Token to represent initial commit in to a repository?

[stevecondylios]

Comparing two commits from the same branch is made very convenient by the ability to place /compare/<first_commit>...<second_commit> behind any repository URL.

For example:

# Regular GitHub repository
https://github.com/rails/rails

# /compare/

# <first_commit>...<second_commit>
db045dbbf60b53dbe013ef25554fd013baf88134...e0b58f6dc089671cb7ed693346fe0f149b2b363d

put together

https://github.com/rails/rails/compare/db045dbbf60b53dbe013ef25554fd013baf88134...e0b58f6dc089671cb7ed693346fe0f149b2b363d

i.e. rails/rails@db045db...e0b58f6

---

This feature is incredibly powerful.

But it gets better still, because if one of the commits you're comparing is HEAD, you can write that directly into the URL without even having to look up its SHA! E.g. rails/rails@db045db...HEAD

Idea

In the same way we use HEAD, it would be extremely useful to have INIT or FIRST or some token to represent the first or initial commit in a repository.

Example:

rails/rails@db045db...HEAD

could become

rails/rails@FIRST...HEAD

which can be written in its entirely without any lookups, clones, or use of other tools like the API - no fiddling whatsoever.

Why?

Many new projects start with boilerplate code. Therefore, quickly and easily examining the diff between the boilerplate (first commit) and the end result (HEAD) is very powerful.

That can be done with the technique outlined above, but, you still have to get the SHA for the first commit, which is less convenient, especially on large projects where pagination (to get the SHA) is a bit of a nuisance, necessitating cloning or use of the API. Admittedly, this only takes a minute, but it would still be great to avoid it.

What I know so far

I asked here if anyone knew a way and some GitHub users mention in the comments that, to their knowledge, such a feature does not exist. So I figured I'd raise it in case it could be made possible.