search

TrustTheVote-Project / NIST-1500-100-103-examples

Other
3 stars 5 forks source link

Use human-readable identifiers for EDF and CVR elements #62

Open ion-oset opened 2 years ago

ion-oset commented 2 years ago

It is much easier to verify and debug EDFs if the @ids are human readable and describe the element and context they are in. For the current test cases we don't need a perfect scheme just a reasonably well-defined one.

Discussion of the details will be ongoing at Human-readable identifiers in NIST documents.

trustthevote commented 2 years ago

Heck yes.

Cliff and I had a schema for this a while back, where there was a human readable that was usually but not guaranteed to be unique with an EDF, appended with an ordinal to be unique.

As I recall contests were like contest-mayor, candidates were like candidate-spacely-contest-mayor and candidate-writein-1-contest-spaceport-board; and question-helium-initiative etc. The ballot section IDs would be directly derived from them like selection-candidate-spacely-contest-mayor and selection-question-helium-initiative-yes

In a later phase I want a generation tools to follow some rubric roughly like this, but for now, the yucky IDs are what we've got to work with on a tight schedule.

ion-oset commented 2 years ago

@trustthevote to be clear: this ticket isn't tied to the current milestones, and we're not trying to solve this before the tight schedule is done. It's here so we can track on it going forward, and because it's come up in BallotLab debugging.

Fwiw we're going to be collecting ideas for that scheme/algorithm at the discussion I linked in the top post.

ion-oset commented 2 years ago

@cwulfman I've realized there's potentially two different issues here.

  1. A general ask to come up with an algorithm or scheme for identifiers (which is what I was posting on based on our recent conversations). That has no immediate milestone or deadline. It might take some care to get corner cases right to our satisfaction, but I think the ones that have been described here and elsewhere are trending towards consensus.
  2. A specific fix to address the needs of a test ballot cited in BallotLab #86.

Do you think it makes sense to split 2 into a separate ticket or is it good enough as it is?