Variant subtag proposal: ALA-LC romanization of Russian

[Frank Bennett]

Not sure whether it matters, but to me, deprecating the
ja-Latn-hepburn-heploc (that I submitted for), in favour of a generic
romanization subtag (say, ja-Latn-alalc09) sounds like a win.  As Doug
mentioned on Wednesday, a "hepburn" fallback would not be available,
but given the ghastly mess that is Japanese romanization "standards",
that's probably not much of a loss -- the backstop at ja-Latn will
probably serve just as well.

This generic tag would let us leverage the work that has gone into
building the ALA-LoC tables in a couple of ways in our context (i.e.
building infrastructure for management of shared bibliographic data).
We not only get to take advantage of all languages in the ALA-LoC
tables without one-by-one filings for new subtags, but we can also use
simpler documentation, since we can just recommend to users that
non-roman script entries in languages represented in the tables should
be supplied, at least, with an XXX-Latn-alalc09 variant.

If a concensus to do this emerges, let me know if there is anything I
should do to help move it forward.

Frank Bennett