here's my anecdote.

i went in for a sleep apnea test a couple years ago based on a hunch my mom had. they wired me up and put me to bed. when i woke up, they showed me all their reports about how my sleep apnea index was off the charts. they gave me a cpap machine (charged to my insurance) and off i went, and i plugged in every night for maybe a year or two.

but eventually i stopped. why? it just seemed like it was more hassle than it was worth. it's not that i doubt what the doctors saw in the report, and nor do I doubt that sleep apnea is a serious condition that some people have... it's that i began to doubt it was as serious as they said it was for me. they do, after all, have incentive to make a big deal about it, in just the same way as the news media keeps itself in business. and people typically go along with it, partly because of the placebo effect.

i'd never heard of "restless leg syndrome" before now, and i have no comment on the diagnosis that you received. but i do not believe that just because you can identify, name, measure, and write a prescription, that it's necessarily serious.

clearly, the issue is connecting the band with the musician, which means not just publicity, but targeted publicity. a genre-specific top 100 list can help because, for example, i can see what jazz music other people seem to like.

i think you're right about the effect of a top100 list making the field smaller, and there were two recent slashdot articles that agree with you. people are discovering that the "long tail" theory of marketing is coming under fire... i guess the good book had it right:

but actually, i'm not sure there's too much to worry about in the long term because... drum roll... this problem will simply create a new service market where profits can be made. some people will be happy with the top 100 hits, but most will eventually want to go deeper, and people or algorithms will need to be in place to help them connect. there are three examples of companies that are exploring this:

• Amazon: for whatever item you're looking at, they'll tell you what other customers bought.
• Netflix: part of what makes netflix successful is not JUST that they deliver DVDs to you, but that they help you explore. Their algorithm is sophisticated enough that they even predict what I would rate a movie to be, and so far, it does a pretty good job.
• iTunes Music Store is again more than a store, it's a service. With their Genius feature, they make it really easy to find music that's similar to a selected song. I've used this to make several discoveries in genres that i know nothing about.

there's certainly a lot of algorithmic work to be done, but i think people will figure it out eventually. that doesn't necessarily mean your music will make it to a best seller list, but it will help you maximize your audience size to what the music is "worth".

Perhaps ironically since now i do computation, what made the most difference in my life as far as getting me interested in science was:

http://en.wikipedia.org/wiki/The_Anarchist_Cookbook

which i could find online. i didn't care about hacking phones or whatever. but i used this to help me understand all the chemicals i could get my hands on at the local drug stores. fortunately, i couldn't hurt myself too much, since where I lived it is illegal to sell nitric acid to people without a license or something. however, i was able to get my hands on or make two highly exciting substances:

http://en.wikipedia.org/wiki/Calcium_carbide
http://en.wikipedia.org/wiki/Nitrogen_triiodide
sodium metal

i know. it's an advertising model that's still being developed, so i can't fault them too much for problems along the way.

however. i think it would be better if they skipped commercial breaks with dupes.

It's not just CBS. I like to watch The Daily Show and Colbert Nation online, and they raise the volume significantly too. They still do it today. It really pisses me off because it's outside of my comfort level, and often times I *have* to mute it.

The other thing I hate are repeated ads. It's super annoying that there are only like 2 ads in circulation at any time, and oftentimes I watch the same ad or from the same company for all 4 "commercial breaks".

I don't understand how the advertisers approve of this, because repetition not only has diminishing returns, it's likely to have negative returns because I begin to associate frustration with their ad.

It's a cross between instant messaging and asynchronous voip.

http://voicebeep.com/sayit

you can talk about it once the case is finished.

that's not always true. i just served jury duty where the defendant was accused of using tracking devices and various technologies to be sure his daughter wouldn't talk about how she was being used as his sex slave.

i'm not saying this makes google's stuff bad, but certainly there are good and bad uses for technology.

I agree with you. I think Songsmith is a neat idea, even if it does seem like that book machine from Orwell's 1984.

I once knew a musician who would invite someone from his audience to sing 3 random notes, and then he would compose a song for them on the spot. Presumably it worked because 3 is enough to specify a major or minor key.

My reason for thinking this is good is because it raises the bar for "professional" musicians. Too many of them sound exactly the same, so if what they do can be generated in an algorithm, then this forces them to branch into new directions.

I am not a volcanologist, but my experience suggests this "microscopic shards of glass" is an overreaction. This article says that Mt Pinatubo released 4 cubic kilometers (= 1 cubic mile) of ash (= < 2 mm tephra). We were within 25 miles of it (we drove from Baguio to Manila that day), while it was blowing up, and nobody complained about breathing it in, and I never heard any discussion of respiratory problems... Actually, I remember standing outside watching the stuff fall down and asking my parents if it was snow. Amusement turned to concern only when it began sticking to our windshield (it was wet), forcing our driver to keep his head out the window, with sunglasses and an umbrella. On that day, we crossed bridges that collapsed shortly afterwards.

Granted, 1000 cubic miles is a lot more, but it will probably spread over a greater area too. My message is sure, get dust masks. But worry more about the roof of your house collapsing. Maybe bridges too. And, regarding one of those articles about water scarcity, I think a coffee filter should strain out the ash, if it came to that.

Come to think of it, people are able to survive dust storms just fine, which on a local scale, are probably worse than what a volcano would kick up. So maybe we should just ask the bedouins what to do.

I was near Mt Pinatubo when it blew. Ash was falling on us, but we weren't concerned about inhaling it... If that were a problem, you could just use an umbrella...

The problem was that the ash mixed with a typhoon passing by, and raining wet ash is a problem because it turned into dangerous sand-rivers. And, it ruined crop fields. In all, it hardly seemed like a big deal.

Certainly this volcano could be bigger, but I doubt it would make a great movie.

yea, well you have the same problem (if it's a problem) the portuguese & spanish have. we "colonials" outnumber you by a lot, so if there's ever a vote...

the "academic route" is probably not going to pay off since they're already busy with their own academics, especially if it is a "good CS" program.

most good scientific software projects that I know about were written by the scientists themselves who wanted to use that software. seriously, as a computational chemist, i know of almost no software comes out of companies. a company might take over and sell academic software, but there are very few projects that i'm aware of that *started* outside of a research group.

a better route might be, if you really believe that this is useful, is to DO IT YOURSELF. you're going to be wanting to add new features anyway, so you'll want to be able to modify it yourself. starting from scratch yourself isn't so difficult these days. just go with java, find a good IDE, load up one of their template projects, and you're 10% done already.

if you need to, submit applications for federal/state grants, and you'll possibly even get a publication or 2 out of the project, AND increased name recognition. when i was a freshman, i worked as a programmer for some faculty who went this route.