> Andrew Conkling wrote:
> > At this point, I don't even care about any sort of CTTS; I'd really just
> > like to see some consistency (at least across a release!).
> I understand that, and I don't think it's too much to ask :)
> But the EditNotes for these changes say they were done as batch edits,
> so I have assumed they were scripted somehow, not done manually. That
> would explain the Release-internal inconsistency you've pointed out.
> (I've come across another as well, where it could seem like a track was
> not recognised well enough to be mapped into what CSGS/M has)
> (These are my speculations. brianfreud can tell us the whole story if he
> wants. And it's worth noting that these batch edits were limited to
> Mozart. He's also the one to ask about KV 412 (the first Horn Concerto)
> as from all the stuff in CSGS/M about it, it seems quite complex.)
Yes, I would agree internal consistency is a good thing. However, as
I've mentioned often, there were some tracks which didn't have
sufficient enough information extant for me to identify which they
actually were. I put these aside until I could have time to properly
identify which title was correct. (For the record, at the time, it
was 1,596 out of just under 20,000). I did them all by hand however,
not via bot - the "batch" indication in the notes was simply an
indication of the edit processing batch, using the tool I wrote a long
time back for batch edit submissions. The edits themselves, however,
were all manually identified.
> I get that
> > copying track titles from CSGS would mean that the track titles may not
> > be consistent, but if that's the case, then I'd argue that CSGS isn't ready.
How is a lack of internal consistency for a release an argument
against CSGS in general? Wouldn't it rather point to those tracks
which, from the start, were insufficiently identified?
> Well, a couple of other side effects of CSGS has been brought to the
> attention already: Inadvertent translating of Mozart track lists from
> French to English (and leaving the RelLang attribute still at French),
> and more than one editor has been picking up non-CSG stuff from CSGS/Bach...
> Another Mozart AddRelease edit a week ago or so by a brand new editor
> had the comment
> "Track names from CSGStandards. Tracks length from IClassis. Doubts
> about naming track 24"
> Track 24 had:
> Minuet for Keyboard in G major, K. 1e / 1 (Minuet) "Notenbuch f??r
> Nannerl Mozart" No. 62: Minuet - Minuet for Keyboard in C major, K. 1f /
> 1 (Trio) "Notenbuch f??r Nannerl Mozart" No. 63: Minuet
> (See http://musicbrainz.org/show/edit/?editid=8351638
> please go over there to give the editor some more support)
Actually owning the release that editor is adding, I would suggest he
properly identified the track. My concerns there were utterly
unrelated to CSG or CSGS, in that he used a hyphen, not the proper
slash, to separate 2 works, as well as titling the release with a
missing space and a misspelled "dis" for "disc". I would have to
wonder, were there not a CSGS list for him/her to have used, how much
worse might the release have possibly been titled, and how much
additional time might the classical editors have spent cleaning up the
> So I agree with you: CSGS is *not* ready for prime-time. (I've argued
> other places that it should be kept as a reference, not as mandatory stuff.)
I fail to agree. While CSGS does need an actual official CSG - which
is why I put my Mozart efforts on hold - I think, on the whole, it's
already presenting much better results than a "here's a 2-year-old
concept of CSG, it's not up to date, but try your best to reinvent the
track titles, and please don't get annoyed when the classical editors
pester you to fix various things that weren't even written in that
version of CSG...". At least on the Mozart front, I know of several
otherwise new editors who have been spending a decent amount of time
doing ARs, as they were able to add 10-40 CD box sets rather quickly,
spending their time then on the ARs, and not on re-creating all the
CSG-ized track titles.