Monthly Archives: July 2011

Why I Hate Science

TL;DR -> The word “science” is being used in so many contexts as to be very bad for the application of real science. Scientists have themselves to blame for this. They need to create better standards of communication both internally and externally if they want their profession to retain the rightful luster it deserves.

I love science, when the word means the use of Bayesian models to come up with new and observable reproducible patterns of behavior. I am deeply intrigued by science, when the word means the groups of people and political processes used to do this model creation. And I hate science, when the word means something used as a club to beat up people you simply disagree with.

That’s the problem: the word “science” can be used in all of those contexts, and it’s never clear exactly which one a person is using. Sometimes it looks like it just means “anything a scientist is saying.”

Listening to some pedantic columnists (you know who you are), people out in the real world are a bunch of babbling idiots, swigging down cheap beer and going to tractor pulls while thinking that science is just a bunch of hogwash for nerds. Usually the explanation is that they got this way because of a poor public education system and a media that promotes conflict over truth.

It’s not unusual in school to hear this religion of science put another way, science is the beautiful application of reason and logic that slowly and inexorably moves from ignorance to truth.

Bullshit.

First, there are things that are known and predictable out to 3 or 4 sigmas. I don’t think anybody believes these things are wishy-washy. But then there are scientists who are experts in one area who go off the deep end picking up little pieces of theory here and there and speculating ad naseum. I don’t think you need to be a kuckle-dragger to tell to see that there is a vast difference.

This kind of pseudo-scientific slime is everywhere. Pick up about any book in the “science” section and you’ll read some of the wildest speculation: what happened 50 thousand years ago to bring about domestication of animals, how the human sex drive evolved, where all the dark matter has gone to, or how it is that conservatives have different brains than liberals.

Don’t get me wrong: there’s nothing wrong with such speculation. In fact, it’s required for science to progress. What I’m concerned about is that the average scientific layman reader — the person who assumes they are just as up-to-date on scientific matters as possible — has no freaking idea where the real science leaves off and where the bullshit begins. One minute the scientists is talking about Carbon-14 dating of stoneware. The next minute he’s talking about tool use in prehistoric societies. Then he mentions inbreeding with neanderthals. Each of these topics has a greatly different aspect of speculation versus reproducible science, but the reader is left to judge the entire essay by what, exactly? Popularity of the scientist? His or her own political opinions and how they agree with those of the scientist? The number of papers published? The name of the publication it appears in? Phase of the moon?

Whatever it is, it has nothing to do with the eventual veracity of the statements made. But I’m not sure Mr. Smarter-than-thou reading it has any clue about that. And it’s an ethical slip that scientists need to stop making.

Once again, don’t get me wrong: I love pseudo-science. Proudly sitting on my shelf is the latest edition of “Secrets of Antigravity Propulsion” I get it down when I need an interesting diversion and laugh. I’m not saying it’s false, and I’m not saying it’s true. In fact, that’s what makes it so much fun to me: while we may have very good guesses about how true it is, we don’t know. As much as we want certainty, there are things we simply do not know. Non-reproducible phenomenon, such as some atmospheric things commonly called UFOs, could be anything. We’ll be speculating about this stuff for centuries. Fun stuff.

And yes, I’ll watch a good UFO show — although they are hard to come by anymore. Probably the best over the last few years was one the History Channel did on commercial pilot sightings. Trained professionals with lots of time in the air seeing unusual things that are cross-checked by ground radar? Very interesting. Fun material for speculation.

But I know these things are wishy-washy. I know that this is all so much BS. My problem is that when I run into statements from scientists that are just as flimsy — perhaps in a less spectacular arena — it’s not so clear.

Take evolutionary psychology, the idea that aspects of our personality exist because natural selection made them the best fit for previous environments which might not exist any more. Great idea — you can certainly hear the music and hum along — but at the end of the day you can make the argument that everything about personalities are evolutionary in nature. ADHD? Useful for surviving in the jungle. Adultery? Useful as a gene transfer mechanism. And so on. You can just throw darts at a dartboard of human behavior and pull some kind of theory out of the air for why evolution made it that way.

Not exactly much of a science. Personally I like Elvis sightings a bit better. Much more entertaining.

How about Anthropology? A fine field of science if there ever was one. Except it’s reached the point where it has deconstructed itself. Anthropologists are unable to make firm comparisons between cultures, thereby reducing everything they are doing to either observation or complete speculation (and sometimes politically-biased speculation at that). Come now, if you can’t compare things, where is the science? I’d argue that private property rights are a critical part of human cultural evolution — that there is a great ditch which is crossed when the private property rights of the individual are recognized. But I don’t think such an opinion is very popular among Anthropologists. Perhaps so. Beats me. It’s their mess. They should clean it up.

The flowering of this idiocy, of course, is when people have strong political or personal opinions and want science to tell them that other people with different opinions are idiots. That way they can appropriately look down on them. I was reading a great article in Mother Jones the other day about homosexuality, science, and politics. The tl;dr version is this: it’s not as simple as “I was born that way” or “It’s all a choice”. For some it might not be a choice. For others it might be perfectly fine to change sexual preferences because, well, just because. Sexuality is fluid. The idea that you’re gay or straight and were born that way and must be that way for life is an artifact of political discourse over the past century or so. But both sides want to use “science” to bolster their side and attack the other. Science, ever so flexible, becomes a big stick to hit the other guy with.

There are many other topics like this. Pick your favorite. The social sciences are full of various ideas that have consensus that are of unknown truthfulness. Economists use all sorts of complex mathematical equations to beat each other up — more and more in public forums, sadly — instead of trying to reach useful conclusions. Got a pet cause? Get you some scientists who are willing to be advocates. The truth doesn’t matter. All that matters is that you are the ones standing for science and the other guys are not.

Average Joe Sixpack is not an idiot, no matter what the surveys may show. He’s aware of this at an instinctual level and for him, it all becomes just so much noise. Show him a new cell phone that uses some tech: then he’ll be a believer. It’s a wonderfully pragmatic approach. Don’t talk. Show me. Show me something that is immediately observable. It’s not his job nor should he care about the fine differences in certainty between inflationary theory and certainty that the Higgs-Boson particle has been found. The ironically tragic part of all of this is that it’s not Joe Sixpack who is harmed by this mess: it’s the guy who loves science and reads as much about it as he can. The guy that can make the biggest difference in a democracy. He’s the real victim. Joe is already using a heuristic that’s imminently practical and workable.

Science needs a new set of ethics. It needs to be clear when reporting medical studies what types of methods were used, what was brought into the study a priori, the sample size, and how it was reviewed. All the data must be open. Scientists should be clear when they are speaking on matters of reproducible science — such as the absorption spectrum of CO2, the Greenhouse principle, the fact that man changes his environment, and black-body physics — and when they are totally speculating, like predicting the global temperature in 100 years. These are different types of information. (I hate to use global warming, but it’s in vogue.)

If scientists want to be considered in the same boat as the guys who think aliens built the pyramids, then fine, they should keep going the way they are going. But I doubt that. So I would encourage, no beg, scientists to adopt much stricter standards of disseminating information. Scientists should love science first. They should not be public advocates for any cause. It’s a conflict of interest. They should only perform open and reproducible experiments. They should encourage criticism. They should reject herd mentality and stop playing the game of trying to publish papers with the least amount of controversial material in them. Peer review should mean something. And the public — of any persuasion — should be brought into the creation and review process as much as humanly possible. Scientists may not like this; it’s a totally new way of doing things. But it’s critically needed.


In addition to the comments and links provided in the article, one of our readers (Lee Killough) sent in some suggested reading. I can’t vouch for any of these, but the list looks interesting. I thought it was worth including. Here it is, along with his summaries. Looks like great resources for further investigation. Thanks Lee!

Brief description:

Scientism and Values: A compendium of economists’ and philosphers’ take on the misapplication of science to questions of value, like conflating means and ends. Just because science shows the most efficient means towards certain ends, does not mean it has any special moral authority towards those ends. Mises.org has put the pdf online, but I have a hardcopy. Murray Rothbard was one of the contributors. I did not read all the comments but I noticed property rights were mentioned, so this book is essential reading.[PDF]

The Mismeasure of Desire: A legally-minded philosopher questions essentialist assumptions of gay genes and critically analyzes some important “gay research”. I pointed you to a radio interview he gave, but the book is available too. Whether homosexuality is a choice, is a separate question from its legal protection, but the two are often conflated. [Yes, and as the linked article above clearly indicated, this use science as a weapon has no impact at all on how we should treat each other. Different subject entirely.]

Against Method: Physicist Paul Feyerabend questions the arbitraryness of the scientific method. This book shook the foundations of science, around the same time as Kuhn’s The Structure of Scientific Revolutions. [Kuhn's book is considered by many to be one of the most important books about how science is practiced written in the last 100 years.]

Who Rules In Science: A moderate in the philosphy of science debates tries to get to the bottom of the debate and proposes realism as an alternative to the extremes in the debate. The Sokal Hoax and other controversies are mentioned.

The End of Science: Horgan surveys a lot of problems in science and concludes pessimistically that it can never have all the answers, despite the confidence of some of its outspoken proponents. (He later wrote a book which expands this idea further into neuroscience/psychiatry.)

Share
If you've read this far and you're interested in Agile, you should take my No-frills Agile Tune-up Email Course, and follow me on Twitter.

Startups: The Alternative Factor

If you ever want your own startup it’s time I told you something you might find uncomfortable: you are probably living in the wrong universe.

Most people live in a universe where smart people study hard to get a good job that’s steady and provides for them and their family. There are a set of rules to follow in life, even though nobody ever comes out and tells you them directly. Those who don’t follow the rules are to be pitied or politely shunned. Among academics there are one set of rules; among business people another. Among laborers there is yet another. We collect and share these rules in clans which take the shape of churches, civic groups, political parties, or professional organizations.

Most people will never have a startup. Worst yet, they have no idea what it means to form and run a startup. Their value system, their experiences, their social structure — everything about their life has purposefully designed them not to understand the startup world. They live in another universe.

It’s difficult to explain this at the level it needs explaining. They’re not wrong — in their universe they are existing and doing things as they should. Or as someone once said, they’re not even wrong. You can’t be wrong if you are acting in a natural way as best as you know how.

I’ve had a passion for startups for several years. I used to think that forming and running a startup would be like learning a new skill, say C++ or flying airplanes. Apply enough hard work, pour yourself into the subject matter, find some examples to copy, then work until you reach your goal.

How wrong I was.

Instead what I’m finding is that all (or at least most) of the things I have learned before I started working on startups have conditioned me to do exactly the wrong things in my startup. Want to write good, solid, bullet-proof code? Excellent goal — become a true craftsman. Only it has jack shit to do with startups and can easily prevent you from ever having a successful one.

I thought at first perhaps this was an isolated incident: that there were a couple of things that work one way in the corporate world yet another way in startup land. But then the examples kept coming. Want to build a vision of utopia — a field of dreams — for your users to come visit? Never works that way. Want funding so that you can develop for a year or so and build a solid product? Forget it — you’re clueless. Want to educate people with your startup about some cool social cause? Bad idea. It’s a startup, not a charity. Want to build a business and flip it? Wrong thing to be thinking right now. Want to take some new technology and do really cool stuff with it? Great. You’ve got a hobby, not a startup. Want to do some stuff to impress fellow hackers? Wonderful. You’re now in a popularity contest, not a startup.

It goes on and on.

That’s not saying that there are exceptions to each of the conclusions above — there are plenty of folks who make a startup around a social cause and do quite well, and there are folks who take cool new tech and make a workable startup around it — but they are the exception rather than the rule. There are even people who don’t believe any of what I’ve said who have crazy successful startups. These people are an example of selection bias: a person with random attributes who gets lucky and then associates their success with one of those attributes. Who knows how much harm these folks have caused simply trying to help other entrepreneurs.

The startup world is effectively its own universe: a place where the normal rules don’t apply.

Paul Graham (a person who studies startups and runs a startup “bootcamp”) said something once along the lines of “We used to think that you had to be smart to be a founder, but we found through trial and error that wasn’t a reliable indicator at all”

From what I can tell of their latest selection strategy, it’s somewhere between a beauty contest and just finding people who are too stubborn to give up and able to figure things out on their own. In a way this doesn’t solve the underlying problem at all — what makes for a good startup — but it works, at least more than other systems do.

The interesting thing I’m finding in my personal journey is that the more I learn about startups and figure out where I’m going, the less I am attached to the old universe.

For example, I was helping a large corporation out a while back. One of the projects they had completed was a huge repository of job instructions and process descriptions for how they develop technology. As an organization that built things, this made total sense: how could you build something without a clear set of rules? But as a startup person, I had two concerns. First, I don’t know what might be useful or not. (I also had some concerns about prescriptive process, but that’s not relevant for this discussion.) Second, unless we’re doing something that actually has value, we should stop doing it. That means measure whatever you do and stop doing things that don’t help anybody.

To me this seemed very natural — after all, I have a dozen ideas around startups. I could spend from here to eternity working on things I find cool. But instead of trusting my “coolness factor” or my own sense of self-confidence for something, I’ve learned to rely on the people I’m trying to help to tell me what I can do for them. That’s the only true metric I can trust — not my instinct or some huge body of knowledge I might have absorbed about some topic or another. But to the folks I was helping, this seemed odd, strange, threatening, and perhaps even rude. Why would we want to do things this way? Are you dismissing all this hard work we’ve done? Shouldn’t we be telling folks stuff? (even if nobody listens) How will people know what to do unless we provide them instructions? All great questions, by the way — but the assumption is that somehow we could sit around and reason what might be useful or not, then never check back with the folks we are actually trying to help. From a startup perspective, it’s just the opposite: assume you know nothing, form a hypothesis, then check. Repeat and rinse.

I could go on with these examples, but I’m afraid it’ll sound like I’m criticizing folks, and that’s not my intention. We just live in different universes.

I’ll close by sharing with you my comment to a teenager who made several hundred thousand dollars in his own business before being shut down by a legal threat. He had described his problems online, and received a sound thrashing for it by people who have never done anything like that in their life. Note the completely different value system I suggest than the “normal” universe:

Notes to people wanting to hustle and form a company/startup.

Note 1: Never underestimate the ability of the press and your fellow citizens to trash whatever you are doing. Because if you are providing value and making money, you are doing more than 99% of the people out there. They will punish you for this.

Note 2: Always expect a lawyer to call. The other 1% who are providing value have learned that the best way to keep making more and more money isn’t to innovate; it’s to use the political and legal system as a club to kill the little upstarts. Be ready for the club.

Note 3: People who end up making a lot of value usually don’t think anybody would much want what they have. First sale comes as a nice surprise. People who have grandiose dreams of killing the market usually wander off into fantasyland and never produce anything anybody wants.

Note 4: It’s all marketing and distribution. Know your customer and be able to get close to them. If you can do that, you can experiment with things until you find something that works. The market comes first, the product second. Great founders weren’t guys with genius ideas who went forth in some heroic journey: they were guys who were able to grok markets; to deeply understand the people they were trying to help.

Share
If you've read this far and you're interested in Agile, you should take my No-frills Agile Tune-up Email Course, and follow me on Twitter.

On Virtue

I was told last week by a follower on Twitter that I should put something in my profile. It seems that what I had, “You should not read this sentence” was not descriptive enough. So I wrote “Agile, Scrum, Kanban, Startups, and living a virtuous life.”

But that didn’t seem right. Seemed as if I were making some sort of religious statement.

So I changed it to “Agile, Scrum, Kanban, Startups, and living a classically virtuous life.

What’s virtue?

Continue reading

Share
If you've read this far and you're interested in Agile, you should take my No-frills Agile Tune-up Email Course, and follow me on Twitter.

3D Isn’t Really 3D

HTC EVO 3D cell phone



Recently I bought a new cell phone with “3D” capability, the HTC EVO 3D. By “3D”, the manufacturer means that the phone takes and displays two sets of images and videos instead of one. This allows one image for each eye.

Of course, there’s nothing 3-D about any of this, it’s just old-fashioned stereo photography. When I was a kid we were doing this, and movie-goers from the 1950s watched stereographic movies with those dorky red and green glasses. We’ve been there, done that.

My friend, who studied film in college, hates 3D movies. To him, it’s a solution looking for a problem. The use of 3D leads to crap on the screen. It’s gimmicky and it can even hurt your eyes.

And that’s not all: the cell phone itself got bad reviews, mostly along the same lines of “gimmicky.”

So why did I get one anyway?

Because, as I predicted, I’ve been having a blast with stereographic photos and movies. At first you do all sorts of hokey things, like having somebody stick a knife “out” at the viewer, or taking movies of stupid things (I have a wonderful 3D video of bananas) but then reality sets in: there’s more to this than gimmick, this is truly a new art form.

And I think that’s what the problem is for fans of traditional movies and photographs: the stereographic format is not “3D”, but it’s also not the old flat 2D art form. It’s somewhere in-between. So you can’t judge it by the old standards, but you also can’t judge it by the way we actually see 3D objects, either. Somebody said it was like when black-and-white media changed to color, and that’s mostly true. I think it might be more like adding one color, though. Is red cool? Sure. Can you have too much red for a certain scene? Most definitely.

Meanwhile, I’m finding some interesting things, like “a little bit will go very far,” or “color, contrast, composition, and depth all interact in new ways” Some things I’m not sure of, like “does the 1/3rds rule of photograph composition holds as far as depth as well?” I’m struggling with how to expose two images that may have different views — do our eyes focus and adjust to light identically? Or does each eye respond on it’s own? What’s the appropriate distance between each lens? Can you vary it wildly, or does a very narrow range work best most all of the time? How much difference in depth works with various subjects? Various lighting conditions? How do you handle when parts of one image over-expose but not the other? How about depth-of-field for each view?

I really wish we had 3D, like the famous Princess Leia hologram in Star Wars, but the only way that’s happening that I can see right now is through the use of large rotating drums and lasers. (Which means it isn’t happening any time soon.) Until that day comes, this is fun. I like having extra data as part of the media I record. I like having a new art form to explore. Even with all the work that’s been done over the decades, stereographic art seems an open field, at least to me. For the first time, I’m thinking about the benefits of purchasing a 3-D TV.

It also makes me want to look into prosumer equipment — something like a 35mm SLR except for stereo work. Fortunately, YouTube will let you shoot 3D from two different sources, say a couple of flipcams taped together, and then it will join it for you automagically. You could probably rig up something with a couple of Nikon SLRS, but you’d have to build the rig, and I think the horizontal offset would be too large. Still, might be worth playing around with.

As programmers and hackers, we’re used to blurring the line between new technology and new art. To many of us, programming _is_ an art, after all. To many of us, it’s art in the sense of being the skilled, predictable work of a craftsman. This is very similar to the attitudes taken by movie directors, I imagine. So this is fun in the same way a craftsman thinks of a new tool as fun, which I’ve found can be a much more satisfying and lasting kind of fun than other kinds.

I just wish they’d drop that idiotic “3D” name.

Share
If you've read this far and you're interested in Agile, you should take my No-frills Agile Tune-up Email Course, and follow me on Twitter.