this blog is girtby.net

Music Insurance

Broadcast radio is changing. It’s going digital, and with this change broadcasters are likely to start embedding watermarks in the audio stream (if they haven’t already). Podcasters, and other internet music publishers are likely to do the same. Watermarks are inaudible markers that uniquely identify the broadcaster, and are also the key enabling technology of a great new business venture that I’ve just thought of, and will now share with you, in the hope that someone, if not me, implements it.

The elevator pitch: music insurance. Insurance against hearing music that you hate.

Here’s how it works.

Let’s say you wake up one day with a revelation. “I would,” you think to yourself, “pay good money to never have to hear Meat Loaf’s Paradise By The Dashboard Light ever again. I would pay until the end of time, should it hurry up and arrive.” Or words to that effect.

Pick whatever example you want here: maybe the Crazy Frog song might be your choice instead? Or Whitney Houston’s I Will Always Love You? I-I-I-I-I-I-eee-I-I-eee-I-I don’t wont to go on, I’m sure everyone has at least one song that is disliked to the point of justifying financial outlay. (For me it would be the majority of the AOR canon, which could get expensive, but I’m getting ahead of myself.)

So you take out insurance against accidentally hearing this song, whatever it might be. Like most insurance policies you pay a premium, and when you hear one of the covered songs, you make a claim and receive some financial compensation to ease the pain. All you have to do is prove that the music was broadcasted publicly and you couldn’t escape it. This guards against fraudulent claims, and is where technology comes in.

We need to make two simplifying assumptions. Firstly we require the music to be broadcast by a radio station or similar public broadcaster, who are motivated to watermark their broadcasts in the interest of asserting their license to do so. This precludes the private playing of insured music; in other words, you’re not allowed to play Celine Dion CDs to yourself and claim on them.

Secondly we require the insured to be in a public place, or specifically nominated private places, for the claim to be valid. This might be at work, at a pub, restaurant, or wherever you have no control over what radio station is played. This precludes you claiming against family members who play music too loud, but I’m sorry no insurer is going to go anywhere near that sort of dispute.

When you sign up for the policy you will download a small application to install on your mobile phone. When you hear an offensive song, you whip out your phone and start the application. This records a sample of the music, the date and time, and the place (either manually entered or from the GPS receiver). This information is transmitted (via internet, SMS, or whatever) to the insurer’s servers. These servers will automatically extract the music sample, match it against the sender’s insured songs, verify its authenticity using the watermark and location data. The watermark is verified against a known list of broadcasters, and the location data against a set of covered locations. If after this, the claim is determined to be valid, the insured amount would be paid.

Of course there is still a potential for fraudulent claims, but — without meaning to oversimplify the actuarial art — this can be countered by conventional risk management techniques. Claiming against Achy Breaky Heart when you’re at a Billy Ray Cyrus concert is obviously fraudulent, for example. And as with all insurance policies, the premiums and claim amounts can be adjusted in relation to specific risk factors, including the potential for fraud in a given situation. I am not an actuary, but feel free to chime in if you are (or can impersonate one on the internet).

Now I know it’s a great idea, but if you make millions from selling musical insurance policies, I don’t even want a cut. I just want you to cover me. Come on in and cover me.

Damn it, 2 minutes after publishing, I read Brendon’s post and realised that animatronic mice form a crucial loophole in my proposal. Maybe protection against animatronic mice can be arranged under supplemental cover?

Pretty weird to come by from my RSS feed and see my name already in the comments. Anyway …

I like the thinking you laid out, even if I don’t buy the particular business model. You might, however, make some use of the technology that would need to be developed for related purposes. How about, say, a chip that could be installed in a radio that has the following characteristics:

1. You can flag a song as reprehensible by pushing a button on the radio (okay, we need a chip and a human interface, but one button should do it).

2. The next time a flagged song is broadcast, the radio automatically jumnps to the next one of your preset stations and/or invokes its seek function to go to the next available strong signal.

3. Support for an opt-in feature whereby the act of flagging a song is counted as a vote, uploaded to a database in the cloud, accessible by station programmers. This could, over time, rid the airwaves of Whitney Houston.

As for the animatronic mice, though, I got nothing, although if I’d change my attitude on one hot-button issue here in the US, I’m sure my Second Amendment rights could suggest something.

No, wait. Maybe v 2.0 of TV-B-Gone could be made to work on the rodent robots?

Yes, but then you have to convince the manufacturers to put a feature in that their customers don’t necessarily want. Why would I buy a radio that would allow you to override my choice of music?

And by the way, the rumours that this post was in fact inspired by hearing Meat Loaf’s Paradise By The Dashboard Light are … completely true and entirely based in fact.

I don’t understand why you think my proposed feature allows anyone but the owner of the radio to control what is (not) played.

OIC, you’re thinking about a radio that can learn your taste in music and help to play only music that is compatible. This is certainly a useful product but it addresses a different problem to the one that I’m thinking about with the insurance policy.

I wonder how this one-radio-one-vote idea would change the type of music that radio stations play. I can’t help thinking that the reason we got Whitney in the first place was from a significant proportion of people who genuinely like her music. I’m worried by what might be called the American Idol effect, where the lowest-common-denominator rules.

On an almost related note, I had immense fun at the Spicks & Speck-tacular show last week. Highly recommended if you’re at all interested in music and/or a fan of the TV show. Anyway at the start of the show the presenter Adam Hills asked the crowd to sing the next line to a song. The twist was that it’s a line that isn’t technically part of the song because it was (AFIAK anyway) never actually sung by the band. Australians should instantly recognise it: “Am I ever gonna see your face again?”. The entire theatre to a person erupted with the next line: “No way, get fucked, fuck off”. (Yes, I didn’t say it made much sense did I?) I find it amazing and amusing that this line was known so widely, way outside of the band’s base of fans.

I agree that my proposed product goes about things a different way, but it does seem as though both of our ideas address the problem of unwanted bad music.

On the voting idea: you might be right about this not getting Whitney ouston off the airwaves – no one ever went broke underestimating the bad taste of the American people, and I presume the same goes for your countrymen and women. On the other hand, it does seem to me that a lot of people who listen to stations that still have “and iiiiiii will always luv uuuuuuuuuuu” on heavy rotation tend to be passive listeners. That is, they like the radio for background, and don’t care specifically what’s on, as long as it’s not unfamiliar. Thus, a concerted effort by a concerned minority could have a real effect on particularly objectionable songs.

That’s a hilarious story about the S & S show. I’d love to have seen that.