Can an article – written by a professional journalist for a national news magazine – credibly claim that there’s a creativity crisis in America? Isn’t the act of writing the article itself creative? Doesn’t that mean something?

Well, no.

By way of Slashdot (here) I found a Newsweek article (here) that made the highly controversial claim that American children (6th grade and under) are less creative than previous generations and advocated project-based learning in the classroom as the “scientific” solution.

I really wish that people who write about science would try learning a little first.Really I do.

Let’s start with the setup: A longitudinal study by E. Paul Torrance (Wikipedia bio; obit; Books by Torrance on amazon) in which young children were tested for creativity, then followed for decades and their creative achievements recorded. The conclusion was that it was a good test, that people who scored high in creativity while very young, often went on to be highly creative adults. Longitudinal studies (Wikipedia definition here), by the way, are hard to do well but can lead to very rich data sets that can be useful for far more than originally intended.

So far we’re in “duh!” territory. The big take-away is that psychologists were thrilled and amazed to find out they could measure creativity. They may also have been wrong but we’ll get to that. Psychologists were also interested to find out that creativity and intelligence did not necessarily go together. Again, “Duh.” Anyone who’s ever seen an interview with Ozzy Osbourne (or any of a hundred others I could name – sorry Ozzy. You’re still great!) could have told you that. Of course, they are not mutually exclusive either. Frank Zappa proved that!

This test has been given lots and lots of times to students all over the world. A researcher, Kyung-Hee Kim (web page here) analyzed the data and determined that the test scores for American children have declined since about 1990. That’s kind of interesting but not, by itself, a crisis. I was disappointed when my grandson refused to wear the TV remote for a hat, even after I modeled it for him, but he’s still a good kid. He might even have some capacity for creativity. Just not in hats.

The Newsweek article, however, describes this decline in scores as a national crisis but, in fact, gives very short shrift to the question of how creative the kids are really or why their scores might be declining. It assumes without the slightest evidence that this is a permanent trend and that America has to do something right away. Then it goes on to describe the prescription: Project based learning in the classroom.

I’m not going to critique project-based learning. It sounds great when the advocates describe it. But the author’s agenda comes out, for example, in the use of the term “drill and kill” to describe rote memorization. “Drill and kill” is a pejorative term used by people who don’t like rote. The author accepts without question that it is bad. There is room for disagreement on this point.

Leaving aside issues of the relative merits of different teaching styles and the very contentious politics associated with them, I was struck by the implication that “science says” we have to start teaching American kids to be creative because they aren’t learning it and if we don’t change the schools to teach it to them, they’ll never learn it and the rest of the world will pass us by! The evidence to back this up just isn’t there.

One point that stood out to me was that the alleged decline in test scores (no reference is given to any validation of either the methodology or the conclusion, therefore I consider the point unproved) means there is also a decline in creativity among the very young. There is a vague mention of TV and video games as possibly suppressing creativity but no real explanation. But, if those things are really involved, how do we know that they aren’t causing children to express their creativity in a way the test isn’t very good at measuring? Would the test results change if the questions were given by bright cartoon characters on TV? What if the reason the kids are performing differently on the tests is because of increased education about “stranger danger?” Not trusting the tester will affect a lot of tests. Has anyone tested that?

What if creativity really is declining but TV and video games are not the cause? Ask any doctor how easy it is to cure a disease without a diagnosis. Treating the symptoms can buy you time but if you don’t have a good idea of the underlying cause, you may be completely helpless to keep the patient alive (Watch a couple episodes of House for a nice, if exaggerated, illustration of this. Most episodes have at least two attempts to cure the wrong problem, followed by the patient getting even worse, then House comes up with a miracle cure. Yes, it’s just TV. Take it with a grain or 12 of salt. I’m trying to make a point here, not prove a case in court!).

In this case, without any clear idea of a cause, we can’t even be sure there is a problem to cure. Psychological tests can be sensitive to cultural and, well, psychological factors that are not necessarily obvious to the researchers involved with them. And creativity, for all the research that’s been done, is still imperfectly understood. What if the (alleged) decline in creativity is due to changes in diet that change the balance of important brain chemicals? Or what if it’s just a passing thing, a statistical blip that will change in the next generation? This is why I get so annoyed with science “reporting” these days. Even a tiny bit of critical thinking would be better than what you get from most science related stories these days, especially any science related article with the word “crisis” in the headline!

I have nothing against project-based learning (though building an entire curriculum around it seems a bit much) and I’m certainly willing to believe that it’s a good thing to teach children to be creative. If nothing else, it makes for a fun childhood. But I don’t see a crisis and absolutely don’t see a basis in this for educational policy. I see opportunities for more research and for developing our understanding of creativity in childhood and even beyond. Let’s try that.

Continuing the subject of bad science (previous installment posted as Who writes this stuff anyway?), we have a study (described here) that explains that people with jobs requiring a lot of creativity often feel overworked and may find themselves sucked in outside work hours.

Sounds like an ordinary IT job to me!

Anyway, Like most science these days (or maybe it’s just science reporting, though I suspect it’s both) they seem to be unaware that correlation is not causation. What that means in this case is that it may NOT be that it’s the creativity required by the job that causes the result. It may be that people who demonstrate the capacity for creativity may get loaded up with work because, well, because that’s what it takes to get it done. Anyone who has ever supervised others knows that for a tough problem, you need someone who works hard, thinks sideways (I was going to say “outside the box” but that would be the opposite of creative, wouldn’t it?) and doesn’t let go of a problem just because the work day is over.  You want someone who will solve it for the pleasure of solving it, not just for the money or because someone who told them to.

When you find those (few) people, you treasure them. You also work them just as hard as you can get away with because there are more problems to be solved than good creative problem solvers to throw at them.

Again: Sounds like a basic (good) IT worker and an average IT job. I suppose other jobs may have similar characteristics. I just haven’t had one of those.

Interestingly, the study seemed to find that these creative people don’t mind having their work impinge on their lives outside the job. The hypothesis offered is that it feels good to them so what’s the problem? The thing to remember here is that, before starting the study, the scientists (and I use the term loosely since we’re dealing with sociology) probably didn’t know that the results would be so trivial. It would have been more dramatic if they had found a high rate of near-suicidal burnout among these creative workers but that doesn’t seem to have happened. But they couldn’t have known that before doing the study, so cut them some slack. I mean, how would social science types know that creativity is its own reward?

A more interesting question to study (take notes for future grant proposals) would relate stress to creative fatigue. If creativity is its own reward, does there still come a point where you’ve been so creative for so long (or so much in one particular area) that you just can’t think of new ideas anymore? How much stress do you feel then? How does it affect behavior both at work and outside it?

Insert joke about your favorite has-been author, producer, composer, programmer or scientist here. I’m going to take the better part of valor and not offer any jokes of my own.

The title of this post is something I once heard a newsroom editor yell (in slightly less family-friendly form) while editing the news. Being a sciency type myself, I am most likely to have that feeling when looking over the science news. The headlines reproduced below are from the last few days and I just couldn’t resist commenting on them.

New gene therapy proves effective in treating severe heart failure

You mean, there’s such a thing as mild heart failure? For the record, I don’t want that either.

Link identified between lower IQ scores and attempted suicide in men

The key word is “attempted.” The smart ones succeed.

Eyes of cattle may become new windows to detect mad cow disease

Yes. Especially when they’re red and glow. Stay away from those cows. (believe it or not, the article actually discusses looking for glowing bits in the retina, under a microscope though. Much less funny when you put it that way).

First Paper ‘Dipstick’ Test for Determining Blood Type

Wait – is this a repeat of the one about suicide? (Dipstick test. get it?)

Apologies may fuel settlement of legal disputes, study says

Well, more so than bullets, I suppose. I want to know what government agency actually spent money to find this out.

Visual system interprets sign languages

Ummm, isn’t that why it’s sign language and not , y’know, spoken? Also repeat comment above. (Okay, okay. The visual system discussed in the article isn’t the human vision system. It’s an artificial system. Couldn’t the headline have said, “device” or something?)

Study finds poker players using drugs to enhance performance

The other guy’s performance, actually. At least that’s what my father taught me. He said when in a hot game, try to drink less beer than the other folks at the table so you won’t play as drunk as they do. Also, repeat comment repeated above.

What are the most effective strategies for secondary suicide prevention?

Maybe it’s just me but I tend to think that the primary suicide would prevent any secondary. Not to mention pretty much everything else. Also, repeat …

(Note: I wanted to work in a dipstick or mad cow joke on this one but couldn’t think of one and still make it to dinner on time. Let me know if you think of a good one.)

Believe it or not, I don’t read the science headlines just to make fun of them. Sometimes, though, it’s almost mandatory. Usually the articles are not nearly as silly. Usually.

If I had a bigger brain, how many more languages would I be able to say, “The check is in the mail” in? Wouldn’t it be nice to be smart enough to answer the important questions (some of them may even be more important than that one)?

The nature of people with big brains has been a favorite science fiction theme for many years. I’ve seen it done in an old episode of Outer Limits and a much newer episode of Farscape, for example. In an excerpt from their book Big Brain, published online in Discover magazine’s December offerings (here) Gary Lynch and Richard Granger come up with some interesting thoughts on this question. I’ll say up front, this was interesting enough reading that I bought the book and really hope it’s not completely obsolete by the time I have a chance to read it (Do you think there might be a flaw in my reading strategy?).

According to a blurb about the book on Discover’s website (here) Lynch is a psychiatrist and Granger a cognitive scientist, which seems to mean they are doing a little more than speculating about the subject. The hook they use to get into it is the skulls of a pre-human species of hominid called Boskop (named for the place the skulls were found). Measurements of the skulls indicate that the brains of the Boskop people were roughly 25% larger than those of modern humans. From this, Lynch and Granger calculate an average IQ for Boskop of 150 which is 50% higher than the human average. But according to the excerpt they’re gone, now. Boskop became extinct maybe 10,000 years ago. We did not.

Were Boskop not as smart as the brain size calculation seems to indicate? Or was intelligence not an important thing 10,000 years ago? Hmmm. 10,000 years ago. Isn’t that about the time the last ice age ended? Maybe their brains overheated as the temperature went up. No, that sounds a little far fetched

Anyway, there are serious flaws in calculating intelligence based on brain size alone. The biggest one is that brain size is only one parameter in intelligence. Whales have bigger brains than humans but are not necessarily smarter. The convolutions in the cerebral cortex make a big difference. Roughly speaking, the more complicated the folding of the cortex, the smarter a species will be. This is why humans are (mostly) smarter than whales. [For a decent discussion of brain size see this article at HowStuffWorks.com]

Nutrition is an important factor in brain development, too and this may have a direct bearing on the discussion of the intelligence of prehistoric peoples. For children to develop to their best potential they need the right kind of nutrition continuously at the right time of their lives. Irregular meals or irregular protein (or sickness or injury or failure of stimulation or who knows what else) in the first two years of (human) life can cripple brain development. After that time, it can never make up the ground it lost early on.

Agriculture and animal husbandry helped humans to improve nutrition and improve brain development. If Boskop people did not adopt those things (and, during an ice age, it might have been harder, or at least different than it became later), they probably did not live up to their potential, despite their brain size. Evidence of their nutritional habits is almost non-existent now. Still, significantly bigger brains would probably correlate to a longer childhood and a longer period of time needed for the brain to develop internal connections. This also increases the window of time when a famine or plague could cause irreparable damage to child mental development. Not to mention more time to develop “issues.” Imagine being a teenager for twice as long! Any creature like that could maybe be excused for having a high rate of suicide, or of teenagericide.

Lynch and Granger raise some interesting possibilities for what beings with Boskop-sized brains might have been capable of if they did live up to their mental potential. For example, their brains probably would have stored much more sensory detail for memories than humans normally do. For them, every memory would include all the sounds, smells and feelings of the original experience. The visual details would be sharp and clear, whereas ours tend to be vague and even mutable. Human brains tend to conjure up only partial memories, and fill in the blanks with imaginary details. This is one reason why two eye witnesses to the same event can have wildly different stories (see here for interesting background on how fluid eye witness testimony can be).

Would super-detailed memories be an advantage in trying to survive in an uncivilized world? On one hand, lessons learned from experience would be more accurate when the memory of the experience was more accurate. Lynch and Granger point out that this does not apply just to individual memories but to whole sequences and hierarchies of memory. That is, connections between disparate things learned are more complex, allowing more to be learned in a shorter time. On the other hand, accurate memories would also be likely to be intense memories – and those can be intensely distracting. Maybe, though, that’s one of the reasons memory is so hazy. Maybe it works best when it’s just a guide and not a complete roadmap.

The uses to which a great brain are put make a difference, too. If these hypothetically smarter people never invented scientific method or mathematics or representative government, their view of everything would have been vastly different from ours. And there would have been a tremendous impact on their adaptation to the world – and to competing species such as Cro Magnon and Neanderthals. Even a truly brilliant race of poets, musicians or bartenders may not have been able to survive as a distinct race in that world.

There is a strong argument that the Boskop people, in fact, were not a distinct race, that the whole idea is a result of sloppy science and poorly defined categories (see here for a good rundown). That explains the seeming conundrum of a smarter species than ours that somehow failed to survive even though we did. They did not die out because they never really existed. The bones ascribed to them belonged to exceptional members of other species. This should not detract from the useful discussion of the implications of a bigger and correspondingly more complex brain than ours. It is worth wondering how a more complicated brain is really different from the ordinary ones we are used to? What
is the role of intelligence in history and human development? How is it tied to culture? What is the point of diminishing returns?

In science fiction stories, smart aliens either view humans as primitives almost unworthy of notice, or become god-like yet pacifistic teachers, depending on who is doing the writing. Imagining more detail about how a bigger brain works and sounds more interesting. At least, it seems to me to be much more interesting to study people who were at least potentially smarter than us, than people who were, comparatively speaking, kind of dumb. Right?

update 1/4/2010: John Hawks, who wrote the post debunking Boskop as a species (see above) has added another post explaining how bad the anthropology is in the book Big Brain. The new post is here. I’ve started reading the book and while it uses the fictional Boskops as a way of explaining concepts about intelligence, the book is not about Boskop. It’s about the brain and the nature and qualities of intelligence with specific reference to how intelligence can be modeled (top-down style artificial intelligence). I’m not far enough into it yet to judge whether the rest of the science is as poorly researched as the anthropology. I thought about leaving a comment on Mr Hawks’s blog to tell him what I’ve just typed here,  but couldn’t find a link or form for submitting comments. His loss.

I’ve discovered a new science. I’d like to say I founded it or invented it but there are already brilliant people doing interesting work in the field. They just don’t know they share a common field.

To begin, consider this story from Wired about a bizarre scientific paper on the development of a zombie plague. The paper itself (link) is a little dry, though it’s interesting if you can wade through the math. If not, read the Wired story. The basic idea is this: Some mathematicians (with quite a bit of time on their hands, apparently) developed the math to model the spread of a zombie infection. They concluded that, unless humans respond quickly with extremely large amounts of violence, the zombies win, civilization collapses and the human race is ultimately annihilated.

The paper assumes slow zombies, not fast or smart ones. It seems reasonable that both of those situations would make things harder for humanity, most likely. It also assumes that normal human replacement (birth and death) does not take place, since newborns eventually die and the newly dead are a perpetual source of zombies, which means the zombies win. The paper models multiple scenarios, including medical treatment for zombieism and the effect of quarantine procedures on the spread. Factors considered in developing their solutions include rates of transmission, the outcome of encounters (fights) and the effect on the spread of destroying zombies so that they can no longer spread the infection. In other words, despite the seemingly whimsical nature of the subject, this is real science.

The authors point out that, while a zombie apocalypse may not be likely, the same general principles apply in real world situations, such as a disease that has a period of dormancy. Presumably, this dormancy would mask the infection so that, like with zombies, it would be impossible to know who to worry about much of the time. Their model is more complicated than usual disease models, yet it has some real world applications. That’s interesting.

From another point of view, the important thing about this model is that it shows that the human race is not well prepared to withstand the onslaught of an evil force hell bent (so to speak) on causing its destruction. The model does not show what happens when human casualties become so great that the level of organization of the response is degraded. It’s an interesting question to consider whether steps could be taken to make civilization more resilient against chaotic forces. Maybe we’ll consider that question some other time.

Another piece of seemingly whimsical or at least offbeat research that fits in with the theme of a new science, was described in an October 2008 Scientific American article about modeling an evil human. This article describes a project act Rensselaer Polytechnic Institute to develop a computer program that could simulate an evil human being. That is, they have been working on an artificial intelligence that embodies all the traits we generally don’t want an AI to have (See my previous post The 99 Gazillion Laws of Robotics). An AI that intentionally causes harm (albeit, within the confines of a computer generated environment).

The idea of modeling an evil intelligence rather than just reading a book about psychopathy (For example the brilliant work of Robert Hare, such as Without Conscience) is fascinating all by itself. It involves problems of defining evil as well as building human-like responses into software, plus the need to develop some kind of environment for the AI to respond to. If this effort is successful, it will allow experiments in behavior that can not, ethically be performed in the real world. Think about it. If you want to know what stressor’s will cause Able to kill Cain (just for a change), you have the problem that if your experiment succeeds, someone dies.

No such problems exist in a purely computer generated world. True, you then run into the problem of checking your results against reality (would Able really kill Cain in an argument about the best thing to feed goats?) but at least you have data to try to check. Without that, you don’t have much beyond speculation and hindsight.

This is why I consider these two separate efforts to be a part of a single scientific study of evil. Maybe I’ll call it evilology (to distinguish it from evil science, which could mean any science misused). This is a science that has gone beyond merely observing the things that do us harm but uses the tools of mathematics and computer science to actively study those forces. This is a science that is capable of developing real, data based predictions about both behavior and potential responses (kill the zombies or we all die!).

During the height of the Cold War, the CIA had a huge Russia desk, populated by people who spoke Russian and had spent years studying Soviet leaders’ interactions, policies and anything else they could get their hands on. The general idea was Sun Tzu’s dictum that it is better to know the enemy, and yourself, than to know only one or, worse, neither.

The study of evil, in all its forms, helps us know the enemy, even when we are our own enemies. It also gives us a way to look beyond ourselves, at artificial intelligences (which are developing), unusual plagues and maybe even aliens. I’m giving serious thought to how to model an alien invasion (with or without genetically engineered zombies).

The advancement of this study out of the realms of history and psychology into scientific modeling is very much to be encouraged.

Robots are in the future. They are in the present, of course, but most people today don’t consider some preprogrammed floating arm on an assembly line to be a true “robot.” We learned what a robot is from science fiction and that’s what we’re all waiting for, often with dread (Don’t think so? Try googling “robot apocalypse.” Wait, let me try it first. 139,000 results. Hey, cool! T-shirts!)

Anyway, in anticipation of the day when robots are the smart, helpful servants/terminators of science fiction fame, lots of people have tried to come up with rules that robots could be programmed to follow to make everything better. Obviously the trend began with Isaac Asimov’s infamous 3 laws of robotics (Follow the link. I’m not going to repeat them here).

Asimov’s laws were pretty good, though his own stories involving them pointed out some flaws at least in potential implementations. Speaking as a programmer, believe me that implementation is an important point with any software. Give 2 programmers the same 3 rules to implement in a very complex system and you will find the two systems do not act quite the same. One programmer checks for compliance at the beginning of a decision, the other checks afterwards. Maybe they have different ways of checking, besides. The outcomes are often the same but there may be huge differences in some situations.

That different people approach the same problem in different ways is just a fact of life that may result in great differences between robot behavior, too. Anyway, because of these and other considerations there have been numerous attempts to update Asimov’s laws. For example a hliarious one I found a few years ago (and can’t seem to find the link for anymore) expanded the 3 laws to 10 (I think) and claimed to have patented them – thus ensuring no one would ever have the slightest interest in using them, even if they turned out to be perfect.

No set of robotics laws could possibly be perfect (see above) and personally I question whether such laws, themeslves are even possible. But it’s an important exercise to try to figure out how to make robots safe and controllable, you know, to avoid the robot apocalypse. An interesting attempt to update Asimov’s laws came out of Ohio State University recently, where some researchers reformulated the laws to make less sense and have even more loopholes than in the original version.

I tried to buy the original paper online but the system was down or something (ironic that the IEEE Computer Society not only charges for electronic documents but then makes it impossible to get them) but an article about the paper (here) reproduces these updated laws and that’s enough for now. According to this article, the first law as advanced in the paper reads:

A human may not deploy a robot without the human-robot work system meeting the highest legal and professional standards of safety and ethics.

This formulation of the law recognizes that the ethical and legal complexities of behavior, especially when robots interact with humans, probably can’t be summed up in a single law. However, it fails miserably as a law that robots can use, which is what Asimov’s laws of robotics were about. It also throws robot behavior into the realm of lawyers, which is absolutely not a good thing. Using this rule I can easily foresee long complicated EULAs requiring robot owners to hold the manufacturer harmless for any damage due to lapses in ethics or ethical judgment. We have gone from “Thou (robots) shalt not kill” to “It ain’t my fault if your robot accidentally offs you. Didn’t you read the license?”

There is no known Turing Test for ethical behavior. How do you certify that robots have ethics without getting someone killed? Maybe we need something like an FDA for robots. Yeah! That’s it! We’ll let bureaucrats decide!

On second thought, I don’t think I want a robot unless I’ve programmed it myself. At least then I’ll know where to find the kill switch.

I won’t go through the other two proposed laws because they are just as bad as the first. They sound like nice, reasonable statements of how robots should be made, if excessively vague, but they completely fail to provide any guidance for the robots themselves.

In the long run, the argument that, since no one can possibly anticipate every situation a robot will encounter, no rule or set of rules will ever be good enough at forcing them to behave the way we want them to, is irrefutable. Developing more vague rules and hoping someone implements them well is not much of a solution though.

That doesn’t mean there is no solution. As the implementation of robotic judgment proceeds, we may need to accept that, rather than giving them laws to obey, we will merely be able to influence them strongly. I have some ideas about how to influence robot behavior. I’ll probably write about them in a future post. For now let’s just say there’s a lot more research to be done.

Every time the power goes out, I have to re-install the driver for my wife’s printer. Every time, including (once again) today. I live in the country, beyond the suburbs into cow country, where it seems sometimes that the power goes out every time there’s a high wind. It doesn’t stay out for very long. Usually no more than ten minutes or so. That’s still enough to make me reinstall the driver. Oh! And VMWare player, which I use on my own computer. I figure that’s a bug that was probably fixed in a newer version but the last time I tried to upgrade, it completely hosed my network connections. After about 4 hours of fighting with it, I downgraded again. It works.

There’s actually a point to this other than just complaining about computers. I make my living (such as it is) with the things. Complaining about them is just part of the job. The bigger point is that, believe it or not, the computer age is still very young and there’s a lot we don’t fully understand about how to make software operate to our satisfaction. Things that should be easy aren’t always and benefits we think we should see sometimes don’t materialize.

Which brings me to the subject of a very interesting recent report (actually a thesis) summarizing studies of how students use software intended to help them learn to do arithmetic word problems. For a short article about the paper, see here. For  the paper itself, go here. Three studies are considered. The purpose was to learn about how students interact with educational software when there has been a breakdown situation. That is, when they get the wrong answer, what do they do?

Anyone who has worked user support or who has even been around people who work with computers can probably answer that without the need for an academic study (or three). What do students do when they can’t figure out the answer to a problem given them by a computer? They do what everybody does: They blame the computer.

Specifically, when students entered the wrong answer to a problem and the computer rejected it, the first thing they tried was entering the answer in a different way – such as using a comma or fraction rather than a decimal point – to see if the computer had simply failed to recognize the syntax they used. I wouldn’t be too surprised if, when that didn’t work either, they tried rebooting the computer and starting over from the beginning. It’s what I’d do.

In another study, the researcher noted that students didn’t just read a problem and try to solve it, they tried to understand what the point of the exercise was in the context of school. That is, they didn’t think “What math concepts are needed to understand this problem?” They thought, “What does the teacher (in this case an unseeen teacher who wrote software) want me to do?”

Not surprisingly, my interpretation of these results is a bit different from that of the researcher, who talked about framing concepts and social understandings. None of that really has anything to do with math. What was going on in these studies was that the students didn’t have good enough math skills to see right off why their answers were wrong (if they did, they might have gotten the right answers in the first place), so they tried to game the system instead.

This could show a weakness in the software. It wasn’t giving them good enough feedback to understand that their answer really was wrong. From the sound of it, the only feedback the students got was a yes/no type response. Not even a hint as to how to do the problem correctly. Without feedback and with rudimentary ability to do the work, the students can be forgiven for feeling lost (And I mean the students can be forgiven. Many of the irate users I used to field calls from should have known better, but that’s an entirely different rant). The software used sounds more like it was intended to drill the students in (theoretically) already existing skills. It was not true teaching software. At the current state of development of the market, there is probably little of the latter. It seems likely that to really teach, software would need a quality of artificial intelligence that does not yet exist.

Another possibility is that there is a problem with math education in general, or more accurately in the things that need to be taught for most people to learn math well. Several times when trying to help people with their math homework I’ve seen that they had the wrong idea about how math problems are done.They thought that, since they could figure out 2+2 intuitively (meaning, with little or no conscious thought), they should be able to do all math the same way. They conjured answers out of thin air and hoped for the best. They relied on what worked in second grade, apparently feeling that anything harder than that was more trouble than it was worth. This may sound like an attitude that would be found mostly in children but I’ve found it in adults, too. Probably no one ever showed them the beauty of math.

Both of these explanations would tend to show that there is still no substitute for a good teacher. There’s another article about the role of teachers in using software as an aid to teaching math here.

The paper concludes with the common sense suggestion that the claims of companies that sell educational software should be taken with a grain of salt and the use of that software in the classroom thought through very carefully. This I agree with. Computers are wonderful things but they are not magic bullets. Even with computers, math is still challenging to learn.

Someday computers will be better teachers. Hopefully by then they will be able to reinstall their own #^*&$! drivers.

Where I make snide – I mean informative – comments about stuff that caught my attention, instead of the usual long-winded ranting. I’ve been thinking about this for a while because often I see something, think of a paragraph or two, then get bored and wander away. But maybe sometimes a paragraph or two is enough! First up:

New Test For Detecting Fake Organic Milk

I couldn’t stop laughing when I saw this one. You mean there’s a problem with knock-offs of organic milk? Of course there is! Damn that supply and demand! People are willing to pay extra for a product they can’t identify, it shouldn’t come as a surprise when there are distortions in the market. That’s what art fraud is all about, after all. Tell a collector you’ve discovered a brand new Vermeer, then sit back and watch the bucks roll in because even the experts can’t tell the difference! (It happened during World War 2. See The Forger’s Spell)

Here’s a thought: If you can’t tell the difference, then maybe it’s not worth the extra money.  (The milk, anyway. The fake Vermeer’s in the book I referenced were terrible. The experts were idiots, which is a lesson we’ll go into at great length some other time)

Martian Volcano Could Harbor Primitive Life

So could my septic tank. How does this kind of empty speculation get to be news? That’s one of my big complaints about science reporting and even a lot of so-called scientific studies. This is nothing! Anyone can have just as profound a thought as this after a couple beers! So what?

Patrick McGovern, one of the scientists whose study of Olympus Mons (that’s the volcano referred to in the headline) is discussed in the article, should be given credit for saying that the implication that life is possible there is the kind of thing that goes at the end of the paper, meaning it’s not the actual point of what he was studying. The point was to figure out the history of that monster volcano, in order to understand Mars better.

So why did the possibility of life get into the headline? Apparently because whatever editor wrote it didn’t think extra-terrestrial volcanoes were cool enough by themselves. That’s weird.

Google’s Eric Schmidt looks down his nose at Twitter

My first reaction was: so? He doesn’t have to use it if he doesn’t want to.

Then I thought, what are the odds this means that Google is going to roll out their own version of Twitter in about 2 weeks?

Were gravitational waves first detected in 1987?

Wow. For those who don’t know, the detection of gravity waves is a BIG DEAL. Because until gravity waves are detected, we don’t really know if gravity is a wave (like we know sound and light are) or not. It makes sense, in the current state of physics knowledge for it to be so but that’s the thing about science: If you don’t have actual evidence, there’s still the chance you could be wrong. Even assuming it’s not wrong and gravity is a wave, detecting that wave would make it much easier to measure and develop better models of how it works.

Anyway, the story says that a physicist named Joe Weber claimed in 1987 to have detected gravity waves but no one believed him. The most likely source of the waves was a supernova that was detected around that time but other scientists calculated that gravity produced by that supernova would have been much too weak for Weber to detect.

Now a Pakistani physicist (they have nukes. You knew that meant they would also have physicists, right?) named Asghar Qadir says there were factors that no one was considering back then that modify the math. He says the whole world may owe Weber an apology (not that he’ll hear it – he died in 2000).

Here’s the thing though: Before anyone can prove Weber was right or wrong, they have to detect gravity waves again. The math can only be validated by actual measurement. There are efforts underway to do the measurement but they haven’t had any luck so far. For my book, the more time goes by without detecting gravity waves, the harder it is to believe that Weber did it in the first place.

The best part is that this article had a headline that made sense, rather than making me laugh or cringe.

At my job we encourage people to use Test Driven Development (TDD). The short explanation of that is that before you write a line of program code, write a test for what it is supposed to do. I confess I don’t always adhere to this. For me the rule would be more like, test early and test often. Testing is a skill and it can be hard. Testing first is also a skill. Like any skill it takes time to learn (and I’m getting better at it all the time!).

For years now, though, I’ve found that even my sloppy and less-than-perfectly skilled approach to testing seems to be too much effort for some people. At a previous job I routinely heard other people complaining that some system/server/software was broken when, in fact, their own code (or configuration or approach or whatever) was really broken. My colleagues and I would say something like, “Did you try it from a different computer?” And, as often as not, when tried on a different computer it worked fine. Reboot and try again. Don’t place blame before you’ve gathered the relevant information. Thank you kindly, call again soon.

See? It’s not just programming. Programming is just an environment where testing is measurable and has well developed tools.

There’s another idea behind testing that shows clearly when you write automated (therefore repeatable) tests for your code. It forces you to figure out your assumptions ahead of time. Often, you quickly realize you should consider other possibilities. What if the incoming number is 0 (zero, null, nothing, empty set) instead of what you expect? Dividing by it would be bad. That’s the fun thing about tests (at least if you have a mind like mine); I often write 3 or 4 tests for a function – 1 or 2 for the way it’s supposed to work, and a couple more to see if I can break it. You see, end users rarely understand the difference between good data and bad data. And even if they do, sometimes a cat walks on the keyboard (go away Kilroy! – that’s my cat) and inputs something you didn’t intend. So you have to anticipate these things.

But, as I said, testing applies to a lot more than just programming, as my “reboot” example above should show. I’ve been known to bring the philosophy of testing everything to areas that have nothing to do with computers. Think about it: You hear about a new TV show (let’s call it Dollplace just as a hypothetical example). You know the history of some of the people involved and you believe it will probably be a good show. But do you immediately start a fan club? No! You watch the show for a while first (and see an okay but dull pilot followed by a better episode that hints at improvements but still isn’t up to the level you want to see from such talented people).

In other words, you test the show.

Rumors are something people rarely test and should. Heard a rumor that your old friend (hypothetically) Eliza has been in training to be a ninja? Well, she has done martial arts before. It’s plausible. But there’s no need to decide right away if you believe it or not. Why not call her on the phone and ask? You might find out that it’s just for a (so far, mediocre) TV show and has nothing to do with actually being a ninja.

I also believe in testing whether the dog will like a particular brand of food, whether the car will get better mileage with a new kind of gas and even whether or not a politician supports policies he says he supports, even if he sounds sincere when he says it (Voting records are much more reliable indicators of belief than talking points). And all sorts of other things. Maybe that’s why I took to TDD in the first place. The logic seems a natural fit to me.

Testing is how we know, as opposed to just thinking we do. When someone tells you the network is down, you don’t instantly rush off and call Al Gore to complain. There are many possible points of failure for an internet connection. It could be your computer’s network card. It could be your router. It could be your computer isn’t even turned on. It could be a bad cable. So first, you run a test. Go to a different computer and see if you can get to the Internet. Can’t? Check your router. And so on. TEST possible causes before making a decision.

We don’t know how to test everything. If I suspect I have some rare tropical disease (apparently one that causes its victims to type maniacally), I call the doctor. And the doctor runs tests. Why would he do that when the web site raretropicaldiseasesyoumighthave.com says I have it? Because he wants to know, not just accept a rumor. Is that such a hard concept?

There were several things that inspired today’s testing rant. One was a workplace discussion with someone who is incredibly smart but who isn’t an extremely experienced programmer. I was singing the virtues of Test Driven Development – even if I’m not the greatest at it, promoting it is part of my job – and she was resisting. Basically, to someone who does not have the habit of thinking in terms of tests, the idea of testing code before it’s even written seemed silly. Like many programmers (not just beginners), she believed that running the code once in a while to see what it turned out would be test enough.

Maybe she doesn’t own a cat. I’ll have to ask.

(Note: She was not against using automated tests and I was able to get her to see that you should test small blocks of code as you write them, so as to catch errors before they get hidden in miles of spaghetti. It was the test FIRST thing she never quite got. I take progress where I can get it. My own testing has a way to go too.)

Another one of today’s inspirations was this video: Luke Francl Explains Why Testing Is Overrated.

He says, in essence, testing is only part of the equation and we shouldn’t rely too heavily on it, especially not to the exclusion of other important tools for ensuring code quality. He even concedes that there may be more than one way to write good code, so his views should not be taken as gospel (for more on his views, check out his blog post on the subject here: Maybe there’s more than one way to develop good software?). Every word is completely true. But I get the feeling that it’s been a long time since he’s had to convince someone that testing at all would be a great improvement. Note that in that last sentence by “testing” I mean writing test code that you can run repeatedly and, if the code still works, always get the same result. I don’t mean opening a web page up in your browser and saying, “Yup. Looks good to me.”

“Looks good to me” worked great for the developer of the Edsel, didn’t it? I mean the car. I’m not talking about some program that used that for its name (Because there probably is one. All the good names get recycled. )

Another part of the inspiration for ranting about testing was that I was working on my wife’s website (I do that once in a while, really I do) and needed a little information about a minor point of syntax. So I googled for it and quickly found what I needed. But here’s the thing: Out of maybe a half dozen blog posts I looked at that claimed to be tutorials NOT ONE showed how to test the technique being taught. I’ve found that’s the norm. Sometimes there’s a little blurb that says something like, “Of course in real production code, you would have tests, too.”

How would you like to see that in a tutorial on driving a car, or gardening? “Of course, in a production garden you would use water.” Really? How much? How often? Does it matter if it’s a little bit yellow?

How about doing your taxes? “In a real tax form you would also fill out schedule C for your home business.” Thanks. That’s good to know. Now how do I do it? And does it matter if I leave it out because there were no instructions? I’m sure the IRS will understand. It’s not like the numbers need to add up or anything.

How about a hydroelectric dam? In a real dam, you would use strength numbers from the actual materials used and calculate stresses for a range of volumes of water since, you know, it might rain someday. But explaining how to do that is beyond the scope of this blog post.

Testing. I’d say it’s not just for programmers – if I could get more programmers to do it.

Never mind me. I’m just ranting.

Is security a science? (I mean specifically computer/Internet security here.) Maybe the question is trivial but sometimes I wonder. The question occurred to me as I was reading a section on cross-site scripting attacks in Ed Skoudis’s excellent book Malware: Fighting Malicious Code, which is the textbook for a class I’m taking. Being a curious sort of guy, I tried it out. I took a prototype web site I had developed for my job and inserted some javascript into a text field, just to see if it would work. It did.

I had the advantage of knowing that I had not included defenses against such an attack in the code because it was a prototype intended to work through a problem, not an actual attempt to build a real live website. It was never going to see real life on the Internet. Well, it seems now that this may not be true. I’ve moved on to other things while that old prototype site has been handed to another programmer to build out into a more complete system. I guess I’d better warn the programmer that he has to include some kind of white listing or tag stripping in the data entry fields before it goes live.

Monday I guess I’ll add it to his backlog. It’s already on mine for the current project (at least, I hope it is!).

This sounds basic, but there’s a deeper question. The prototype site was built using Ruby on Rails, a wonderfully powerful web development language with lots of deep bindings backend magic. I ran a test because I wanted to be certain that Rails did not already handle these issues behind the scenes. I didn’t think it did but it never pays to assume that there’s something Rails does not do.

So what does any of this have to do with science? For openers, science is a process, not an end in itself. Scientific method describes how to enhance konwledge through careful and creative inquiry (Sounds nice when I put it that way doesn’t it? Don’t bother telling me about all the times it doesn’t seem to work. I know). My little test to find a flaw that I was reasonably sure was there wasn’t science. It could have been a preliminary. I could develop a theory of security architectures from that and begin a series of tests of the characteristics of those architectures.

If I had more time (and more skill. And enough money to be able to pursue these little whims. And … never mind), maybe I would create a fork of Rails and build some enhanced security systems in, just to see what would happen (and to spare myself the embarrassment of finding out I overlooked something I shouldn’t have). Would it make much of a difference? Would programmers take advantage of those features if they were added? Or would they turn them off to make life easier? Or would they complain that they were implemented wrong?

If the features were used correctly, would it make a difference, or would crackers (that’s like hackers, except us highly educated security types like to think it’s more accurate) just shift their attention to different attacks or different platforms?

Without solid measurements all we can do is speculate. That brings up another interesting point. If security is a science, I often think it’s a social science. We can measure the relative strength of security against certain types of attacks. We can theorize about the resilience of systems. But everything in security depends on the behavior of people. If the programmer programs in a security measure, or (like me) sloppily leaves it out because of an unreasonable belief it won’t be needed, how does that influence the behavior of other people – both innocent users and hack crackers, alike?

The interesting thing is that computer security, since it involves computers, is rooted in computer science. Data structures and logic and computational complexity are more the order of the day than research design with volunteers. Computer science education is heavily weighted toward the hard sciences, not the social sciences at all. This isn’t just the way the field is, it’s the way the people in the field think. In fact, having always thought of myself as a hard science kind of guy, I’ve tended to think of the social sciences as being not science at all.

Maybe I’ll have to modify that somewhat. Security as social science means being able to model the strengths of weaknesses of a complex system AND how the people interact with it. Like social sciences in general, I don’t think the processes or the results are very mature yet. Not enough rigor and not enough reproducible behavioral data. If anyone can overcome that, it should be us hard-thinking computer geeks science types.