I think an explicit statement should be made about hold votes of that author.
Downgrading them to novote would be Bad Idea ™, as they might be present for parts
which have 2 cert votes and an Admin cert vote already, thus bringing them immediately
to "certified" status, despite that problems were discovered which led to that hold.
Leaving the votes as-is, as currently described, is the best option, as this forces the other authors
to re-submit a file where the leaving author has put a hold onto.
However, I think this should get mentioned.
The same problem arises when a parts reviewer (not author) leaves us.
We should apply the same procedure then IMHO.
Downgrading them to novote would be Bad Idea ™, as they might be present for parts
which have 2 cert votes and an Admin cert vote already, thus bringing them immediately
to "certified" status, despite that problems were discovered which led to that hold.
Leaving the votes as-is, as currently described, is the best option, as this forces the other authors
to re-submit a file where the leaving author has put a hold onto.
However, I think this should get mentioned.
The same problem arises when a parts reviewer (not author) leaves us.
We should apply the same procedure then IMHO.