HNHacker News
TopNewBestAskShowJobs

jahooma

188 karma · joined October 10, 2013

submissionscomments
jahooma··on Can Family Secrets Make You Sick?
Notice how the lone medical expert represented was skeptical of the findings.

That's because among serious psychologists and those who study genetics, the view advocated here is pretty much exactly wrong: your childhood experiences have almost no effect on the adult you become.

The public and the media tend to believe otherwise, and regularly put out stories like this one, perpetuating the myth. But it's just not true.

We've known for years that the genes you inherit from your parents are far more important than whatever parenting they do. Adopted siblings raised in the same household grow up to be no more similar than strangers plucked off the street. And identical twins raised apart grow up to be ridiculously similar -- almost creepily so -- despite having no shared environment. When you run the correlations, genes turn up as a large factor in every single measurable outcome (education level, income, IQ, height, weight, personality, choice of college major, and so on), whereas shared environment -- where you grew up, how your parents raised you, your socioeconomic status -- begins as a moderate impact while you are a child and then slowly disappears in your teens, until at age 25, your IQ and personality have basically nothing to do with whether your parents read to you at night or spanked you or talked to you or ignored you completely, or what siblings you grew up with, or what house you grew up in, or what town you grew up in.

(Coincidentally, ACE is a relevant acronym here, but it stands for something different than from what it does in the article. Additive genetics, Common environment, and unique Environment are the components of the outcome of twin studies. Common environment is the factor of your parents and household, and unique environment is every other environmental factor. http://en.wikipedia.org/wiki/Twin_study#Methods)

Now, there is a small caveat to the above. If you were severely abused as child that can impact you later, but it truly has to be severe abuse. To lower a child's IQ, you'd have to really starve them or deprive them of key nutrients -- but this pretty much never happens in the US. A diet of only McDonald's is more than sufficient to prevent this kind of stunting. Likewise, the sexual or physical abuse would have to be pretty extreme to impact you later -- much more than this story let on.

-----------------

So, the real story behind this NPR piece is that people with bad parents often grow up to have problems not because of bad parenting, but because their bad parents happened to pass them bad genes. In other words, the common cause was that both the child and the parents had bad genes. You should never trust a study about parenting or the effects of the environment if it doesn't take genes into account.

And this is where people get uncomfortable, because no one wants to believe that something you're born with, something you cannot change, is so important to who you are. The "truth" about the effects of genes and your environment does not fit in well with mainstream ideals of equality of opportunity and free will. Thus, the impact of genes are routinely downplayed everywhere you look, and environmental effects are raised up, like in this piece. There is broad, systemic bias at work, making the world out to be the way we want it, rather than the way it really is.

But we do ourselves a profound disservice in this (often unintentional) act of twisting the truth to fit our ideals. It causes grief when we blame fat people for being lazy or lacking self control, when in reality their obesity is largely a product of their genes and the modern phenomenon of cheap and tasty food. We praise smart kids with lucky genes who get A's, while putting down their less bright peers, even if they worked harder, because they only managed to get C's. We love beautiful people, but shun their less appealing peers.

So I ask you: Where is the compassion for those with unlucky genes? Where is the support for those who don't have the talent to live out the American Dream? Why is it that no one knows or cares about genetic luck -- probably the greatest source of inequality and unfairness in our world?

jahooma··on How to Dial Down the Earth’s Thermostat
Careful. It's dangerous to dismiss ideas based on intuitive feelings rather than evidence-based analysis.

Consider nuclear power. If we had scrapped developing it because it sounded dangerous with the argument: "What. Could. Possibly. Go. Wrong?", then France today wouldn't generate 75% of it's electricity from nuclear power. Many countries around the world would be forced to use more coal and fossil fuels, adding a lot more pollution to our environment. And we'd be deprived of the source of energy that causes the fewest deaths per amount of energy produced.

But more broadly, this type of reasoning just isn't scientific. We can only judge geoengineering by carefully understanding what it would do. Yes, this would have to include a tough analysis on possible risks. But dismissing it out of hand and depriving it of funding because it feels wrong in your gut is not rational.

It's easy for us to approve studies that, say, fight disease. Everyone knows it's important to get information on, say, what the health impact of a new chemical is. But those are the easy cases. Doing science well is about getting it right in the hard cases. I believe geoengineering is one of those hard cases, because it feels wrong to us that we'd be able to control our environment. It doesn't seem like "our place" to tinker with nature.

But let's see. Volcanos erupt making abrupt, profound changes in our environment. As mentioned in the article, when Mount Pinatubo erupted, it shot enough stuff in to the atmosphere to cool the earth by nearly a degree Fahrenheit. Given that the world didn't end because of this, perhaps there's a place for controlled intervention by humans which more efficiently and precisely alters the world.

We don't know yet if anything we think of would work. We need to do more research. But do you really think we should quit because it doesn't sound nice?

The fate of the world rests in our hands. Are we going to throw away one of the few potential solutions to our problems because it doesn't appeal to our human intuitions?

It's time to grow up. Let's do science properly before we fuck ourselves over.

jahooma··on Is Social Psychology Biased Against Republicans?
Well, there's only one truth. It's perfectly plausible that one set of ideas (conservative ideas) could be less scientific than another set (liberal ideas).

However, ideology can influence us, since we are not perfect at science. Without the strong objective feedback of the hard sciences, social psychology is particularly vulnerable.

The bias alleged in this article actually starts to look more plausible when you consider the history of social psychology and related fields. Back in the 1970s, it was taboo to mention ANY innate qualities of human nature. Gender was purely a social construction. Babies were identical in every way--people were different only because of their upbringing. Prominent biologists like E.O. Wilson who argued genes might play a role were attacked and demonized. Nothing could contradict this doctrine of the blank slate -- that we're born perfectly malleable. I highly recommend Steven Pinker's 2002 book, titled "The Blank Slate" [0] which brilliantly debunks this theory, and lays out the best characterization of human nature I've ever come across. Seriously, read this book, it is a masterpiece.

Anyway, from what I've read recently, it seems many fields like social psychology are still captive to lesser versions of the blank slate fallacy. Human nature is not as malleable as they think. Our instincts are still there (gotta eat, sleep, procreate), and even "higher" areas of the brain for things like language, emotion, and thought seem to be heavily innate (to think at all requires machinery, and there are many ways to craft that machinery). Also, genes really are more important than most seem to realize. The twin and adoption studies show that the majority of variation among people in intelligence and personality is due to genes. How your parents treated you and brought you up has almost no effect by the time you are an adult in important metrics like IQ and personality tests. Smart people are largely born smart.

These sort of "deterministic" ideas fly in the face of traditional liberal values. Topics like gender discrimination and societal inequality are undermined by these ideas, so that provides a reason for a liberal thinker to push back against them.

If the gender gap in computer science is due more to innate differences in interest than to discrimination, and if inequality of income is due more to innate differences in talent than differences in opportunity, then that makes it harder to argue for reform. I believe thinking along these lines is the major cause for bias today.

So yes, I agree social psychology is quite biased. In the future, we will look back with horror at how we let politics and ideology interfere with science.

[0]http://www.amazon.com/The-Blank-Slate-Modern-Denial/dp/01420...

jahooma··on The Most Valuable YouTube Video: A Brilliant Gimmick
Interesting idea, but I doubt it can work. There's no incentive to bid after $20, once the novelty has worn off.
jahooma··on Google's Self-Driving Car Can't Navigate Heavy Rain or Most Roads
The common sentiment has been that self-driving cars are just around the corner. The reality is quite different.

The remaining problems to solve such as navigating the elements or obeying construction signs or interacting properly with pedestrians or police are orders of magnitude more difficult than the problems that have been solved thus far. These problems are fundamentally different in that they can't be solved by current AI techniques and vision algorithms. The progress made so far has been quick, but it relies on technology that Google has already mastered. We'd be mistaken to think that the remaining challenges will be solved as easily.

It's a repeat of the classic mistake that has plagued the field of AI since the beginning: we underestimate the difficulty of problems that humans solve easily. We simply aren't aware of the incredible complexity involved in our simplest decisions, such as pulling over to allow an ambulance to pass. This is simple right? Just slow down and move off to the side of the road. But when is it OK to move off to the side -- what if there is something in the way, what if a pedestrian didn't expect you to move there, what if the car behind you suddenly gets in the way while it's pulling over, what if you're on a bridge, what if the ambulance behind you turned already and no one expects your car to suddenly pull over? Similar or more difficult problems arise when there's debris or potholes in the road, other poor drivers, bad weather, jaywalkers, policemen giving orders, road work, detours, etc.

What you find is that the last 5 or 10 percent of the capability required to make self-driving cars feasible represents a category of problems which we don't know how to solve, requiring a level of sophistication far beyond the current state of the art and perhaps approaching general intelligence in some cases (such as interpreting signs).

Better approaches involve shooting for more modest goals instead of full autonomy. Car companies are making investments in these more practical, incremental improvements, like automated parking and advanced cruise control.

But unlike car companies, Google isn't in this game because it thinks it can make a profitable and successful product. Instead, it's obvious that the main function of developing self-driving cars is as a PR tool (and the same goes for the rest of the Google-X projects). Google has gotten a lot of positive press for their self-driving cars, and they even use it to attract new employees.

However, I predict this positive press won't last (this article being an early example) because people's expectations are way too high. As years and years go by without much progress, Google's self-driving cars will increasingly become a PR liability and will be compared to the promised flying cars of yesteryear.

jahooma··on Solar Roadways almost funded
Interesting. Solar-powered roads is an appealing concept.

However, one has to wonder whether it is efficient enough to be worth doing. Are roads really the best place to collect sunlight at scale? Current solar farms are set up in specific places at tops of hills or what have you in order to maximize energy production, whereas streets have no such guarantees and require special resources and engineering to withstand the forces involved with supporting cars and trucks. Doesn't sound so ideal to me.

Even more concerning is that they never address basic feasibility questions like this, or say how much it costs or could cost if done at scale. All that their crowdfunding campaign seems to say is "Wouldn't it be cool if we could put solar panels in our roads?" without any analysis of whether or not we'd actually WANT to do that. (Hint: We probably don't because it's probably massively inefficient.)

It's obvious to me that this idea has very little potential and that this family is putting on a show to extract money from green, eco-friendly types who lack critical thinking skills.

Think about that before you give them another cent.

jahooma··on $1,000-a-day miracle drug shocks U.S. health care system
But how could it be any other way? Isn't it an inherent gamble to develop drugs since most of the drug-creation attempts are bound to fail? You could just pay researchers to "try" to come up with drugs, regardless of whether they get results, but that doesn't seem like it would work very well. Is there any other way to smooth out the rewards?
jahooma··on Why TypeScript Isn't the Answer
Hmm, that was pretty one-sided.

The author fails to acknowledge any of the good consequences of typescript. Adding static typing, classes, modules, enums, shorter anonymous method syntax, default arguments, etc. etc. make a world of difference for writing clean, structured code that is rid of a whole class of bugs at compile time. It's simpler to read, easier to write, and more likely to be correct in typescript. And the intellisense in Visual Studio is phenomenal. Contrary to what he says, I think typescript does tackle head on the biggest problems of javascript and makes massive improvements.

Regarding type definitions, I've never had problems getting correct type definitions for all the libraries I use, and it's also always possible to just not use type definitions as a backup. He forgets to mention how great it is to be able to use any javascript library straight up (unlike most compile-to-javascript languages) -- and that in 99% of the cases there are existing type definitions which add on static types to regular javascript.

That's right, in perhaps one of the most brilliant features, typescript lets you add types after the fact to any javascript library, and it works well. Thus, IMHO, most javascript libraries are actually better in typescript!

Though typescript doesn't fix every problem (like the mechanics of what "this" refers to, although it helps here too), you almost don't want it to fix these, because only this way can you use regular javascript and its vast libraries so simply.

Typescript has always been a sort of compromise between making a better statically typed language and retaining a javascript core for compatibility, but this balance is executed so well that you just about have the best of both worlds.

← PreviousPage 3 of 3