Strange statement in 2.6

[riking]

It's very cool that the UI is communicating this to us! And, it's also why the format of change IDs is pretty cool: by using letters, there are less likely to be conflicts, which means these are very often very short, until your repository grows pretty large.

Change IDs are reverse hexadecimal, so they're exactly as expressive as regular hexadecimal. I don't understand what "by using letters" is trying to say here.

[steveklabnik]

It could be improved for sure.

a digit has ten possible values

a latin letter has 26

this means all-letters is denser than letters + numbers

[riking]

I do think that thinking of them as "letters" even though they're actually "the last 16 letters of the latin alphabet" does seem to make it easier on my brain-keyboard interface, since I'm not constantly switching between numbers and A-F.

[hintron]

From https://jj-vcs.github.io/jj/latest/glossary/#change-id:

A change ID is a unique identifier for a change. They are typically 16 bytes long and are often randomly generated. By default, jj log presents them as a sequence of 12 letters in the k-z range, at the beginning of a line. These are actually hexadecimal numbers that use "digits" z-k instead of 0-9a-f.

(emphasis added)

I think the statement "by using letters, there are less likely to be conflicts" (I'm assuming you mean in comparison to hex) implies that change IDs have a denser encoding using A-Z, which isn't true. It uses K-Z. (And it couldn't even be A-Z anyways, because A-F is already reserved for hex commit ids. So, the best non-overlapping encoding could only be G-Z.)

And that gels with what you already mentioned earlier in the tutorial:

One really cool thing is that they use a disjoint set of identifiers: qzmzpxyl can never be a commit ID, but must be a change ID, and bc914fcd can never be a change ID, but must be a commit ID.

So I guess if it were me, I would reword it to be something like this:

It's very cool that the UI is communicating this to us! And, it's also why the format of change IDs is pretty cool: by only using letters K-Z, there will never be a conflict with a commit ID.

BTW, @steveklabnik, just went through the tutorial and really enjoyed it. You have a real talent for making things clear and understandable (I've also enjoyed the Rust book and IntermezzOS). Thanks for the work you put in!

[steveklabnik]

Thank you so much!

I'd take a PR with that change, would you mind making one?

[hintron]

Ok, sure. I guess I have to make the change only using jj, as a proof of concept though, so bear with me :)