Fog Creek Software
Discussion Board




Success in science

I read the _National Review_ article that Joel referenced in his blog entry.  I'm not going to comment on the main point, but the author is wrong when he says that success in science is highly dependent on "sheer intellectual firepower".

Sure, you need a certain amout of brains just to get into graduate school and get your PhD.  Beyond that, though, your success is far more dependent on your level of dedication and on your ability to get along well with the people around you.

(Caveat: I think physics might be more dependent on brute-force brainpower than the other sciences.)

I spent about five years hanging out with biologists.  The most successful biologists were the ones who practically lived in the lab, and the ones who were good at networking (helps for getting grant proposals approved).  The slacking brainiacs -- and the arrogant loners -- didn't do so well.

Alex Chernavsky
Thursday, October 17, 2002

Plus, the article furthers the "form a hypothesis then work toward it" they teach grade-schoolers. Most discoveries are accidental, or side effects of what-ifs, or deductive analysis. Sure, mini-steps are made on the hypothesis-then-test model, but not many of the big leaps are.

Troy King
Thursday, October 17, 2002

You guys might like reading "The structure of scientific revolutions" by Kuhn.  Here's a good outline:
http://www.emory.edu/EDUCATION/mfp/Kuhn.html

80min
Thursday, October 17, 2002

I've read Kuhn's book.  I don't see how it's relevant to the matter at hand, though.

Alex Chernavsky
Thursday, October 17, 2002

Then you haven't read Kuhn's book!  It's all about what constitutes success in science.

80min
Thursday, October 17, 2002

I didn't see this article as about science in particular.

I thought one of the main points is about the very difficult job of managing talent.  This is a constant topic on forums.  What attracts talent, what does it take to manage talent, what kind of management does talent respect: Certainly not always the nicest guy or best politician.

Another point is the risk of aiming talent in the wrong direction.  Or, more topical, how easy and dangerous it can be to squander talent.  In other words all the talent in the world may be on a wild goose chase.

tk
Thursday, October 17, 2002

For theoretical physics, in which I receive my Ph.D., sheer intellectual firepower is the most important factor determining success. Motivation and communication skills have secondary importance.

The article has a simplistic portrayal of the scientific process, but that's irrelevant to his main point.

The intelligence agencies need brilliant analysts and leaders, but cultural and organizational problems are equally important. For example, the rivalry between the FBI and CIA kept them from sharing information with each other. The traditional FBI focus on investigating crimes after they've been committed was an ineffective approach towards fighting terrorism. The FBI has a poor computer infrastructure, in part because they value agents in the field far more than desk jockeys.

CIA director William Casey was a very scary person, even if was as brilliant as the article claims.

Finally, after reading a National Review article about the importance of brilliant leadership, I can't help wondering whether that same reasoning applies to US Presidents.

Julian
Friday, October 18, 2002

To me, all he was saying was "we need superman".

Clutching at straws I think, if what he says is true then the ills of the various intelligence agencies are endemic and will take years to rectify, if ever.

Nobody ever expected such a large number of people willing to die for a cause. Suicide bombers are most "unscientific" because science tells us that survival is the paramount human instinct.

Mix poverty, religion, youth, lack of education, a cause, and a willingness to die and you have a potent weapon.

The best wepon of the west is to reduce poverty and educate, as well as military intervention.
It needs to be a multi-pronged approach, we need some smart people to remove the causes as well as stopping the effects.

We need some "human scientists" as well so that the solution can be worked from both ends.

Alberto
Friday, October 18, 2002

Any intellect above the "minimal requirement" (which IMHO is very low) is a handicap in science. People with more brains tend to be fairly critical of the so called "advances". This includes their own "accomplishments". As a result, they tend to publish less, if at all.

Most successful scientists are narcissistic egoists that truly believe they have pulled the world from their anus every time their mediocre grey matter comes up to speed with a basic outline of an idea that has been rehashed to the bone 100 times before.

Science died when genuine curiosity was replaced by greed and stage play. The world is shocked when they hear about the forgeries at Bell Labs. One guy is ousted, none of the others is held responsible. Renowned publications amongst which Science and Nature have the audacity to state that their review process is beyond reproach. The dirty little secret is that everybody in the scientific community knows that faking results is common and daily practice in the murkier fields, where independent verification is non-existent.

Just me (Sir to you)
Friday, October 18, 2002

I remember reading a paper written by this total asshole director of a lab.  He described a (in his opinion) brilliant young applicant who didn't want to start by coming up with his own theories, but checking the theories of others.  Being a sounding board, helping expand on ideas, checking things rigorously.  And he blasted the applicant for not being ambitious, that no one wants a fly on the wall scrutinizing their work.

I know his perspective, some teams can't handle such a person who isn't "producing" all the time.  But to blast him shows that people don't want an Erdös, someone who could be a great collaborator and eventually learn enough to make profound theories himself.  The Erdöses of the world just have to look forward to only owning a briefcase while others get the glory.

So to go back to the initial point, I don't think "science" today is ready for great minds.  The infrastructure in general is not there -- it's available for the mediocre minds because they make up most of the people.

Rant off.  I needed that. ;)

Tj
Friday, October 18, 2002

Tj, you bring up an interesting point. It's long been acknowledged that functional teams (the basic unit of research in industry) need more than one kind of worker/personality to do their best.

As a chemist-turned-multimedia developer, I can tell you that my productivity was much greater while working in a team than when working alone; unfortunately this aspect of science does not get much press. If you doubt the importance of teamwork, just review the Nobel prize winners of the last couple of decades.

The value lies partly in combining team members having varied talents. But there is another, equally important function of teamwork: collaboration. This takes the form of stimulating discussions and brainstorming with peers as well as the dreaded post-mortems or drawn-out meetings with management.

After all, how does one get support for new and good ideas, or figure out whether to not to trash poor ideas, without hobnobbing with your fellow wizards? The bull sessions, stimulated by present company as much as coffee or beer, turn out to be as much a part of scientific development as meeting with cohorts at international conferences.

In truly functional research teams, all members are considered peers, and all opinions matter; otherwise there is not much point in working within a team. Prima donnas notwithstanding, the work *and thinking* is done by all team members, not just the head honcho.

I suspect that the intelligence community needs to look at this model, instead of continuing with business-as-usual: that is, top-down management, with everyone in the organization looking to protect his/her turf while simultaneously protecting his/her ass (the model almost universally followed in our society).

Jim McCarty
Friday, October 18, 2002

[So to go back to the initial point, I don't think "science" today is ready for great minds.  The infrastructure in general is not there -- it's available for the mediocre minds because they make up most of the people.]

If history tells us anything, its that great minds don't give a shit what the anyone is ready for.

The article highlights an excuse for the intelligence mistakes, is that they didn't have enough smart people.  I'm sorry but that is just a piss poor excuse. Are they actually trying to imply that if you have enough brains in the intelligence area that they can predict the future? Give me a break.

His whole motive for writing that article is not to move us toward a better tomorrow, but instead to comment on political society and generate income from those comments (see the book he is seeling at the bottom). Everyone has an answer for how to change yesterday, it's tomorrow that baffles even the genius.

trollbooth
Friday, October 18, 2002

Nice try, Justme, but those of us who have worked in the field aren't falling for it.

David Clayworth
Friday, October 18, 2002

Great post, Justme, those of us who have worked in the field know you speak the truth.

Ed the Millwright
Friday, October 18, 2002

Humans are involved in all fields.  There is no harkening back to the good old days when science was pure and without ego, ambition, politics, or greed.  In fact those things are primary drivers.

The bright folks are no more or less altruistic than anyone else.  Never have been, never will be.

tk
Friday, October 18, 2002

Of course if I consider myself a successful scientist I have to recognise that someone else's answer is as likely to be valid as
mine.

David Clayworth
Friday, October 18, 2002

It seems to me that "passion" is a characteristic of anybody who has solved, cured or discovered anything of significance (atom splitting, penicillin, electricity etc).

Like all forms of work, the successful participants are passionate, and the rest of them are just working the cogs.

Scientists are no different?

Alberto
Friday, October 18, 2002

I don't think the comparison between the Russians in the 80s and Al Qaeda now is valid.  First of all, the CIA knew the Russians were a threat back then.  I'm not sure that anyone can say that the CIA knew before the attack that Albert Qaeda was a major threat to us.  Second of all, Big Al Qaeda consists of about three absolutely nutty people that are scattered around the world.  Russia was a huge country controlled by somewhat logical individuals that knew they would be killed in an all-out nuclear war.  The Al Qaedians don't care if they die.

The two situations are totally different.  Russia was big, easily identified, and a known threat.  Al Qaeda is small, hard to find, and was relatively unknown.

Additionally, the main threat from Russia was an intercontinental nuclear missile.  We had and have systems that would tell us if one of those puppies were to launch.  That sniper in Virginny could be Al Qaeda.  They just want to disrupt our society and kill as many of us as they can.  And, they have God on their side, so if they die in thier "cause," it's no sweat as they'll be getting BJs from virgins for eternity.

I'd much rather have been a CIA agent trying to stop a nearly broke Russia, than a CIA agent trying to stop someone who believes if they die trying to kill me they will spend eternity in heaven.

jar jar
Saturday, October 19, 2002

I thought Meyer's article was a highly irresponsible suggestion on how to improve foreign intelligence.  In a nutshell, his theory is this: come up with plausible prediction first, and spend your resources trying to find evidence to support this prediction.  Does anyone have any faith in our intelligence/criminal prosecution system to think that this doesn't create a Procustean bed to which all observations will be neatly cropped to.  His example of the US's view of the Soviet Union economy in the early 80s suffers exactly from his own suggestions: the CIA had a hypothesis of what they thought they should be proving, and were prejudiced towards evidences that supported this hypothesis, because they were primed to be on the lookout for it.  Turns out this hypothesis was incorrect, and the CIA wasted a lot of time trying to justify it.

Meyer's suggestion only works out if the hypothesis is correct from the get-go.  "Get it right the first time" is not a helpful methodology.

Scientists know that ninety-nine out of every hundred hypotheses will turn out to be crap, but the last one generally turns out to be a doozy.  Unfortunately, I doubt foreign intelligence can benefit from such a slow, laborious, and random process that depends so much on these rare and uncommon "a-ha" moments.

Alyosha`
Saturday, October 19, 2002

What was irresponsible was missing what, in hindsight, should have been more apparent.  In fact many, if not most academic Middle Eastern scholars, whom we wrongly presumed to be scientists or at least researchers, the experts, were tragically wrong.

Of course, all they risked is their reputations, which in fact seem mostly unscathed.  After all, history, the social and cultural sciences don't have the burden of the hard sciences.

tk
Saturday, October 19, 2002

In other news, IBM has made gazillion-dollar screwups.  And Microsoft is forecasted to be bureaucratic once Gates leaves.  It is impossible to sustain the energy of an organization through generations unless the need is always great.

Of course you want intelligent people, especially in an organization that calls itself 'intelligence.'  What company doesn't.  The trick is having intelligence up and down the ladder decade after decade.

Anyway this whole thread is about the survival characteristics of intelligence.  I think intelligence goes to wherever is most interesting and full of need.  Intelligence is a liquid, if flows around.

Tj
Sunday, October 20, 2002

*  Recent Topics

*  Fog Creek Home