Pandora

I recently went back to using Pandora, the internet radio station that selects songs based on your musical preferences. I tried it a few years ago and got a few interesting suggestions, but it was pretty hit and miss.

After a gap of two years or so of not using it, it's gotten way better in the interim. I think it's to do with the way they predict which songs you'd like. Originally, they had a bunch of guys who assigned attributes to the songs (acoustic guitar, major chord instrumentation, 'wispy male vocals' (really) etc.) and these were prominently featured ('this song came up because you seem to like [attribute X]). Coding up vague song descriptions struck me at the time as an incredibly labour-intensive way of doing it, but they seemed to be giving it a go, so good for them.The Pandora About Page still describes this process.

While this is just a hunch, what I suspect they actually do now is use the much richer data set that their users provide when they select which songs they like and dislike, and which songs they skip. This is outsourced to thousands of people, not fifty, and reveals actual likes and dislikes (not just a presumed love of all 'wispy male vocals', however defined).

I suspect this is the case, because:

a) songs that come up feature less musical overlap than before on song type, and more on taste - Owl City came up on my station for Jack's Mannequin the other day. Musically, they're quite different. Demographically, they're both near the centre of the bullseye of Stuff White People Like.

b) The 'Why did this song come up?' tab has now gone, as it has to when the answer to every question is 'because that's what happened when we inverted the giant matrix and extracted the principle components'

and

c) It's gotten a lot better, as it does when you start using really huge datasets with really good information extraction mechanisms. The Jack's Mannequin station is almost a pitch perfect playlist for SM's music tastes.

And I suspect this was the plan all along - the whole Music Genome thing is mainly a seeding mechanism (and for new songs, which don't yet have any likes or dislikes), and the real information was always in user choices. At a minimum, if these guys aren't using this information, they're crazy.

Put it this way - the algorithm is now good enough that I listened long enough for the frequent ads to bug me, and actually paid for the upgrade (perhaps the first 'freemium' product where I bought the extra part). Proof that good matrix decomposition can really add value!

Signs you may be watching the wrong show

When an ad for yeast infection medication comes on, after being talked into watching an episode of 'Bones' on Hulu. I note in passing that such ads don't feature during House, The Simpsons or Southpark. Maybe those marketers are on to something after all.

Thoughts on The Social Network and The Olympic Games

I watched 'The Social Network' this evening, and found it excellent. I think Scott Adams got it right when he said that (regarding the portrayal of Zuckerberg) the movie was "massive respect from a genius in one field (Sorkin) to a genius in another". The guys who should be really cut are the Winklevoss twins, mercilessly portrayed as pampered stuck-up rich kids without enough talent to start a real company.

Anyway, the movie left me with a feeling I can best describe as 'life choice envy' for Silicon Valley types starting their own companies and attempting to change the world. Actually, 'envy' isn't quite the right word, as there wasn't any sense of resentment of their success.It was more a feeling of being impressed by people who really are trying to do something big with their lives, and a slight hollow sense that one's own aims and achievements are insufficiently ambitious.

I originally encountered a similar sensation when watching the Olympics as a teenager. There was a dawning sense that a) this was something that I'd abstractly wanted to do when I was a child, and b) I wasn't ever going to get there. There are two potential ways of interpreting this. The first is that one should have tried harder, as in aiming for world-beating success in some endeavour, and that one's own decisions were too slack and unambitious. This was always the initial feeling I got when watching the Games. The second interpretation (which usually came after a day or two's reflection) is that one is simply envying an outcome, not a decision - what you really want is the feeling of great success and winning a gold medal, not the whole package of the gold medal plus what's required to actually get it. 

Once I learned enough economics to think about the problem in econ terms, for the Olympics specifically it became apparent that it was probably a very poor thing to aim at. The Olympics is a classic tournament, and people tend to be overconfident when estimating their chances of winning, and hence overinvest in tournaments. The other big-time mistake is that people screw up the expected benefit by focusing on the highly salient outcome of the gold medal, and not the many, many non-salient parts, including:

a) The guys who came 2-8th in the final and always wondered 'what if I tried a bit harder?'

b) The guys who didn't make the final at all, and felt like complete failures

c) Everyone, winners and losers, spending years and years denying themselves tons of pleasures to do nothing but train, a cost that is spent before you have any idea whether you'll succeed at all, and

d) The winners themselves after their sports career is over, feeling hollow and aimless as the rest of their life will always pale into comparison with that one high point, drifting ever further into the past

Salient, but not frequent

Frequent, but not salient

So the Olympics itself is mainly option b) - you're really envying a feeling, not a decision. I personally think that the Olympics bet has a negative ex ante expected value for the vast majority of people actually competing, let alone for guys like me who started without an amazing natural ability. (I'm athletic, but well aware of where I fit in the distribution).

Which brings me back to The Social Network. For me, computer science was in many ways the Road Not Taken. I'm happy with the road of finance and economics, but my pursuits are mainly academic. It's hard not to feel a sense of being impressed by guys who actually start tech companies, employ people, struggle to meet payroll every month and eat ramen instead of pulling several hundred K salaries, all because they're aiming at creating a world-changing company. I remember thinking this when meeting for the first time a friend of a friend who is my age, and is out in Silicon Valley doing just that. Starting a tech company, in other words, was the Olympics that I might have actually had a chance in. 

All in all, I don't think I would want it, for similar reasons I don't want to enter the Olympics - starting up a business isn't actually a tournament, but it does have highly skewed payoffs, and tends to attract overconfident people who underestimate their probability of failure. 

But I did leave with the feeling that I used to be more ambitious in the tasks I am doing. And that I should aim higher, and be willing to fail. Risk and Reward are generally correlated, in life as in finance. That is the correct lesson, I think - focus on picking bets with good expected payoffs ex-ante, but don't be afraid of working hard and betting big if the price is right.

As Daniel Burnham put it:

“Make no little plans; they have no magic to stir men's blood and probably will themselves not be realized. Make big plans; aim high in hope and work, remembering that a noble, logical diagram once recorded will not die.”

Preach it, Mr Burnham! Onward!

Free advice for musicians composing song lyrics

If you're ever tempted to make one of the following crimes against the English language, you should probably shoot yourself in the face. For some reason, a lot of the worst ones relate to the area of love. Behold, The Most Annoying and Clichéd Rhyme Schemes in the Song Lyrics World:

-Rhyming 'Maybe' with 'Baby'. Apparently equivocation is the most common sentiment people feel vis a vis their loved ones.

-Rhyming 'Making Up' with 'Breaking Up'. Oooh, such a clever contrast! When you do this, my hemorrhoids start acheing up.

-Rhyming 'Love' with 'Glove', 'Dove', 'Above' or 'Uv' (Mark Steyn had a great essay about this). For love, you're best off not putting it at the end of a line.

-Rhyming 'Die For you' with 'Lie for you' and/or 'Cry for You'. What kind of loser would want these to be their expressions of love anyway?

-In much the same fashion, rhyming 'Knees' with 'Please' (double points off if you're 'on your knees, begging please'. Get up off your damn knees and show some backbone, man!)

-Rhyming any combination of 'mad', 'bad', 'sad', glad' or 'had'. Honestly, this is so hackneyed, obvious and infuriating

Doing any of these will place you permanently on my $#!7 list - not that this on its own is reason to shoot oneself in the face, but your lack of creativity will be.

Things that make me dislike Justin Timberlake less

Via Kottke, check out this awesome video of Justin Timberlake and Jimmie Fallon doing a rap performance.

Tropes that make no sense - the 'I Love You' pickup line

I was watching the first episode from the new season of House the other day, and found myself irritated that they decided to repeat one of Hollywood's most nonsensical staples: characters who declare that they love someone, either immediately after the first kiss, or (even worse) before they've even kissed for the first time.

Honestly, is there anyone who is a) above the age of 13, and b) not a functional retard who has actually done this? Is there anyone, man or woman, who would not frantically start looking for the exits if presented with this pearler of a pickup line?

I suspect part of the reason for this is that scriptwriters need a quick way to create emotional tension. Just putting a margin note to the effect of 'have a bunch of slightly awkward pauses and uninspired sounding conversation as you both look a little too long at each other, then look away pretending it didn't happen' might be a bit hit-and-miss for some actors. But Hugh Laurie? Really?

In the episode, Cuddy remarks that 'I told you I loved you, and you didn't say it back.' House has some throwaway joke to brush it off, which sounded jarring given his character. Surely someone with such a piercing (albeit cynical) understanding of human nature, along with a healthy dose of misogyny, wouldn't just let that slide?

My answer would have been 'No $#!7, eh? That's because we hooked up for the first time less than 24 hours ago, and that's not quite how love works for me.'

Yellow

I've often thought that many songs are made by a single good line.

I think Coldplay's 'Yellow' is a great example of this. The subject is a girl, and it's a love song. If ever there was a more tilled field in the annals of poetic inspiration, I don't know what it would be. It's damn hard to say anything original about love. Most of it is pure boilerplate, too:

'Look at the stars.

Look how they shine for you,

And all the things you do.

Yeah they were all yellow.'

Not very inspiring stuff.

But there, right in the middle, is a line that always stuck with me:

'For you I'd bleed myself dry,

For you I'd bleed myself dry.'

It works so well, for many reasons.

They say it twice, with the pause and drop in the instrumentation before the repetition to make sure you pay attention. Then the great guitar riff comes in powerfully. But that's it -it's never uttered again, even though it's clear that the whole song is building to that point. The worst thing you can do with a good idea is bash the audience over the head with it by repeating it over and over.

I think it's the most interesting sentiment about love that I can recall being said in the last 10 years.