Joshua Lock

@rdimitrov do you think you could take a look at this?

@ethan-lowman-dd can you help us reproduce the results in your screenshot? The PR and merge commit you linked to don't show any Check failures here as we haven't merged the...

As #232 is merged can we rebase this and get it reviewed @znewman01?

np, let us know when you're ready for review!

When/how trusted metadata is loaded for non-root roles is also not part of the current specification.

I really like this idea. It reminds me of @trishankatdatadog's [suggestion](https://github.com/theupdateframework/specification/issues/91#issuecomment-639582774) to include a copy of the root metadata, or at least its hash, in the code. This approach would...

I'll take a stab at this next week.

Rebased on the latest master with a new version and date added. Please take a look @trishankatdatadog @mnm678

I'm not one of the original authors, but my interpretation of the spec and the Diplomat paper matches the expected outcome suggested here.

See related bug in the reference implementation theupdateframework/tuf#920 where adding the same key with a different signature scheme enables a single key to fulfil a threshold of > 1