I realize that this is only one person's experience, but his description of what is 'wrong' with Pandora happens to be exactly why I like it so much. His argum

I realize that this is only one person's experience, but his description of what is 'wrong' with Pandora happens to be exactly why I like it so much. His argument is that "Pandora didn’t solve the real problem – the lack of a human DJ" and yet this is exactly what Pandora does for me better than any other music service I've tried - it picks out a playlist of amazing songs that I've never heard of, but which 99% of the time I like. On top of this, I constantly get compliments from my friends on how cool my music is, though it's hard for me to take any credit -- all I do is press the thumbs up button in my Pandora app. To me this is also a feature -- I don't have to spend hours scouring music stores (real or virtual) for cool music, and then spend more hours constructing the perfect playlists; Pandora does this all for me, better than I ever could, in less time, and for free! (I actually pay for the ad-free service). My FM radio experience consisted primarily of Z100 in the mid-late 90's, so maybe I never really got to experience a 'truly great DJ' in comparison to whom Pandora would seem like a pale imitation, but the fact is that those DJ's can only ever be the exception to the rule. However I do think he's right that the other streaming music services have little to offer (notice how he leaves Pandora out of that list), they have the basically the same music and basically the same service offering. And they do precious little to actually improve circumstances for the music artists themselves. In contrast, I think Pandora probably does the most of any of these services for the artists themselves, despite the author's argument that it's profit model is primarily about advertising and branding. The key here is the technology -- the Music Genome Project, which the author dismisses right at the start, really does improve discoverability for musicians, as it analyzes and recommends music (and hence artists) not based on whether the artists are or are not subjectively cool-and-trendy (which, let's face it, is how the mainstream music business works), but rather on the actual qualities of the music, and it can then compare that to an individual listener's historical music preferences. If this were the primary mechanism of the music industry (and on the merits, why should it not be?) I imagine it would really free artists up to spend as much time as possible on making the coolest and most awesome music they can come up with, and save them a whole lot of money on things like PR, etc., which has little to do with music itself.