Any difference between these two "September Song" works? Or merge?

Looks like there’s two different copies of September Song, and both have quite a bit of recordings attached to them. One is listed as a “version of” the other, instead of being a different arrangement.

Should these be merged? Or is there an actual difference between the two? They’re both used the same way and I don’t see any differences based on the recordings and albums, so I’d say merge… but since its a pretty big merge I’d like to get an extra opinion.


welcome to the forums, Peel~

I’m not familiar with the piece, but looking into the edit history, they both were created around the same time (2010 and 2009, respectively), and there have been several edits by @HibiscusKazeneko and @chabreyflint to move recordings from one to the other (and to some of the arrangements), so they probably know more about the differences between the two. way back (2016), there was talk about merging these two, but that obviously never happened. :wink:

either way, I think there should be an annotation to explain the differences between these two (if there are any), because they do look identical at first glance, at least to me.


Thanks for the research, I was looking to see if this had been talked about before.
I think I’ll merge them and let the votes decide, but first I’ll leave this open for a day or two to see if anything else pops up.

1 Like

Sorry, I didn’t notice this post until I entered the merge edit.

1 Like

the works merge
the recordings do not
Walter Huston recorded 1939, then re-recorded 1944


Back to topic: Seen from the classical (score-based) point of view, every new arrangement of a score makes a new Work Style / Classical / Works - MusicBrainz, (Arrangement Works). Since it is often difficult to track the exact version for a given arrangement, the workaround is to create a Work with the disambiguation “catch-all for unknown versions or arrangements” Style / Classical / Works - MusicBrainz (Different versions). Thus, having two different Works is clearly in accordance with our guidelines and would ask for keeping these separate.

Remains the problem when this “classical” guidelines should apply and whether this particular Work “Knickerbocker Holiday” should follow these guidelines. This is not always easy to decide, particularly for Composers whose Works are at least partially written in a “classical” tradition (Kurt Weill, Leonard Bernstein, George Gershwin, to name just a few). So far “Knickerbocker Holiday” has been edited consistently following the classical style (with the overall title preceding the Song title) Musical “Knickerbocker Holiday” - MusicBrainz. For other theatre Works Weill composed in America, mostly for Broadway, we have a mix of classical and theatre style (with the overall Work in the disambiguation) Work “Lost in the Stars, Act I” - MusicBrainz

Unfortunately the guidelines for theatre Style / Specific types of releases / Theatre - MusicBrainz do not mention Works, but just say “Depending on the release, it may be more appropriate to instead apply classical style (as used for opera)”. Whatever we decide to do, these inconsistencies should then be normalised.

See also my comment on Edit #87474608 - MusicBrainz

1 Like

I am a bit lost with all the variations of the Work in Classical. All I ask is to make it clear for us non-classical editors. Sometimes we have a track like this on a compilation or soundtrack we want to link to a work, so it helps if something is in the annotation or disambig to make it clear which we use.

In this example I am voting yes to the merge as it seems to have overlapping artists on both works. Nothing seems to make them unique from each other. I understand where there is a different arrangement noted, but don’t see that here.