Fog Creek Software
Discussion Board




Why some programmers are 2+ times more productive?

What do you guys think of this article?

The programmer's stone
http://www.reciprocality.org/Reciprocality

After spending several hours reading all the chapters, I still cannot exactly grasp the important points.

If any of you can elaborate it'd be greatly appreciated.

Mac
Sunday, December 01, 2002

I don't know why the tailing "/ro" were cut off.

The full path is

http://www.reciprocality.org/Reciprocality/r0/

Mac
Sunday, December 01, 2002

There is not much new under the sun.  These ideas have been around in different belief systems / religions for a long time.

From what I can see, there are many half-truths followed by faulty conclusions... or should I say non-sequiters.

IMO, there is a reason it is hard to understand... and the problem is not with you.

Scot
Sunday, December 01, 2002

Although "M0" keeps coming up and seems to be some kind of key concept, I can't find it defined or explained anywhere on the site.

The whole thing seems like a bunch of gibberish to me. 

J. D. Trollinger
Sunday, December 01, 2002

Hoo boy!

One of my favorite sites... breaking into the domain of jouel's word.

When World's Collide!

(It's like that episode of Seinfeld where George didn't want his girlfriend to meet his other friends.)

I admit this guy sounds like a crackpot but I read all his stuff a while back and I like it. The way he presents it makes it sound kooky, particularly with this 'M0 entity' thing, but I thing it's a sound hypothesis worthy of further exploration.

His central thing is that people are socialized to become addicted to dopamine, and hence to 'boredom products', meaning things being the same today as they were tomorrow. People don't like change. They are threatened by it. They will kill to stop change. Why is that? He says because we are socialized this way. THe process through which we are socialized is called 'the M0 entity' which he talks about liek its some sort of supernatural being with a mind of its own, but its really just a name to describe the idea that the process itself is self-perpetuating. I don't think he actually believes the process is a 'entity', just that describing it as one conveys more of the idea of how powerful the process is by virtue of it's addiction.

Anyway, so he proposes that when things don't change too much, you get happy due to elevated dopamine levels. He thinks you get addicted to this surge of opiates - just like a heroin addict. Children who ask too many questions of the teacher threaten the teacher's state of boredom and routine and thus must be medicated with Ritalin. Engineers who ask too many questions threaten the company's bordom state and are laid off or run out of town or put on Prozac. Politics as usual prevails. People are asleep. Are we attacking iraq? Well then we must always attack Iraq. Are we not attacking Iraq? Well then we must never attack Iraq. Things must stay the same. Internet chat rooms must stay the same. Anyone who threatens the internet chat room's stasis must be driven away. Blah blah blah.

Anyway, he suggests some ways in which his hypothesis can be proven correct or disproved -- he is not married to his conjecture, he just thinks it's interesting and explains a LOT.

As do I. I think it's a really cool idea.

Yes, the m0 thing is explained in there somewhere but I too had a hard time finding it. It took me a while to digest all the stuff and I got rather excited about it.

It seems fighting the M0 entity means not backing down from fighting BS and flushing your Ritalin and Prozac down the toilet. Maybe spending less time on the internet and playing video games. Maybe taking up swords and overthrowing the government. Who knows! It's worth looking into and I found it a really exciting idea.

X. J. Scott
Sunday, December 01, 2002

much of that website is pure crackpot, but a lot of it is very interesting.  good link. not sure if labelling people as "mappers" and "packers" helps the author's case, but his examples do seem to correspond to those folks who have a "clue" and those who don't, within my sample space.

.
Sunday, December 01, 2002

there are some interesting ideas, but serious problems with that author's "research." For instance:

From http://www.reciprocality.org/Reciprocality/r3/index.html:

"Little Known Fact: When he was a student, Neils Bohr's tutor was Soren Keirkegaard, the creator of existentialism, who personally bit him on the neck and turned his mind so far inside out it nearly went all the way around."


Soren Kierkegaard died in 1855, Niels Bohr was born in 1885.

intern
Sunday, December 01, 2002

What an excellent question. (but for sure a not a good website!!).

I have to admit, that Peopleware and a good number of books talk about how a top developer is MANY TIMES productive as the others on the team. However, none of those books actually makes an effort to understand why!

Is there any traits that developer has, or any methods that the developer uses that could analyzed, and trained / explained to other developers?

One could perhaps just dismiss the whole issue, and just assume it is a issue of intelligence. Just like some students perform much better then other students. Is there any reasons beyond just better intelligence that makes a top student better then lessor performing students?

It would bet that it is probably is just one of intelligence. However, there still may be some things that can be taught.

Is there a list of habits, and approaches that top notch developers use? By the way, the numbers for studies show that on software teams there is a MUCH larger variance then just two times. In fact, several studies show 28 times the difference.

That means one developer working for 1 month will produce 28 months worth of work compared to other developers. That is over 2 years!! In fact, this means often that 1 developer can output more work then whole rest of the team combined.

Gee, I certainly for one would like to find out if there is anything that can be learned from these kinds of talented star developers…

Albert D. Kallal
Edmonton, Alberta Canada
Kallal@msn.com

Albert D. Kallal
Sunday, December 01, 2002

I think being productive in programming equates to knowing lots of little shortcuts which add up to a lot of time saving.

One of my colleagues refuses to use regular expression and recently spent 8 hours building a routine to validate an email address in PHP.

I built the same code using a regular expression in 30 minutes. In this case I was 16 times more productive than my colleague.

Matthew Lock
Sunday, December 01, 2002

I agree with Matthew's point.  The more stuff you know about the more possible *quick* solutions you have at your finger tips.

I'm not a Java programmer but my Java friend who is very fast says that the main thing about Java is knowing all the libraries inside and out because more often then not there is already a library that already does what you need.

Obviously that's not all of it but it helps.

Gregg Tavares
Sunday, December 01, 2002

"It would bet that it is probably is just one of intelligence. However, there still may be some things that can be taught."

I wasn't aware that the world has settled on standard definitions of "intelligence." ;-)

I know many very intelligent people that completely suck at computer programming.  I do think the reciprocality guy might be on to something or other,  but since so much of what he writes is anecdotal, sketchy, false, or nonsensical,  you kind of have to discredit the whole enterprise.

The recent fringe psychology trend (which isn't so "fringe"...it is mentioned in many mainstream articles) of linking attention deficit disorder and autism with attributes necessary to be a good programmer are suspect at best. I know people diagnosed with attention deficit disorder, and they would not make good programmers. I know some very good/productive programmers, and they don't seem to exhibit any ADD symptoms. 

Probably 80% of what it takes to become one of those 28-times-as-productive programmers is just caring enough about your work to be that productive, and continually improving yourself so you can be that productive. I don't think that figure is unique to programming. There are surgeons who are 28 times better than other surgeons. There are writers 28 times better than other writers. There are pro basketball players (Michael Jordan) who are 28 times better than other pro basketball players. People have looked into this type of "genius" performance in the past, and have narrowed down a lot of the variables to simple "desire" and "practice."  Check out http://www.gladwell.com for an interesting article about "physical genius." 

intern
Sunday, December 01, 2002

It all comes down to training and talent.  Other endeavors have people who are 2-10 times as productive as the average Joe.  That's just the variability of the human being.

For example, most people cannot run a single mile in 6 minutes.  Yet there are hundreds of marathon runners who can run 26 consecutive miles at that pace or faster.

In basketball, some players average 2 points a game, and others average 20+ points.

Some people can play chess against a dozen people simultaneously and beat all of them.

There are math problems that can be solved by some students in a few minutes while others will take hours or never solve it at all.

What is interesting is that the most productive programmers (in terms of time to complete a given task) tend to write the least number of lines of code, and the number of lines of code per unit of time does not vary much from programmer to programmer.  The reason the superprogrammer is so productive is not that he/she is churning out ten times the lines of code per day; it is that they use far fewer lines of code (in a proportionately smaller amount of time) to accomplish the same task.

T. Norman
Sunday, December 01, 2002

"flushing your Ritalin and Prozac down the toilet."

Er. Ritalin and Prozac have, approximately, opposite effects. One decreases brain activity, the other increases it (roughly speaking). Although people with mild uni-polar depression tend to be easily distractable in a similar way as AD people, it's not for the same reason that attention deficit people are.

And it's probably not a good idea to suddenly stop taking either. In fact, it's the thing associated with all the hoo-ha with SSRIs. The instructions in them specifically say "don't suddenly stop taking them", the medical profession is fully aware that patients shouldn't suddenly stop taking them and tell patients this.

People keep suddenly stopping taking them, having rebound depression and killing themselves and somehow this is the medical profession's fault.

Katie Lucas
Monday, December 02, 2002

"People keep suddenly stopping taking them, having rebound depression and killing themselves and somehow this is the medical profession's fault."

Well alright then I'm glad it's that simple. Sorry there was a misunderstanding, let me clarify in case there was any confusion: "The medical profession proctices pure science and always places the heath of the patient ahead of profit and politics."

XJS
Monday, December 02, 2002

I think for the most part, people in the medical profession _are_ trying to put the health of the patient at the forefront.

Doctors and other staff at clinics or hospitals don't make any money off of pharmeceutical prescriptions (aside from your shady storefront doc selling human growth hormone and viagra). The people who want kids on prozac and ritalin are largely schoolteachers, and the kids' own parents!

You shouldn't really be glib about starting/stopping drugs like prozac and viagra. Stuff like that really does mess up your natural serotonin levels, and abruptly quitting can send you into a deep depression. the affect is akin to the low after taking ecstacy...you feel horrible...for a month! Kids, don't do drugs! (even the prescribed kind. ;-))

prozac
Monday, December 02, 2002

"The people who want kids on prozac and ritalin are largely schoolteachers, and the kids' own parents!"

This is true too, but unless I've missed something, the schoolteachers and parents are not the ones glibly writing the script.

"prozac and viagra...really does mess up your natural serotonin levels"

I agree with you.

"the affect is akin to the low after taking ecstacy...you feel horrible"

I believe you.

XJS
Monday, December 02, 2002

"This is true too, but unless I've missed something, the schoolteachers and parents are not the ones glibly writing the script."

Often in the case of ritalin, the school nurse writes the initial script. Prozac is prescribed by doctors. But keep in mind that prozac is typically prescribed to people who have serious depression problems, whereas ritalin is prescribed to kids who are merely being a bother.

prozac
Monday, December 02, 2002

[Anyway, so he proposes that when things don't change too much, you get happy due to elevated dopamine levels. He thinks you get addicted to this surge of opiates - just like a heroin addict. Children who ask too many questions of the teacher threaten the teacher's state of boredom and routine and thus must be medicated with Ritalin. Engineers who ask too many questions threaten the company's bordom state and are laid off or run out of town or put on Prozac. Politics as usual prevails. People are asleep. Are we attacking iraq? Well then we must always attack Iraq. Are we not attacking Iraq? Well then we must never attack Iraq. Things must stay the same. Internet chat rooms must stay the same. Anyone who threatens the internet chat room's stasis must be driven away.
Blah blah blah.]

That theory is wrong.
We are not programmed to hate all change and love all sameness. We are programmed to like a certain balance of change and stability. We don't like boredom, but we also don't like constant upheaval and revolution.
We learn from past mistakes, to some extent, and from past success. If a previous war succeeded, people might be too willing to get into another one, possibly forgetting the ones that failed because they are less recent than the successful one.
People are not asleep or stupid. Life is complex and change usually has to be gradual to work out well. There are times when sudden radical change is needed also.
But this guy (and I didn't read the article, only what was posted here) is just trying to justify his own bias against stability. But if he actually had to live in a world of constant change, he would quickly change his mind.

PC
Monday, December 02, 2002

"Often in the case of ritalin, the school nurse writes the initial script."

Thanks for this - I was completely unaware that that was even possible.

"But keep in mind that prozac is typically prescribed to people who have serious depression problems,"

I agree that this was Eli Lilly's idea of their small expected target market but I am skeptical that this is any longer true. It is quite common for family physicians to dash of a Prozac script without doing any sort of psychological evaluation beyond hearing that the patient feels blue, is tired, or has trouble falling asleep. Most people with mild depression taking Prozac or other SSRIs would be far better served by cognitive therapy or even diet and exercise, particularly since most of these drugs, although they do provide a benefit, do not provide a benefit that statistically differs much from that of placebos when subjected to the cold harsh light of he double-blind study. Is it worth the risk of the side effects I have to ask. Another issue is that bipolar disorder is different from depression and SSRIs should not be used to treat bipolar disorder - it needs to be screened out before giving a SSRI, lest we end up with a Columbine cocktail. Almost no doctors screen for this, are aware of the problems, read journal articles, or even scan the PDR's list of side-effects (which include psychosis).

"whereas ritalin is prescribed to kids who are merely being a bother."

I agree this is the usual case

XJS
Monday, December 02, 2002

Sorry, pc, but you should really read the article before posting.

I have read the reciprocality site quite a few times and have pondered for a long time about the theories presented there.

This thread seems to be rejecting the ideas because of small mistakes found in the article. Get the big picture: There is something terribly wrong with a lot of the worlds population. There are a lot of people making terrible mistakes, just because they are to lazy to think.

Ever noticed, how hard they try to stop you from thinking at school? Ever wondered why? Becaus thinkers just don't fit in!

Software Development has to do with abstractions, reasoning, thinking. Above, someone mentioned good programmers writing less lines. That's only half of the story: Good programmers write the wright lines! Why? Because they understand the problem at hand. Because they collapse the problem space and produce an elegant soluction.

But yes, don't take reciprocality too serious: Don't believe, try to understand. Try the concepts in the Programmers Stone. Apply them to your surroundings. See if it fits. Learn from the experience, even if you prove the theories false.

Oh, to actually digest the site could very well take you a month or two (ponder whilst in the train/bus/subway).

Daren Thomas
Monday, December 02, 2002

Ok I'll read it, but I am VERY skeptical about the idea that most people are stupid/lazy.
I am definitely a thinking person, by the way, and I acknowledge that most are not. But then most people have better common sense than I do, because I have to analyze everything, which can get me pretty confused.
People are different and most varieties have their worth.
Anyway I'll read it before commenting further.

PC
Monday, December 02, 2002

With regard to Prozac:

About a third of depressed people feel substantially better after taking an antidepressant.  Another third feel slightly better.  And one third don't feel better at all.

This is the same proportion of people who seem to benefit from placebos, and from faith-healing:

http://www.astrocyte-design.com/pseudoscience/index.html#thirds

Antidepressants are starting to lose their sheen.  A few years ago, it was rare to find studies that were skeptical of Prozac and its cousins.  Now such studies seem to come out every few months.  See, for example: http://tinyurl.com/rfu

Alex Chernavsky
Monday, December 02, 2002

Each time there's a new drug that's supposed to make you happy, almost everyone falls for it.
I'm starting to get tired of seeing the same thing over and over.
On the other hand, I do have a relative who had severe mental illness and is only able to function because of Paxil.
But the drug companies are trying to sell it to practically everyone. Parents take drugs that supposedly make them happy, and then try to convince their kids drugs are bad.
Drugs are bad, because the mind/body is a complex system and if one of the neurotransmitters is low you do not solve the problem simply by taking a drug that raises it.

PC
Monday, December 02, 2002

*sigh*  Most people who talk about ADD don't know what they are talking about.  There are a lot of basicly wrong ideas about ADD going around.

I could believe that an ADD sufferer would be drawn towards computing.  But ADD does not a good programmer make.  Far from it, in fact.  ADD can do a good job to prevent brilliant people from doing great things.  This also goes for programmers.  An ADD programmer has 20 little programs being written at once, none of them complete.  An ADD programmer misses all of the little details and writes code that's either buggy or doesn't work to spec.  An ADD programmer is too busy websurfing to code.  You get the picture here, so I'll stop.

w.h.
Monday, December 02, 2002

PC wrote:  "the mind/body is a complex system and if one of the neurotransmitters is low you do not solve the problem simply by taking a drug that raises it."

People are too eager to make a scapegoat out of some alleged chemical imbalance in their brain. 

A deficit of neurotransmitters is not to blame when you feel depressed as a result of a bad job, poor eating/sleeping habits, conflicts with your spouse, substance abuse, lack of exercise, etc.

Common-sense solutions are too often overlooked when treating depression.

Alex Chernavsky
Monday, December 02, 2002

Scope management cannot be underestimated. I've never been involved in a prject that couldn't have been chopped in half with decent scope management.

pb
Monday, December 02, 2002

[A deficit of neurotransmitters is not to blame when you feel depressed as a result of a bad job, poor eating/sleeping habits, conflicts with your spouse, substance abuse, lack of exercise, etc.]

I agree, and I actually think lack of exercise underlies most of the above. The importance of exercise is not really appreciated, even though there are lots of studies showing that mental and physical health is improved by exercise. People don't take it seriously, because the causal relationship is not understood. Why should exercise be good for you? So they disregard all the advice, because they don't understand it.
Anyway, the fact is our health depends on the voluntary muscles being used to some extent. Health in the US is a disaster, but people don't worry because we keep hearing how we're living longer than ever. The drug companies and the medical profession want everyone to think health depends on drugs and surgery, not on a natural lifestyle. After all, people used to exercise and they didn't live as long.
Well maybe people live longer now but it's a miserable existence when they have all those chronic diseases.
And I'm skeptical about the statistics on longevity anyway. They always give the mean but that statistic alone can be very misleading.

PC
Monday, December 02, 2002

You know, I don't think the medical field, most especially the general practicioners, have such a conspiracy that they would pass up good health for medication.  Most general practicioners would love it if you would go out and get some exercise.  They are still guaranteed business because it's only a matter of time before you hurt yourself while exercising, resulting in yet another doctor's visit for you.  Simple as that.

Most of the quick pill attitude is courtesy of the parents who want their kid to take a little pill and become a 4.0 student because they will settle for nothing less than perfection.

Really, what they need to do is functional brain scans on people.  You can tell pretty easily if there's a neurochemical imbalance that way.  The problem is that most people want a quick fix and are too busy to really figure things out.

The reason why depression and ADD and other problems are coming out of the woodwork is pretty simple.  People are actually researching why people get depresssed/hyper/etc and are trying to figure out what's going on, instead of just assuming that the person's whacked up in the head.  Behind the glossy veneer of day gone by is the dirty little secret of abuse.

The brain is pretty good at coping with stuff.  People tend to self-medicate.  ADD sufferers will seek out stimulants like cigs, caffiene, or illegal drugs because they do some modicum of help.  Even the clinically depressed can work themselves out of a depression for a while.  The foods that you creave might contain things like Tryptophan (boosts seratonin levels) or caffiene or carbs in general.

Some things you used to be able to get away with you can't anymore.  You pretty much have to get a college degree these days.  100 years ago, you could get away even without high school.  You could beat your wife and kids.

Some random points from up the thread.  Ritalin is a stimulant.  It works by boosting dopamine levels in certain lobes of the brain.  In a person with a true case of ADD, this has the effect of allowing certain areas of the brain that control things like attention span, distractability, etc.  SSRIs like Prozac work by preventing the uptake of seratonin.  Not the same thing as a stimulant, but both increase brain function.  An individual case of, say ADD, can be treated by a combination of a dopamine and a seratonin booster and both will help different things.

Why are some programmers 2+ times more productive?  Simple?  Some people are really sharp in the skills that coding uses, in the right job, with the right sort of self-management, with the right environment.  Everybody else is either not quite as bright, not cut out to be a programmer and wasting time compensating, spending too much time up late, etc.

took-the-blue-pill
Tuesday, December 03, 2002

[Most general practicioners would love it if you would go out and get some exercise.]

Yes, but since they don't have a theory about why exercise is healthy they don't think of it as the most important thing. There are studies showing a connection between exercise and health, but no theories about why (I have a theory, but never mind).
It's like many things in the health professions -- noticing correlations and connections but no coherent theories. Maybe it's because the sort of person who goes into software development, and who is accustomed to analyzing complex systems and finding out WHY things happen, is not the kind of person who goes into medicine.


[They are still guaranteed business because it's only a matter of time before you hurt yourself while exercising]

Medicine would not be very profitable if they had to wait for people to get hurt exercising. I never get hurt by it. There are smart and dumb ways to exercise, and even if everyone exercised in self-destructive ways, it wouldn't make up for the profit the drug companies make on depression and chronic diseases.

It isn't a conspiracy and I agree the patients want pills for everything and they're too lazy to work at their own health. On the other hand large amounts of money are involved so you can count on some degree of deception.

PC
Tuesday, December 03, 2002

took-the-blue-pill wrote:

"Really, what they need to do is functional brain scans on people.  You can tell pretty easily if there's a neurochemical imbalance that way."

This isn't true.  Mental illnesses are diagnosed based on symptoms, not on brain scans. 

In any case, the whole "biochemical imbalance" hypothesis is speculative at best, and downright wrong at worst.  The actual data is complex, partly contradictory, and not easily summarized into a simplistic theory of how "neurotransmitter X" is the culprit in depression, or schizophrenia, or what have you.  Of course, you wouldn't know this, based on the accounts given in the popular media.

Pharmaceutical companies and psychiatrists are largely responsible for popularizing the "chemical imbalance" theories of mental illness.  I'm not claiming that there is some kind of huge conspiracy here, but there are definitely people who benefit from the idea that disordered neurotransmitters are responsible for psychological distress.

Neuroscientist Elliot Valenstein has written an excellent book on the subject.  The book is called, "Blaming the Brain".  Here's a link to the Amazon page:

http://www.amazon.com/exec/obidos/ASIN/0743237870/

Alex Chernavsky
Tuesday, December 03, 2002

took-the-blue-pill wrote:
"The reason why depression and ADD and other problems are coming out of the woodwork is pretty simple.  People are actually researching why people get depresssed/hyper/etc and are trying to figure out what's going on, instead of just assuming that the person's whacked up in the head. "

(These are personal observations)
How much does the environment (immediate surroundings) play as a factor to those who "have" ADD/ADHD?
It seems to me that attributes of ADD/ADHD ( http://www.adhd-becalmd.com/selftest.php?id=17363 )are similar to behavior of a child who is not intellectually challenged at school.

Within the past 10 years or so there has been a shift in the makeup of classrooms in gov't run (public) schools.
When I went there were 3 types of classrooms; classes for the "slow" kids, classes for the "regular" kids, and classes for the "superior" kids (thats what it was called in NY).

Now (at least in NC) the make up of classrooms is a mixture of children (at least in the K-5 grades) and more attention is paid to the children that have a tougher time with the curriculum; leaving the smart kids; bored, working with an unchallenging curriculum.

apw
Tuesday, December 03, 2002

apw,

bingo.

X. J. Scott
Tuesday, December 03, 2002

I'm curious as to what the metric is for measuring which developers are "2+ more productive"?  It seems like every shop has there own metric.  For some it is developers who pound out code the fastest .  For some, it is developers whose code is rock solid(least bugs, fast algorithms, etc).  For others, it is develpors who apply the most elegant design solution (i.e UML, Design patterns etc).  Is there a standard to go by to measure developer productivity?

ST
Tuesday, December 03, 2002

====Begin quote====

It not surprising that it would take Forbes, the magazine that, for years, had as its slogan "Capitalist Tool," to point out that the way money is spent in the field of education is truly bizarre.

In their November 21, 1994 issue, an article by Peter Brimelow asks the question: "Would any management worth a damn put most of its dollars into its weakest divisions and starve the promising ones of capital?"

The next sentence answers the question: "Not and live for long."

Yet, as the article goes on to show in eloquent detail, that is exactly what is happening with funding for our brightest and most gifted children in the US educational system. According to the Department of Education, state and local spending on gifted and talented children is less than two cents per hundred dollars spent. And federal funding is never more than one tenth of one percent.

====End quote====

http://www.adhd.com/educ/smartfail.htm

J. D. Trollinger
Tuesday, December 03, 2002

In the studies I read about, they gave hundreds of different programmers the same specification, and took metrics such as the time taken to write the program, the number of lines of code, and the number of defects.

Not only did the fastest programmers write the least lines of code, but they also had the lowest number of defects.

T. Norman
Tuesday, December 03, 2002

T. Norman has the right answer.

The commonly quoted number 'ten times as productive' means ten times as many lines, ten times fewer errors, ten times faster and ten times more features *all at the same time.*

The studies have been pretty informal though, being comprised of programming competitions done on small projects over a weekend, but its the best we got.

X. J. Scott
Tuesday, December 03, 2002

Forgot to mention that the best (top 5% or so) are ten times better than average, not better than bad. So an average 'best' programmer is worth 10 average programmers (at least). An average 'best' programmer is worth 100 average 'bad' programmers (or infinitely more if the bad ones have negative productivity which is the usual case.)

You can save a sheetload of money by hiring the best programmers. Hire one at $150,000 and that's like hiring 10 good programmers at $15,000 a year!

But how do you find the best? Most companies don't have a clue.

And more importantly, why would they want to work for such a lame ass company? That's the real reason companies are stuck with average people.

There is also the top 0.005% who may by 100,000 times as productive as average -- they are a different matter entirely. One of these guys can be worth tens of millions a year at least, but few companies are intelligent enough to realize that. And that is as it should be!

X. J. Scott
Tuesday, December 03, 2002

One issue is that a programmer who is 10 times more productive than average, typically only makes 10% more money than an average programmer.

Rather than your top gun making $150K and your average guys making $50K, the scene is usually your average guy makes $68K, and your top gun can make $80K, max.

I'd assume this isn't the case for _good_ software shops, like Microsoft. But, if you are working somewhere else, that is how it usually plays out.

If someone could post a list of software shops in NYC that actually pay a "10x" programmer say, $180,000, that would be great. I'd love to start trying harder. ;-)

programmer.
Tuesday, December 03, 2002

My aunt, who was a schoolteacher for 40 years, likes to say that ADD stands for "Absence of Dad's Discipline" :)

Andrew Reid
Tuesday, December 03, 2002

>>There is also the top 0.005% who may by
>>100,000 times as productive as average --
>>they are a different matter entirely

...and they'd do in one hour what an average programmer would take 52 years to do! Where can I hire one of these magical developers?

Andrew Reid
Tuesday, December 03, 2002

100,000:1 is an exaggeration, but 1000:1 may not be so far off from reality.  But you can't hire them because those are the types who create their own products and run their own company.  Doom was written by only 4 programmers in a couple of years.  Put a team of 300 average programmers together and they probably still couldn't do it in 5 years.

And like I said before, it's not that the superstars write much more code per hour than anybody else.  They just get a lot done with the few lines they do write.  The whole Doom source code is just about 1.3MB, uncompressed and including comments.  A half-dozen Java classes written by average programmers can easily add up to more than that without doing anything nearly as much.

As long as companies only want to pay 10-20% more than the average programmer, and hire people based on "X years of experience in the Y language" and frame their interviews around useless questions like "what are your strengths and weaknesses" that do nothing to test your actual ability to do the job, they'll be stuck with nothing but a truckload of average Joes and Janes and worse.

T. Norman
Tuesday, December 03, 2002

Ha ha! You guys snapped readily at my baited hook!

Larry Wall is one such 100,000:1 programmer. Each hour of his time results in 100,000 or more hours of gained productivity in total. (Much of it shared in by others.)

These sorts of ratios kick in when someone creates some immensely powerful and useful programming language or such like C or Perl, or some reductionist operating system that is good enough like Unix. (Rather than inserting my usual bit about the millions of man hours squandered on Multics, I'll just let you all pretend you have just read a very convincing discussion of the subject for the umpteemth time.)

Maybe the number is even a billion to one, but I didn't want people to think I was simply being ridiculous. I'm quite serious about the 100,000 though, but there are very few people who are operating at this level.

X. J. Scott
Tuesday, December 03, 2002

ADD is not the Absence of Dad's Dicipline, nor is it what happens when a gifted kid is denied sutable learning opportunities.  Anybody who maintains that opinion is about as knowlegable as your average quack doctor.

Generally most people don't WANT to believe that there's something actually wrong with 3-6% of the US population and listen closely to a bunch of useless loudmouths who tell them that we just need to go back to the way things were in the 50s.  There, how's that for a conspiracy theory?

Your average ADD kid will be shuffled between a gifted program and a regular program.  He's too bright for the regular program, but give him the gifted program and he has problems there too, because he can't manage to turn in homework on time.  His parents may have already tried diciplining him, but your average schoolteacher would much rather blame the parents then themselves, leaving them with a clear conscience to teach you all sorts of stuff that you don't realize is wrong until College.

took-the-blue-pill
Wednesday, December 04, 2002

Blue Pill -
No one with with access to a library, a notebook, and a pencil can truly be denied a suitable learning experience. Lack of self-motivation on the other hand...

No one
Wednesday, December 04, 2002

" Maybe it's because the sort of person who goes into software development, and who is accustomed to analyzing complex systems and finding out WHY things happen, is not the kind of person who goes into medicine."

The author of my favourite book on trouble shooting  networks, said he found the procedure  easy because he simply adopted the approach his Dad, a doctor, used for diagnosing patients.

In general doctors, mechanics and  network engineers all need to use the same kind of logical processes. You are however trying to find out why something doesn't work, not why it does.

Stephen Jones
Thursday, December 05, 2002

*  Recent Topics

*  Fog Creek Home