[skippy the wonder dog]

The problem with M@W is that it's diffuse and unfocused as an album, and it doesn't flow.

I really don't understand the concept of an album's "flow". I simply don't care if track 12 sounds nothing like track 1.

I'm right there with ya. I never understood why "flow" matters, unless the goal is a "concept album". If someone tries to claim that the album is worse because "Wild Mountain Honey" is preceded by "Toronto #4", then perhaps real the issue isn't that the record is "disjunct", but that the critic is obsessive-compulsive.

It's a rock album, not a Broadway musical.

What you're struggling wth is not the product of my obsessive-compulsive disorder, but of your failure to separate the general from the specific, and the objective from the subjective.

People favour the familiar and repetitious in music. That statement is backed up not only by a large body of research (you can start at <!-- m --><a class="postlink" href="http://www.psych.mcgill.ca/labs/levitin/">http://www.psych.mcgill.ca/labs/levitin/</a><!-- m -->), but by the simple and oft-repeated observation that the music industry gets people to buy crap by repeatedly forcing them to listen to it.

That's true of everyone to some degree, but individuals vary in how they respond to music and how actively they listen to it. It's not a stretch to suggest that people who listen to music less actively tend to favour greater familiarity and less change; this is supported by the nature of Muzak and pop-music clone acts, whose listeners surely are the least active listeners around. It's also not a stretch to suggest that the less active listeners are the larger group; again, this is supported by the nature of music that has the broadest appeal.

From that, it should be pretty clear that an album will have broader appeal if changes in tempo, timbre, etc. are gradual and minimal, rather than significant and abrupt. It will have broader appeal because it is more "listenable," that is, less demanding of the listener's attention. And this leads us to the suggestion that a less "listenable" album may sell more poorly, even if the individual songs are appealing, which is the case with M@W.

This can be verified by a simple experiment. You compile two playlists. One playlist groups a set of songs that flows, and the other jumbles them at random. You then subject a large number of volunteer listeners to one or the other of the playlists, and record their impressions -- not of the flow but of (a) the individual songs and (b) the playlist as a whole.

My suggestion is that the listeners of the jumbled playlist will not only rate the playlist lower as a whole, but also rate the individual songs lower.

If I actually do that experiment and post my results, then you can call me obsessive.

To say an album is "bad" is completely subjective. It depends what your yardstick is. If your concern is with whether you like the songs individually, you can call M@W a great album. But in calling it a "bad album," my concern is with why it was a sales failure.