[x] Did we ever decide?

What happens when a band is recording as a fully formed band with (it's usually) a drummer who gets replaced mid-recording and whose parts then get re-done entirely by a new drummer while further recording takes place. The recording takes place over several years with both the first and second drummers in place for a long period of over a year.

Do we have line-ups:

a) 2 line-ups including both drummers even though one isn't on the finished product?
b) 1 line-up and disregard the first set of recordings even though we believe that music from those recordings is on the final output?
c) 2 line-ups excluding the first drummer even though there was never a partially formed line-up that excluded a drummer?

I know we've discussed before - but every time it comes up I can't get clear the answer we've arrived at. I believe it's c) [see recently-added Boston extra line-up for Third Stage]

hmm i don't think i'm particularly consistent but would tend to go with (b). (c) would imply that any given single track would have multiple lineups and i don't much like the idea that one song could have multiple lineups, but that's just me. i think (b) is a little simpler.

Apologies in advance - I know we're revisiting old ground here but we have no barrier to two line-ups on 1 track.
[www.bandtoband.com] http://www.bandtoband.com/index.php?Page=Search&AlbumId=7579

I'm not sure we should have such a stance in any case - that would lead to 'What's a song?' discussions when you get a concept album or segued piece of music.

We've previously discussed and agreed upon the concept of "ghost members" when a member participates in a recording session and some of the material featuring that member is actually released. This approach is meant to make sure that we don't present unnecessary partial lineups in situations in which a song that is missing a band member (perhaps an instrumental) appears outside the context of the rest of the recording session. An example might be Van Halen's "Eruption" (which lacks David Lee Roth) appearing on a various artists compilation. Would we list a DLR-less lineup for that release? No.

When a member is fully removed from the entire recording session or failed to participate at all, then there's simply no appearance to document anywhere. In that case, we do not list ghost members. This will occasionally result in transition lineups. One example might be Fastway's first album, for which the bass tracks were re-recorded after the co-founder left the band:
[www.bandtoband.com] http://www.bandtoband.com/index.php?Page=Search&AlbumId=1340

It sounds like this second category is what misterpomp is talking about here. Therefore, I think the answer is b or c depending on the circumstances.

That Whitesnake single is the least appealing example of multiple lineups on a single song. Two other examples are these:

[www.bandtoband.com] http://www.bandtoband.com/index.php?Page=Search&AlbumId=32761
and
[www.bandtoband.com] http://www.bandtoband.com/index.php?Page=Search&AlbumId=37824

Here is an example of a b circumstance:
[www.bandtoband.com] http://www.bandtoband.com/index.php?Page=Wiki&ActionField=edit&AlbumIdFi...

That album featured three new Van Halen songs begun in a transition period when the band had no bassist. Long-time bassist Michael Anthony re-joined the band, thereby restoring it to a quartet, and added vocals (but not bass) to the songs. We decided that the transitional three-person lineup was not meant to be documented.

Whenever possible, I lean in favor of b because I think it's a cleaner representation of band history (and perhaps one that the bands themselves would prefer). But that's not always possible or even desirable.

I suppose another approach, not recently suggested but in place in the early days when we could display only one lineup per release, would be to simply ignore incomplete/transitional lineups. That would clean up a lot of entries, including all of these:

[www.bandtoband.com] http://www.bandtoband.com/index.php?Page=Search&AlbumId=38483
[www.bandtoband.com] http://www.bandtoband.com/index.php?Page=Search&AlbumId=1308
[www.bandtoband.com] http://www.bandtoband.com/index.php?Page=Search&AlbumId=7579

For me the difference is that someone 'in the band' DID play bass on those tracks. That can be extrapolated to say that the band was not, therefore, fully-formed at the time and so we would be documenting what were merely works in progress when the need to complete the band with the recruitment of a fourth member is obvious. It's still a fine judgement call IMO, but that's how I could rationalise ignoring the 3 man line-up.

However, when a fully-formed band makes recordings and then continues to record with one member changed but re-records all of the prior parts of that departed member, I think b) is only a cleaner representation because we haven't got the functionality yet that allows us to simultaneously recognise the line-ups that existed AND the subsequent erasure of a single player's music. I favour c).

Another benefit of going with c) in the situation that you described at the top of this thread is that doing so at least hints that something is amiss (no pun intended). Even if we do not document the removed member, we at least document the existence of the removed/missing member's lineup.