Assorted links


'...gospel music comes in at the lower end of the spectrum — or, as Griffith puts it, is music for dumb people.'

Yep, Jerry Lee Lewis, Carl Perkins, and Elvis Presley were certainly among the dumb people that liked gospel.

'One of the most important things to remember about this album is it's really just four guys in a room shooting the breeze, goofing around, and stumbling through a few old songs. This wouldn't be especially interesting under most circumstances, but the three guys in question happen to be Elvis Presley, Jerry Lee Lewis, and Carl Perkins, which, as you might imagine, makes quite a difference. Perkins was doing a recording session at the Sun Records studio in Memphis on December 4, 1956, with Lewis playing piano on the date, when Elvis, back in Memphis after a stretch on the road following his breakthrough as a major star, stopped by to say hello. Elvis, Perkins, and Lewis began casually jamming -- mostly on old gospel tunes they remembered from a shared Baptist upbringing -- and Sam Phillips had the presence of mind to record the proceedings.'

And a youtube link to The Million Dollar Quartet - Just A Little Talk With Jesus

Yep, Jerry Lee Lewis, Carl Perkins, and Elvis Presley were certainly among the dumb people that liked gospel. Although I loved Jerry Lee Lewis but he never hit me as 'smart'. Let us on forget:

Also include Sam Cooke (Started out Gospel), Ray Charles, Aretha Franklin, Little Richard, most doo wop groups, etc. As Elvis once stated that"Rock and roll is basically gospel (music), or rhythm & blues (is too)" I still think the sacrareligious rock song is Little Richard swiping Mahalia Jackson licks on Tutti Fruitti.

Musical tastes are a decent placeholder for race. Paging Steve Sailor.

Yep, Jerry Lee Lewis, Carl Perkins, and Elvis Presley were certainly among the dumb people that liked gospel.

Yes they were. Your point? Based on his high school records one of Presley's biographers claimed he probably had an IQ of 70. The Army must have tested him when he was drafted and I doubt they would have taken someone so dumb. But the record of the test must be out there somewhere.

Lil Wayne on the far left? Racist!

#4: before jumping into conclusions look at the distribution for Radiohead. It looks like a nice normal distribution that goes from 1000 to 1500. Why it is a small blue sign that implies that ALL people that has Radiohead as favorite music scores 1200?

Ps. If you said Beethoven it's not the music, it's your tiger mom.

"Ps. If you said Beethoven it’s not the music, it’s your tiger mom."

+1 signaling

Also, how seriously can you take it when they are taking average SAT scores from a university and then polling Facebook pages of the students.

"Also, how seriously can you take it when they are taking average SAT scores from a university and then polling Facebook pages of the students."

Make ASSUMPTIONS about the bi-variate distribution between individual SAT scores and music taste at the university.

Ecological fallacy.

Beethoven is interesting.... I love classical music, but somehow I never quite got grasp of Beethoven. Being a string player, Beethoven is just too hard to play. Unless you are really good (i.e. solist with orchestra) /in really good company (very good string quartet) or play in a symphonic orchestra, you just don't get to play him. The quartets are so hard; lot of work and the effect just isn't quite there.

Seeing him as the only classical author, the signalling really seems the obvious answer.

The survey isn't of classical string musicians, it's who normal people like to listen to. Yes probably some signalling, but for most there's almost nothing greater in the classical canon than Beethoven symphonies. The legend has outstripped the reality, but the reality of Beethoven shouldn't be forgotten: probably the best ever.

Same thing with the Beatles...they get overhyped for sure, but the facts are that they were awesome.

I don't understand what's going on with this graph. All the genres are super low, not just Gospel. Even Classical falls on the "Dumb side". When looking at the genres even country beats classical. There's something very off with these generalizations and the data itself.

Looks like most music falls somewhere in the middle of the distribution, with music on the lower end of the distribution correlating with what poor, underprivileged people might listen to and with music on the higher end of the distribution correlating with what snobs might claim to listen to.

That's pretty classist to imply that poor people are stupid and rich people are smart.

Huh? I am suggesting that poor people have lower SAT scores because of the disadvantages they face in life, and that people with high SAT scores are pretentious. You literally can't get less classist than what I'm saying.

You are absolutely being classist, but you're apparently of the belief that you can only be classist in one direction. Next, tell us how it's impossible for women to be sexist, or for minorities to be racist.

Fantastic goods from you, man. I have be mindful your stuff
prior to and you are just too wonderful. I actually like what you've
acquired right here, really like what you're stating and
the best way during which you say it. You're making it enjoyable and you continue to care for to stay
it smart. I can't wait to learn far more from you. This is actually a great web site.

people with high SAT scores are pretentious.

You forgot to exempt yourself from that category.

Classist maybe but certainly true - stupidity is what causes poverty. I would expect everyone here can agree the poor are poor due to their rock bottom IQs.

Crime is a function of poverty! Except crime by white men. I think we can all agree.

"poor, underprivileged people might listen to"

"snobs might claim to listen to."

Condescend to those on the bottom, attack those on the top. How very snobbish of you.

4. Boy, this is a gift to racists.


You neatly miss the question. It does not matter if it is a gift to racists or not. It matters if it is true.

#2: too bearish. Lots of people drive from point A to B at morning and B to A at sunset. Even if the changes in your neighborhood quiet streets are not mapped, it would be great to push the autopilot button once you get into highway/main streets.

A. To buy a google car, you must first buy an office at the googleplex to work
B. To buy a google car, you must purchase a weekly remapping of your travel zone subscription for $1000 per week

It's silly to talk about "how hard it is to keep a database of the nation's streets." You can keep that in your pocket if you want, and it's only going to get easier. And of course you don't need the entire nation's DB in your car at any one time.

Plus, these cars will all be talking home to the mothership. You only need the first car to notice a new temporary traffic light and pass that information back. They've already demonstrated that they can detect a new traffic temporary light in one demo.

LA comes to my mind, freeways are less than 1000 miles but average annual daily traffic is more than 4 million. I bet drivers would be happy to push the autopilot on freeways and get control back when they get into the "unmapped" territory.

One big thing going for self-driving car features is that it's a technology that's very amenable to incremental deployment. We can creep up on it by continually adding driver assistance features to normal cars without needing to deploy everything at once and hope it works.

This is a reason to bet on the more tech-oriented car-makers rather than Google.

#1 - no big surprises here. People usually move 1.d4 or e4 or c4, castle king side, and protect their k-side pawns, though f2 and f7 are often sacrificed to open the rook file. Knight is worth less than the bishop is worth less than the rook is worth less than the queen is worth much less than the King (100%). Standard theory for chess.

I don't know why we should necessarily expect this to correlate with value - plenty of chess games begin with the two queens being exchanged pretty early on.

Maybe at the patzer level but good players don't trade queens unless looking for a draw, or up material which is rare "early on".

OK, so what explains the discrepancy between d2 and d7? Just the Sicilian? And c2 survives much more than d2 despite the Queen's gambit? I find some of these surprising. Maybe you could elucidate those as well.

@Michael, as you say both the apparent discrepancies are probably due to the Sicilian. Queen's gambit would get rid of c2, as would the Nimzo, but in Sicilian it typically survives (and actually becomes a target that has to be defended when black plays Qc7). Have a good game.

4. This is all social class based

Yup. Smart people listen to music they have been told is for smart people. Quelle surprise.

"It is unpossible for music to appeal to different IQs. Except country music, that obviously appeals to idiots."

- coastal social justice warrior elitist

I think there is another important variable.

#2 - So, technology which is not ready for prime time is not ready for prime time? Errr...

Obviously we aren't all going to be riding in driverless cars in a month. No one has claimed otherwise. I have no doubt there are many challenges and that the cars have 'a ways to go'. However, to say that it won't work is ridiculous. Piloting a car is exactly the kind of problem a computer is going to excel at, given enough data. It is fixed and finite and solvable.

I also doubt that Google's long term plan is for the cars to be dependent on ultra-detailed maps. The final Google car will have the mapping car 'built in', and so all its maps will be up to date at all times.

I don't know what Google's long term plan is. Maybe they don't even. I've definitely become more bearish since learning that they aren't really working on the problem of autonomous mapping. I can't speak to it's difficulty, but if they're choosing not to tackle it, that must mean it's fairly difficult.

"It is fixed and finite and solvable."

You see the world that way? Really?

"Piloting a car is exactly the kind of problem a computer is going to excel at, given enough data."

But if the data is from yesterday, then piloting a car today can be really hard.

I've been to cities I don't know and rely 100% on the navigation system to find my way. It is stressful to look at every crosswalk, traffic light, road sign trying not to make a mistake while every time you look at the mirror there's the homicidal look of someone who thinks you're too slow. As a human I'm doing the mapping of a new city.

Compared to driving at home, basically I don't read any road sign because I know my daily commute. I'm not doing the mapping, I rely 100% on memory. Not looking at the signals is not only a robot car problem, it happens also when you feel comfortable.

#2. The fact that the google car has to have a traffic signal in its map isn't a problem that can be solved with more frequent map updates. The problem here is that google is trying to make up for crappy eyesight with an awesome 'mental map' of the environment. Think of it sort of like a 90-year-old retiree with glaucoma trying to stay behind the wheel by driving only roads he knows by heart. This is a fundamental problem. If a Google car can't drive around safely without a detailed map (including the location of all traffic signals), it simply can't drive safely. And if the system were able to see and comprehend its environment well enough to update its maps on the fly, it wouldn't need such detailed maps in the first place.

Nobody said it was impossible. The MIT guy, for instance, said it wasn't likely that he'd see a fully autonomous car in his lifetime, and he's only 49.

Your suggestion makes sense, but you have to wonder why it something like that isn't being talked about right now.

I doubt will see a successful autonomous car that relies on a map for anything other than routing. Driving will need to rely on sensors and decision trees.

I think this particular article is too Google-centric. A lot of companies are working on various degrees of autonomous vehicles and they are showing a lot of progress. Indeed, it's to the point of putting a lot of these features into their current high end models. Google not having an autonomous vehicle, is not the same as saying no company will ever build one.


I'd bet heavily on Google being unsuccessful here. They have had a bunch of embarrassing failures trying truly disruptive products.

Mercedes seems to be the technical leader, but there are several interesting developments at other firms. A self-driving S-Class makes a lot more sense than Google's self-driving death-trap.

The reason that people focus on Google here is that their technology is the farthest along. It strikes me as deeply unlikely that Mercedes or VW is going to make the major advances in computer vision that would be necessary to have a practical map-less driverless car. The reason you don't see discussions of that is that they don't have their technology up to the level where it even makes sense to try.

Mercedes-Benz S 500 Intelligent Drive Autonomous Car Self Driving Car

It's got a vision system and it's autonomous. Indeed, it's driving in light traffic. I don't know how well it compares to Google's current system, but I know it's far better than anything that existed 10 years ago.

Ironically enough, Mercedes technology is much cheaper than Google's technology. And their already being a car-maker gives them a path for incremental feature deployment. Google pretty much needs to be all-at-once.

None of those links should be cause for you to imagine that Mercedes is close to solving this problem. And the RADAR-first approach described in the only meaty article is your first clue that Mercedes is way behind here. RADAR isn't going to help you with any of the really hard judgment call problems that Google is grappling with -- it inherently can't read signs, can't tell what color lights are, etc.

RADAR might well be a better sensor choice than LIDAR (certainly, it's better at handling weather, though it's not like LIDAR is without virtues -- it'll have a much, much better resolution). But the computer vision problems here are core hard computer problems. You should prima facia doubt that Mercedes is really equipped to solve those problems.

It would be much simpler and safer to have autonomous passenger aircraft, in fact, it's already been done. But we don't hear about Google flying any 747s from Kennedy to LAX.

There will never be robot planes, not because there can't be, but because people won't fly in something without a human pilot. Right or wrong, they want to know a person is in charge in case the robot fails. Also, they want the pilot to be dead too if the plane crashes and kills them. Weird but that's just how (most) people are.

In fact that's why I imagine driverless cars will only succeed if they allow a human to manually override and drive the car. Just in case.

#2: As the driverless cars get on the road, they themselves, in a distributed fashion update the maps databases.

And, how hard would it be to persuade cities and states to start mandating that things like traffic signals and the like have to broadcast by signal, as well as sight, their presence and the information they are supposed to be conveying?

Seems like infrastructure could be made smarter simultaneously with making the cars smarter.

This could be a good idea, but I don't think it will happen until after the driverless car already exists. It will just make them marginally more reliable. But they will have to get there on their own first.

One could imagine this being useful for safety systems in regular non-self-driving cars, so maybe it will happen regardless.

But updating government equipment is something that will take a long time.

The reason they need the maps is because the sensors they can put on the cars are not good enough to understand the environment around them.

Driverless cars can't update the maps because the POINT of the maps is to shore up the bad "vision" of the driverless cars. You're effectively saying, "Hey, let's take a person with uncorrectable really bad vision and give them a map so they don't need to use their vision. How will we make the maps? Other people with uncorrectable really bad vision will tell us what's there!"

Wrong. They don't have the hardware/software to process the imagery to a sufficient extent in realtime

That "hardware/software" doesn't EXIST. The detailed maps that Google created are human-corrected. This isn't a matter of slapping another CPU or a few more gigs of RAM into the car. The car's sensor array costs $80,000 -- it's not like they just cheaped out on the computer.

Funny how the main chart for the music link leaves out Beethoven and says smart people listen to Radiohead, but if one goes down to the lower chart one finds listeners of Beethoven simply way ahead of all the competition. But thn, maybe classical music is "classist."

On France's problems, there is so much media overdoing them. So, we get this big headline that it will grow more slowly than the EU as a whole, but when looks at the difference, it is essentially miniscule. In many ways, France if very average for the eurozone, partway between the troubled southerm Med economies and the better off northern ones, led by Germany.

"On France’s problems, there is so much media overdoing them. So, we get this big headline that it will grow more slowly than the EU as a whole, but when looks at the difference, it is essentially miniscule. In many ways, France if very average for the eurozone, partway between the troubled southerm Med economies and the better off northern ones, led by Germany."

I think there is some truth to that assessment, but it doesn't mean France is doing fine. It just implies that the Euro Zone is, on average, economically stagnant. (

Looking at the Eurostat graph on Annual GDP growth, The EZ has close to 0% growth since 2008.

#5 is pretty silly. For one thing, they weren't disguising the McDonalds as "Gourmet Snacks" they were asking people if they liked their "organic" fast food. When framed like that, the expectation is going to be that it will most likely taste worse than "regular" fast food, so it's not surprising that people would be pleasantly surprised that it tasted like actual fast food. It's not like these people were expecting fine dining and mistook McDonalds for the French Laundry.

Secondly, I would argue that the people they were giving the food to weren't "critics" or "experts", they were people at a catering convention. I don't know about anyone else, but most catered food I've had is barely a step above McDonalds. I remember spending $150 per plate at my wedding and thinking how much I had wished I just got the In-n-Out food truck to come instead.

The only thing this video shows is that people are really dumb when it comes to buzz words like "organic". Just because its organic a piece of fried chicken is still not healthy food.

There's nothing unhealthy about a piece of fried chicken unless it goes down your windpipe.

High calorie. Lots of animal protein (carcinogenic)

Calories are why you eat food. If you think otherwise, try living on a no-calorie diet, just drink water, see how long you last. Animal protein is what carnivores eat exclusively, they don't all get cancer. Wrestle with a lion or tiger and get back to us on how unhealthy it is.

Paleo alert!

Read all the comments about item 4 as applying to item 5.

Good point JayT. My take is they were just being polite with the camera on them.

4. "Smart" people like John Mayer over Pink Floyd?

Really? I thought it was a music survey.

If anything completely destroys the validity of the survey, that conclusion is certainly the one.

SAT tends to be popular along the coasts, ACT tends to be more popular in the middle of the country.

The highest SAT scores by state are found in the middle of the country where the fewest people take the SAT (e.g., Iowa and Minnesota).

Given that university populations are regionally skewed, this is going to mess with the accuracy of the data.

Smart country music fans in Iowa take the ACT andgo to Iowa State University.

Smart John Mayer fans in NY take the SAT and go to NYU.

I don't know if that tells us whether or not the Iowan is smarter or dumber than the New Yorker.

Maybe it makes me dumber, but I'll still take my classic rock, please.

#4 Jazz seems to require further breakouts, but considering it's all to low status on the intellectual spectrum, one might conclude that jazz (or something called jazz) might reflect Kenny G fans, certainly not John Coltrane fans.

Just as I thought, Rush is off the charts, well beyond Beethoven, into the 1500s...

2: In the future, instead of asking "where are my flying cars?" or "where are my hoverboards?" we will be asking "where are my driverless cars?".

4: If they're correlating college SATs and campus music preferences, the middling score that they give to the Grateful Dead seems dubious to me. The campuses with the most Deadheads that I've ever seen were MIT and Caltech.

If driverless cars don't materialize in the next 20 years, “where are my driverless cars?” will be a much more legitimate question than "where are my flying cars?" ever was.

Most smart people seem to think driverless cars are quite likely, with some caveats and skepticism around the edges. Very, very few people with degrees in engineering ever thought the flying car was a reasonable proposition.

You're so awesome! I do not believe I've truly read through a single thing like
that before. So good to discover someone with some original thoughts on this issue.
Seriously.. thanks for starting this up. This site is one thing
that's needed on the internet, someone with a bit of originality!

Comments for this post are closed