As GitHub explains in their blog post[1]:

_It will also include works which explain the many layers of technical foundations that make software possible: microprocessors, networking, electronics, semiconductors, and even pre-industrial technologies. This will allow the archive’s inheritors to better understand today’s world and its technologies, and may even help them recreate computers to use the archived software._

[1] https://github.blog/2020-07-16-github-archive-program-the-jo...

There's no git usage, they're just storing a snapshot of each repo's current HEAD in a TAR file (and then compressed and QR-encoded).

> The snapshot will consist of the HEAD of the default branch of each repository, minus any binaries larger than 100KB in size—depending on available space, repos with more stars may retain binaries. Each repository will be packaged as a single TAR file.

The guide included on each reel about how to access the data is public, git isn't involved in the process: https://github.com/github/archive-program/blob/master/GUIDE....

No, it is not QR-encoded.

This is the method:

https://earth.esa.int/documents/1656065/3222865/170922-Piql-...

In the link you post it states "piqlWriter: data writtenas high-densityQR codesEncode binary data to 2D barcode (apply Forward Error Correction)Modulate light using a Digital Micromirror Device (DMD) to project the barcode on the film."

It's a lie. It's not a QR code. I don't know why they say it is. Look at the code. It's nothing like a QR.

They should have said "2D Barcode" but they might be assuming an ignorant audience.

Haha, yes, I have this image in mind of future scientists trying to understand the mess the git CLI is! I'm sure they will have a bunch of theories in which git is considered a religious relic from past humans, used to summon some gods by writing weird incantations, that's already how it feels to me some time :)

The git-cli is an inconsistent mess, agreed.

The data-model, however, is not. It is actually a very smart, mathematical (cryptographical) tree-structure[1].

I'm pretty sure some far future generation will be able to figure out the math behind a "git database". I'm actually more confident they will be able to do so, than your average "excel" or "winrar backup".

-- [1] I really started liking git and understanding what it all is about (and how to use the weird commandline --flags - or was is -flags or -f or -F or flags) after reading the Git From the Bottom Up book: https://jwiegley.github.io/git-from-the-bottom-up/

My understanding is that the vault also includes instructions on how to access the stored projects. I would guess this has to at least explain the minimum ability to replicate git or understand git's format?

The git source repo is probably included in the archive itself. So meta.

Pretty sure the repos are stored plaintext in tar balls