I proposed to add discussion about partial clone (blobless, treeless, and shallow clones) to the book.

For the text, it can be inspired from this GitHub blog post.

In summary:

When you want to clone a repository, but you don't want full clone (for example because you're on metered internet connection), Git provides three ways to get partial clone:

1. git clone --filter=blob:none <url> - blobless clone. Trees and commits are downloaded, but blobs are downloaded on-demand after checkout. Recommended for developers
2. git clone --filter=tree:0 <url> - treeless clone. Only commits are downloaded. Blobs and trees are downloaded on-demand. Recommended for CI builds that needs access to commit history
3. git clone --depth=1 <url> - shallow clone. Download commit history until specified depth level (roughly corresponds to number of commits), in this case 1 commit. Recommended for CI builds.