Merging classical Recordings: basis for sufficient confidence?

sufficient-confidence
confidence
Tags: #<Tag:0x00007f309a4c1270> #<Tag:0x00007f309a4c1130>

#1

I’m seeing classical Recordings being merged on the basis of:
Same performer.
Same composer.
Similar time.

To me this is an inadequate basis for merging classical recordings by the same artist.

My expectation is that many different Recordings by the one Artist of the same Work are going to have similar lengths.
Or to put it differently:
Same performer, same composer, & similar time are what I’d expect of different Recordings.

Additional information about the Recordings seems necessary before we can have confidence Merging.
Information like:
Same AcouticIDs,
Millisecond time matches.
Documents showing that each recording was made at same short-term recording session.
Documents showing that each recording was made on the same day or at least in the same short space of time.
Recordings being within same Release Group.
Documents showing the Artist has only recorded that Work once.
Any other documents that show the 2 Recordings were actually one Recording.

I’m not after proof here.
Just a reasonable basis for confidence.

(I’m not seeking to stop this particular batch of merges as, from memory, they are just a drop in the bucket of merged classical Recordings. So this post is not intended to say that the Merging editor is not acting in line with existing or current community standards.
Rather I am seeking to create discussion around community standards and perhaps changes in community standards.)

(If this reads like the post of someone who has been interacting with a deliberately obstructive and wilfully obtuse Australian aged care provider company about the inhumane mis-treatment of a relative there is a very good reason for that.)


#2

In this case, it seems it’s also the same label (EMI) for one case and the entire same tracklist for the other, so I would probably have considered merging too. In fact


and

look like they might be merged as releases as well, or at least be a re-issue + original.

Generally, I’d say it depends on a few more things. I’d be happy merging on performer, work and time for obscure works - ones that I expect have only been recorded once by this performer, or even once ever. For something like a Beethoven symphony or sonata, I’d generally try to find the date before merging, but I’d be more confident if (like in this case) the two recordings were in releases with the entire same tracklist (or in a boxset that contains the entire same tracklist as one of its discs).

(And sorry about your relative, hopefully you can get the situation to improve!)


#3

Hi!
I don’t agree with the milisecond time matches: when a recording is integrated into a compilation or a reedition of the disk is make, the recording times can change. An additional problem is that the track times in the booklets can be seriously off (by up to a few seconds) so even if someone carefully copies the track times from amazon/discogs/booklet they can be wrong. Also, acoustids are tricky, as discussed in How reliable is acoustID for merging recordings? (By the way, I’m wondering if it can be a sufficient basis for merging recordings with matching track times, works and where one with [unknown] artist) . I feel concerned because I made a large set of merges for Bach works, and bringing it up here is a good idea I think.

I looked at a few works with a lot of recordings, checked for identical (or mutually containing) recording dates, artists and acoustIDs.
Then I looked at recordings by the same artists and with similar times (±5 seconds), the labels corresponds (on the back cover the phonographic copyrights help also since labels usually pass recordings around), the tracklist were similar enough (several tracks with very close track times in a row), not speaking about additional data like discogs that can most importantly provide the recording dates.
I must admit that I have much less attention to detail when it comes to Alfred Scholz’ 1001 pseudonyms, there I generally do not look further than artist and timing.
For compilations, I tend to merge it with the recording with the closest track times, if there is one (and only one) with close enough track times.
I wonder if there is a set of guidelines specifically for merging classical works, since I believe that there are not that many recordings out there by a given artist, even for famous works, one just needs some merging.

(also, I’m sorry for your relative, that kind of stuff can be pretty soul-crushing)


#4

I think I agree with everyone :slight_smile:

Same performer/work/duration seems a bit short as a justification since we want to avoid hard-to-revert merges.
Usually I’m also looking for:

  • same label
  • same date (it’s the case here)
  • same acoustID or AcousticBrainz signature

I am also much more confident with less information if several successive tracks are merged (it’s the case here) and all durations correspond (with the last track possibly different in live recordings due to applause being faded out differently in remasterings)

I also often refer to a short list a discography websites to check whether a performer recorded the same work several times: here Landowska recorded BWV 903 twice apparently, in 1935-36 and in 1948 (“2nd recording”) according to bach-cantatas.com

I just copied the list of sites I use frequently on the wiki: https://wiki.musicbrainz.org/User:Loujin


#5

hi dosoe, I’ve just left a comment on one of the Merges. before I realised that you’d commented here.

Millisecond timings are, as you say, not good evidence that Recordings are actually different Recordings.
But if they match then I’d see them as adding a lot to the confidence that the Recordings are the same.

From what you write (and from what others have written here) you are doing the work necessary to have sufficient confidence around non-Scholz-ey Recordings.
I know it is a pain but if you could include something in your well-presented edit comment to document that extra knowledge you have then your much appreciated Merge edits would stand as presented.
Would something like “Match in at least one of; Label, recording date/s, track sequences and/or sole occurrence” make it clear to beginner editors that Merges are being done on the basis of enough evidence?

(thanks for the commiseration. Ay yi yi. The cognitive tilt towards thinking that it can’t really have happened like that does win out frequently. But I’m off to add another document to the paper trail anyway.)


#6

I will try to be more clear in future merges.
Also, a question: How to handle Alfred Scholz? Hiw pseudonyms make it significantly to spot duplicate recordings and when they are merged, which pseudonym choose?


#7

I discovered yesterday guidelines on the wiki written by @monxton and @Hawke. They look reasonable, although I would use Scholz as recording artist instead of special artist.


#8

These guidelines don’t answer my question: If I merge several scholz-recordings with different pseudonyms, which pseudonym should I use? And yes, there are scholz recordings that exist with different pseudonyms for the same recording, that’s the whole point of making these pseudonyms.


#9

Well firstly, they aren’t actually guidelines in the sense of being endorsed by the powers-that-be. They are just what I thought at the time of writing. I have in the past (I mean, the past before these forums) tried to engage the community on this topic but never got much response.

Having said that, my recommendations do answer your question. They say that what you should do is:

  • preserve every artist in the ARs
  • if there is only one credited set of Scholz artists, use them in the ACs
  • if there are no credited artists, then use a special purpose artist. I did create [unspecified artist on Alfred Scholz compilation] for this, but then I didn’t use it much (!)
  • if there is more than one credited set of Scholz artists, then use special purpose artists in the ACs - ex:[unknown Alfred Scholz conductor]. I am almost sure that I did create a set of special purpose artists for this, but if so they have been merged away (or atrophied after been replaced by specific Scholz pseudonyms in the credits), as they no longer exist.