Names such as viaf2603144.viaf002i.perseus-eng1.xml are obviously hard to work with. Can we do better without introducing new complications?
jebb1891 is a starting point that makes intuitive sense to scholars: author name + publication date. jebb1891a, jebb1891b etc. if we have multiple pubs in one year. If we had multiple Jebbs publishing in one year, that would be a problem.
For now we treat translations as versions of the original.
https://scaife.perseus.org/reader/urn:cts:greekLit:tlg0003.tlg001.1st1K-ger2:front.1 -- this brings up the intro to a German edition of Thucydides. In that case, we kept intro and translation in the same file.
Commentaries are not transformations of the source text but they do build on the source text citation scheme.
Instead of 1st1K-ger2 could we use something like
jebb1891-eng1
?