FYI: Metamath databases, executable, and book are archived as part of the GitHub Arctic Code Vault

[David A. Wheeler]

FYI:

The Metamath repos “set.mm” (databases), metamath-exe (program), and metamath-book (book) are now archived along with lots of other data as part of the “GitHub Arctic Code Vault” project.

The GitHub Arctic Code Vault project made a snapshot on 2020-02-02 of many projects on GitHub and stored them in the Arctic World Archive (AWA), a very-long-term archival facility 250 meters deep in the permafrost of an Arctic mountain. The archive is located in a decommissioned coal mine in the Svalbard archipelago, closer to the North Pole than the Arctic Circle & near the World Seed Bank. Svalbard is regulated by the international Svalbard Treaty as a demilitarized zone. The data is stored on 3,500-foot film reels using silver halides on polyester; the medium is expected to last over 1,000 years.

This archive has a lot more than Metamath. The “GitHub Arctic Code Vault” made a 2020-02-02 snapshot of every repo that met at least one of these criteria:

- had any commits between 2019-11-13 and 2020-02-02

- had at least one star and any commits from between 2019-02-04 and 2020-02-02

- had at least 250 stars.

All three of those Metamath repos met those criteria, so they were all included.

The snapshot consists of the HEAD of the default branch of each repository, minus any binaries.

In set.mm the default branch is “develop”, so that is what was archived.

Note that all of the commit histories are NOT recorded, presumably that would have taken too much space.

I believe that the set.mm repo version stored is commit 34834894a0e8aef244ce100c044623f9c9fcf5b9

(By Jim Kingdon), "Add ltmul1a to iset.mm”.

More details here: https://archiveprogram.github.com/

Podcast discussing it: https://changelog.com/podcast/411

I think it’s *great* that the databases are archived in a long-term vault. I know others here have also been concerned about recording information long-term. While there’s always more that can be done, it’s still good to know. So much important knowledge has been lost in the past, and this is at least a step to prevent that from happening again.

— David A. Wheeler