Easy. It’s not code.

NYT regularly creates interactive stories and those creative developers have no idea how to deal with assets. I deal with that category often and our little projects reach half a gig unless I catch it in time. No PRs, just commit all, merge from remote and push.

One node_modules directory?

Multiple forks, huge blobs of data that someone stuck into a repo for some reason (had issues where someone would git add -A without a good .gitignore and wind up with several ~50-100 MB generated files in a repo)...

The New York Times is a mini-game company with a newspaper attached.[1]

[1]:https://x.com/ballmatthew/status/1774413955940639199/photo/1

Images/Video maybe?

Though I would not want to use that git repo!

It doesn't have to be their source code. They mught just have local fork of every single public repo they have used at least once in a project. I don't know a single company that does cleanup of repos that are unused so it can grow quickly.

Here’s the repo list: https://files.catbox.moe/jx7ksm.txt

It’s a lot.

They don't.

The person who tweeted/xed/whatever doesn't know what he's talking about.

270 gigs of data? Entirely possible.

270 gigs of source code? YeahNo.