservation strategies and stores more than just codes. On February 2nd, 2020 GitHub plans to capture snapshots of all active public repositories and store them in the AWA.

It is not a solo venture 

In order to ensure the safety of its data GitHub partnered with other companies and organizations. As these open source projects are one of the pillars of modern civilization it is important to preserve them no matter what. These partners include the Internet Archive, the Long Now Foundation, the Arctic World Archive, the Software Heritage Foundation, Stanford Libraries, the Bodleian Library, and Microsoft Research. Since hardware can outlive storage media, we need to ensure how to deal with software susceptible to bit decay, and according to Github, the Archive Program will be able to do that. 

How It Works

GitHub plans to store the code on film reels that are around 3,500 feet long. These film reels were encoded by Piql which a company that specializes in data storage with incredible longevity. It is expected that the code will be able to survive in the archive for at least 5 centuries, but the simulated test suggests that data will make it for an entire millennium. There will be instructions and guides on how to use that data if access if ever needed and the storage will be QR-encoded.  

Is this really necessary? 

AWA itself is one of those structures that you never need until the day comes that you do. So, we should all hope that the day never comes. As far as the code is concerned, who knows. Currently, lots of elements of modern civilization rely on it, which in a way makes it a crucial building block. However, there are tons of sloppy or bad projects in that code, since it is open source. This means we will be preserving some of the quality works, but sifting through all of that and finding quality software might be a bit of a chore.