I'm not sure that I agree 100% with everything, but I do think Erik has said a lot of wise things here.
Some thoughts:
1. FreeDB contains many duplicate or near-duplicate entries. It's designed around the ability to have a software program on my computer "magically" recognize what CD I just put into the computer, in order to show the album track names. There are many slight variations on CD's, apparently, so the same "album" (in a sense) will often appear multiple times in their listings.
2. Even if that problem were solved by sticking to unique artist/title combinations, the data is of limited value for our purposes. If there ever did arise a big clamor for that sort of thing, we could probably serve that interest better by setting up a software engine to allow people to easily include FreeDB links into articles. I'm thinking of something like an 'album:' namespace that interfaces realtime into the freedb listings, without actually importing them as wiki-text.
I'm not advising that we do that, I'm just saying that it would be a heck of a lot better than free-texting data that's already formatted by someone else.
- Make another policy that pages about bands should *not* contain
links to the individual albums. These links should be created on demand, not for all album titles.
I think that the first sentence is too strong, but I think the second sentence softens it nicely. Basically, I think contributors should use good judgment here -- some albums are of historical importance and will of course need an entry. Others, well, maybe, maybe not.