John's comments on this phenomenon sparked a blogversation between John, Glenn, Dave, and now me. And maybe others. (John even graciously invited me into the discussion.)

The blogversation is essentially about the question of whether this is how blogs will help digitial content get discovered.

My personal take on it: yes and no. Yes, blogs or blog-like sites will be a big part of how music is discovered. But no, the serendipity of "I saw John's mention of Anil's mention of Kevin's music" is not how music will principally be discovered IMHO.

Rather, I like imagine a world in which people play the roles that radio station "disk jockies" like Wolfman Jack or Scott Muni or Pete Fornatelle (the latter two at WNEW-FM) played when I was a kid. You would tune into the station, and be introduced to all sorts of great music. And they had interesting personalities too and injected their own personal stories into the proceedings.

Radio stations, except for some college stations and independents, have become so dominated by playlists and other formulas for control that they are worthless for this now. On the Internet side, collaborative filtering mechanisms such as those used by Amazon provide recommendations but there are no personalities behind them. You can see bios of people who review things on Amazon but those are very limited and are not "engaging" in any sense.

I think the Internet will enable those old-time radio DJ's to be replaced by individuals who are doing something a lot like blogging and who supply music recommendations and/or audio streams as part of what they do. If the "blog posts" were audio, then they would be doing exactly the same thing those DJ's did. If written, they are blogging.

And blog comments and blogrolling lists will enable communities to form around each of these blogs (an improvement over the old model).

The really big question is: how do these emerging DJ's get known to their potential audiences, so that they can make hitherto unknown music known to its potential audiences?

One could say, the same way blogs get known today. Other bloggers link to them; hits are counted and displayed and sites are ranked by that popularity measure; etc..

And those mechanisms will certainly be a part of it.

But I think technology will take us a lot farther by using recommendation engine technology, such as Amazon uses for albums, to provide personalized recommendations of blogs. Imagine Amazon without it's "People who bought this also bought..." and other recommendation features. It's hard because Amazon users are so used to those features. But that is how the blog world is today. It's lacking a recommendation engine.

A recommendation engine can say, in effect: Out of the music-related blogs that are focused on the same general kinds of music that you like, these are the ones that people have found most reliable for music suggestions.

Now, it is obvious that recommendation engines in the blog world don't only have to be about music. They can be about news or technology explanations or any other subject. The common thread is that they will help us find the blogs we will most benefit from reading.

The blog world needs a real recommendation engine.

Note 1: If we had a feature on EM that supported blogs for our users, we would have a kind of music-blog recommendation engine like what I imagine above. EM figures out who is really reliable at recommending music that a lot of people like, and doing so before others have discovered it. If such people had blogs, and were good writers, they would have an opportunity to play a role in people's music lives like the one Scott Muni played in mine. If you have ideas about what would be the best way to set something like that up, let me know. Should we do it in-house, or work with an established blog company?
3:30:20 PM    

Deborah noted that not detecting a particular smell or taste may kill you, whereas mixing colors together is unlikely to. So from an evolutionary point of view, our brains should handle those sensory inputs just as they do.

But I find it interesting to try to imagine what it would be like to look at a green object that is really composed of yellow and green and to be able to "see" that fact. I wonder if it's even possible to imagine what it would actually be like; it's imagining a nonexistent sense.
2:17:51 PM