ZZ Top

That looks promising. Nice find.

Since nobody protested, I've added the Billy Gibbons & Co. release. I'm curious to see how it goes.

Hello, ZZ Top!

I really like the collab rule as it is. It'd be hideous if you let in all these long lists of performers in all their different permutations from the many tribute albums out there. Fair enough if they have the mindset to give their collaboration a band name, but you've got to draw the line somewhere. The next step down the slippery slope might be where bands release videos or recordings of auditions or live jams. Every time someone plays with a band but doesn't join them, does that give you a whole new band?

I'd prefer not to see every major rock band being connected to 3 or 4 of these, especially when the entity's entire output is one track:
[www.bandtoband.com] http://www.bandtoband.com/index.php?PageField=Wiki&ActionField=var&Album...
[www.bandtoband.com] http://www.bandtoband.com/index.php?PageField=Wiki&ActionField=var&Album...

For what it's worth; without being able to find a rule that bars it, I hate the way ZZ Top have got in. I would have (I think) preferred that we widened the net of '& Friends' to bar '& Co' etc. rather than loosened it completely.

Generally I agree with Bloopy and misterpomp, I think "& Co." is equivalent to "& Friends," especially in a context such as this one.

(However, as much as I personally dislike Bloopy's other examples, I would allow those, because each person is billed equally.)

Err, I was saying I like the rules as they are, with Billy Gibbons & Co. being a valid band name. The thread where the & Friends rule was thrown out was convincing to me.

It could be argued that Metallic Attack and many other tributes fail rule 6a, because there's no collaboration keyword like And/&, merely commas. So the line is drawn there at least.

Rule 6a isn't worded very well, since we've got plenty of examples of collabs lacking a "collaboration keyword". Any connector will do, even punctuation (and when punctuation is lacking, we use a center dot).

But all of that is irrelevant to the Metallic Attack type situation. Those aren't collabs by our rules, since the parties involved don't all have prior independent releases (Rule 6b). In other words, they're bands... even if the bands have ungainly names. And God knows ungainliness hasn't stopped us before, e.g.:

[www.bandtoband.com] http://www.bandtoband.com/band/julie-tippetts-ian-maidman-annie-whitehead-didier...

The Shoppe discography is on Facebook now. Lineups given with each picture if anyone wants to tackle them:

[www.facebook.com] https://www.facebook.com/media/set/?set=a.214291535302858.55348.130006833731329&...

Their new album is in queue, I think we already have the drummer in the db.

Queued up some albums from The Shoppe including the 1986 s/t one. Turns out it makes a connection to band Little Texas, so there's your nicer link for ZZ Top.

I would like to note that I strongly support breaking connectivity for collaborations. If it cleaves off huge parts of the DB, boo hoo.

The point of this site is not to document all recorded output. That's discogs.com, allmusic.com, etc. The point of this site is to link _bands_. A collab is rarely a band.

Furthermore, when reality and the ability to write clear rules conflict, reality should always win. If previously in this thread people are all agreeing that some group "isn't" a band in the same sense as others, and merely disagreeing over whether the rules allow it, then the answer is clear: we shouldn't allow it. Rules (including what I wrote above about collabs!) should be treated as guidelines that tell you what the right answer typically is, and how to bias your thinking in the face of ambiguity; but the final answer is always a judgment call.

We've evolved rules as complicated as the geocentric astronomers' to work around complexity, instead of just sticking with simple guidelines. And the simplest one that cleaves off a lot of silliness is to ban connectivity through collaboration.

The rule on connective collabs is anything but complicated. Neil Young & Crazy Horse is a band, Slayer & Ice-T is not. That's basically it.

I've come up against "judgement calls" and it's awful. Like it or not, the aforementioned rule allows for eg. 8 jazz musicians with established solo careers to collaborate with a 9th newcomer and allow that newcomer to make connections. There are dozens of such entries in the live DB. So I entered a pop collaboration with the same measurements, but it got frowned upon and left to languish in my queue with no basis in the rules as to why. So why did I bother researching it in the first place? Just to be pissed off for time eternal, apparently.

I think a goal of the site should be to document as much recorded output as possible, so long as it's in the spirit of B2B. Things like jazz collaborators linking to jazz collaborators or musical casts linking to musical casts are in the spirit of bands linking to bands.

A better solution than deleting at this point would be to allow all collabs, but then provide site visitors with a way to exclude collabs when searching. But either way, if you say Neil Young & Crazy Horse is a band, but Paul Dunmall · Paul Rogers is a collab, I'll still demand a precise rule to get that result every time. I just cannot see judgement calls ever delivering consistent results.

I'm much less worried about being inconsistent than categorically allowing things that are clearly not "people being in a band together" because we can't write a "consistent" rule. One gets it wrong sometimes, the other gets it wrong all the time.

I will continue to disagree that documenting as much recorded output as possible is a desirable goal. That biases us towards continually bending the rules to allow more and more things in. If jazz musicians don't tend to play in "bands", then I'm OK with them mostly not ever being represented in the DB. Making music together with someone is not the same as being in a band with someone.

Otherwise, if everyone else thinks I'm wrong, then I think we should put session musicians on albums, because at this point what we're documenting is shared musical creation, not shared band membership; and session musicians participate in the former. Note that those are both perfectly valid things to document, so this isn't about saying that we're "devolving" from one thing to another. But the focus is very different, and we should own that, not straddle this weird line where we're trying to rope broadway casts, collaborating jazz artists, and bands into the same tent but exclude other musicians.

I'm not convinced we need to be changing anything and I don't think we should get hung up on the word "band". For the purposes of the site I take "band" to mean "the musical entity credited with the release of a record", and if that means it's two or three solo artists collaborating or a Broadway cast, so be it.

The rules are clear on why we don't recognise session musicians - they weren't part of the named entity that recorded and released the music. E.g. Spike Edney played on various Queen records but was never a member of Queen; Paul Dunmall and Paul Rogers are very clearly the two members of the musical entity called "Paul Dunmall · Paul Rogers".

You'd presumably argue that "Paul Dunmall · Paul Rogers" isn't a band, but why not? If they'd released a record as "Dunmall & Rogers" or "The Dunmall-Rogers Duo" there would be no argument. What's the difference?

The difference is that "Paul Dunmall" and "Paul Rogers" are already established as recording entities. If, when they combine, they see fit to maintain both those entities and simply join them with a joining word or a joining symbol, it's a logical position to say they haven't created a new entity whereas if they use a different styling, they have created something novel. It's not the only possible position but it is logical.

Would it maintain logic but resolve most of this problem if your rule was to be flipped and instead of "as long as any one member of the collaboration has not released anything under that name, it's a group" to "as long as no members of the collaboration have released anything under that name it's a group"?

Also, I think you could limit that loophole to the first instance.