I'm experienced enough to know that statements like "There's an excellent chance I will never discover artists like Steve Harley again" are (i) too topical to be of any interest to you, dear reader, and (ii) very likely of no lasting power beyond, say, a few months. (Music writing is especially difficult that way, and to think otherwise is to ignore history.) But with things being what they are, I'll swallow the shame and forge ahead. (As for boring you, well, that is not exactly without precedent.)
Anyhow, as for Steve Harley, my introduction to his music was through that ancient but ever trustworthy guide, Starostin, and I really don't see his kind surfacing again. Not with the explosion of access to music and the perceived knowledge about the subject that that instils in otherwise well-meaning listeners. I've speculated about what I find lacking in many of the new guard, and I think it stems from a basic difficulty when dealing with popular music, which is the culture of judging things based on novelty, rather than resonance. It's easy to get recommendations for obscure artists from years past, but ones who also happen to be good are a trickier ask. Harley is a good example of someone who's unquestionably weird, but only slightly less unquestionably good. I can imagine him being summarized by any of the new guard to sound like just your typical weirdo, and consequently lamenting how many other artists there are who face the same fate.
Even aside from Starostin's rigor, the old fan sites at least made it clear why they liked the music they did. I miss them, not just for nostalgia's sake, but because there was an obvious sense of love and attention to detail (okay, maybe it was often obsessive). I don't sense the emotion amongst the few sites I read nowadays, and so it is all the more harder to project one's own musical makeup onto the review and infer if there is likely to be a match. To say nothing of the sense these sites imparted of an album being part of a larger story, be it an artistic or emotional movement. You don't get that with fragmented one-off reviews, no matter how many of them are accumulated.