You weren't meant to have a boss | Hacker News Reader

You weren't meant to have a boss

paulgraham.com

239 pointsdfranke298 comments

I used to work for a defense company that eventually got bought by a mega defense company. It was ~3000 people in various parts of the country and it was started by a man who deeply believed in the power of the entrepreneurial spirit. It had that free for all structure that PG says he'd never seen in a tech company. It was the wild west. We had different groups competing for the same government contracts. Managers and hackers alike got whopping bonuses for beating out other groups and they got to decide which contracts they wanted to bag. Entire groups were fired if they didn't bring in revenue. Fist fights, rancor and IP theft between teams were commonplace. But with all that they created some truly mind expanding tech for their time. They owned every angle of a highly lucrative market and showed no signs of slowing down... Until they got bought. The good news is, the fist fights, duplicate (and sometimes triplicate) efforts were stopped and everyone is one big happy family that hasn't done anything new in 7 years. Their market share is in free fall but they say it's fine b/c they are moving away from products in to large scale integration which is too boring to even type about. All the cowboys have gone to other places and I went to a startup. It was fun while it lasted.

I've heard some Amazon employees say something similar (but maybe not as enthusiastically) about Amazon. One person described their structure to me as "like terrorist cells": it's understood (and encouraged) that 3 different teams are all trying to implement the same thing, in completely different languages and with completely different interfaces. They might not even know the others exist, until they ship. Hopefully at least one will survive.

In some cases, more than one does. Amazon has at least 3 different data stores you can use, last I saw.

Theres a view in economics that you 'can not play market' within the boundaries of the firm. That is to say if there is an overseeing committee that can end a project prematurely then you are not really having market forces operate 'within' the firm and are doomed to failure. There have been a couple of companies that have tried to do this. It is documented in the following book by Nicolai Foss:

http://www.amazon.com/Strategy-Economic-Organization-Knowled...

Isn't it ironic that you say Amazon's strategy is "doomed to failure" at the same time that you link to a book for sale on Amazon?

The company(s) mentioned in the book still exist, but they no longer try to 'play market'. The more incentive you give people/subgroups to compete with one another, the more incentive they have to not share information(best practices) with one another. Managers see this and reallocate resources. Groups feel alienated that their resources are taken away from their idea to fund someone else's idea and then they go back to keeping their real ideas to themselves and/or starting a separate business outside of the parent firm. Innovation in the mentioned companies increased dramatically when the 'play market' strategy was initially adopted though.

Amazon has multiple paradox resolution teams competing to negate each other's efforts.

Entirely possible that my previous company would have never lasted had they not been bought. Perhaps those opposing internal forces would be too much to bear. But it did work for a while.

That actually sounds like fun. Friendly competition indeed :D

Being from the defense sector, I'd love to know what companies you're talking about. Although I bet you don't want to get into that. Anyway, in a week I'll be leaving my 1000+ people company for a 5 people startup! Interesting times ahead...

Yeah, for whatever reason it doesn't feel right. If you're headed to / live in the bay area, drop me a line and we can have a beer. As far as your transition goes, congrats! You'll miss the epic nature of your previous projects but you won't miss the bureaucracy.

Are you advocating fist fights literally or metaphorically? And if it's a metaphor, I'm not sure what the literal suggestion is.

It is important for people to have a fighting spirit. Otherwise they'll be controlled by those who do.

"Mediocre hires hurt you twice: they get less done, but they also make you big, because you need more of them to solve a given problem."

I would say Mediocre hires hurt you THRICE. Twice for the reasons you mention, and the third one is all the good folks leaving since they can't stand mediocre people.

One big company that seems to do well, at least from the outside is Semco in Brazil. Watch Ricardo Semler's "Leading by Omission": http://mitworld.mit.edu/video/308/ and read his book "Maverick": http://www.amazon.com/dp/0446670553. This is of course an anomaly since most companies don't work this way.

One of my favorite books.

And note, this company (originally) had nothing to do with hacking or technology. If it works for a manufacturer of heavy equipment, in a unionized shop, how much better can it work in software?

I have lived in Brazil for 11 years and never heard of this company. You can check out their website: http://semco.locaweb.com.br/en/

Something about that design and the icons makes me feel vaguely happy. That's a good emotion for an industrial equipment manufacturer to elicit...

Everything written in the article rings true. One effect he didn't mention is that the poisonous confidence-destruction and the lack of learning are self-reinforcing. It creates a cycle which discourages quitting.

I think, though, that the effect only hits ambitious, independent people. Those with less ambition or less independence seem to enjoy themselves, and do well. (There are also a few ambitious, independent folks, who aspire for the top of the corporate ladder and have a good chance of making it. They do well when their skills are matched to the needs of a large organization.)

I used to think that when big company employees "settled", it was a gradual process that snuck up on them. Not always. I've actually had a conversation with someone in which they told me about their conscious decision to settle -- that it was better to accept the circumstances and make the best of evenings and weekends. She became upset when I tried to remind her of the higher aspirations she had just the summer before.

[I work in a big company. I'm getting out in June when I'll have the funds to not use credit cards.]

Woo, you sound just like me. I recall a conversation in the canteen with a colleague recently where I said something along the lines of "... this place is so mediocre ... there has to be something better ..." and she replied with "well its just a job, you have to work somewhere, there are plenty of places worse".

She has clearly decided to settle!

I also agree with the less ambitious people doing well - I liken it to Seth Godin's book The Dip - those people are willing to lean into the organisation and just settle for its way. The more ambitious can sometimes see just how wrong things are and lean against it, hence doing badly in the company. If you are going to do well in a big corporation, you have to learn to play by its rules and settle unfortunately ... I haven't done that yet, but its a daily struggle!

Most people I know of from college don't even care to look elsewhere for a job. All they wish is to get into some big company and switch jobs every 2 years to another big company. Also for those people, getting into anything small is not good enough to be told among friends.

Regarding Ambitious, Independent people.

I tend to agree with you, but there's another class of people that doesn't fit your high ambition/low ambition paradigm.

I don't know how to label these people, and they confuse me greatly, because I'm not one of them and don't understand their motivations. These are people for whom ambitions and motivations are unrelated to "career choice"

I've met a great many people who work fairly-routine/moderate-pay/low-stress jobs, who have no desire to do work any more mentally taxing, and who profess to "live outside of work."

Now, some of these people are just not ambitious or independent. But some really seem to be extremely passionate, ambitious people. Some of these people are the "cause people" -- people that are heavily involved in political campaigns, in "causes," do large amounts of volunteering. They seem to derive their fulfillment from this work, and the paid job is a means to the freedom to pursue these causes.

Yet a different class of these people are the "family-focussed." They work, they may volunteer, but they appear to derive their fulfillment from... well, to be blunt, from raising kids and helping/coaxing/celebrating/watching their kids grow and succeed.

I'm not suggesting at all that these two other alternatives are in any way "lesser" (or "greater" for that matter) -- nor am I suggesting that other people aren't interested in, or don't do these things.

But, to generalize from just one example, (everyone does it -- or at least, I do...) though I don't do much in the way of volunteering or raising of a family just now, even if I do either of these at a future date, I don't think I would ever define myself by these things. I would identify and define myself more by the contributions I make in, for lack of a better name, the financial marketplace. Since this is a forum of "the startup crowd," I imagine that most of you define yourselves similarly?

Anyway, I went on so long, I forgot to make my point, which was that some of these people are perfectly happy in a job where they are not expected to be ambitious, so that they can save their energy and ambition for the part of their life that matters to them. I do not understand this, but I have recently come to notice its existence. I thought I'd point it out, since we're talking about ambition and lack thereof.

W.L. Gore and Associates (http://en.wikipedia.org/wiki/WL_Gore_and_Associates) is a big company that appears to try hard to avoid hierarchical structures. From a 2004 FastCompany article (http://www.fastcompany.com/magazine/89/open_gore.html):

"So Bill Gore threw out the rules. He created a place with hardly any hierarchy and few ranks and titles. He insisted on direct, one-on-one communication; anyone in the company could speak to anyone else. In essence, he organized the company as though it were a bunch of small task forces. To promote this idea, he limited the size of teams -- keeping even the manufacturing facilities to 150 to 200 people at most. That's small enough so that people can get to know one another and what everyone is working on, and who has the skills and knowledge they might tap to get something accomplished -- whether it's creating an innovative product or handling the everyday challenges of running a business."

Anyone who works for Gore care to comment on the actual practice?

I worked in the Gore HK office in the late 80's and even met the man himself. He was adamant about keeping groups small; small enough that everyone knew everyone else personally. New locations would start with about 10, and whenever an office/factory grew larger than about 50 they would split into two groups of 25. Another credo was that there were no "assistants" or staff positions. Everyone either produced product or sold product. Two roles. There were however senior and junior personnel, and decisions were made by rough consensus among senior personnel. However, all remuneration was openly discussed and agreed upon by the entire group in the context of production or sales targets.

I liked being there, although my role as a freelance systems analyst/programmer was not an authorised one.

"Work for another company if you want to, but only for a small one..."

My first job out of school was at IBM. This was invaluable not because I learned a lot but because I witnessed firsthand the bureaucratic inertia described in the article.

When I left to launch a startup, I never looked back or second-guessed myself because I knew I wasn't missing anything.

I'm also a recovering IBMer. It was awful... like 'Programming in a time of Cholera'

I'm a recovering SBCer (now AT&T, again?).

IBMers and SBCers compete over whose company was the basis for Office Space & TPS Reports. (i.e. the TPS Report zero that started it all)

I think that's crucial. I also worked for a huge software company for a couple of years before striking out on my own and the experience was worth it just to satisfy my curiosity about it, alone. The knowledge of what it is like and resulting lack of second guessing is valuable.

Lastly, I think this sort of testing, if you will, is worth doing in general. There were a couple of things I was really curious about trying as a freelancer. Each time, even when it didn't pan out was invaluable in terms of getting to know my own motivations more and refining what I liked, was good at and what I didn't. Sometimes you think you love something but you really don't know until you do it. You might like only parts of it but its worth doing it to learn that.

When I started reading this I thought it was going to suck. I thought he'd rant on big companies employees and make startup founders seem cool because of his background. When I was about half way through it started to change; what Paul said really made a lot of sense.

I want to be a researcher, but I also want to make somewhat of a living before getting a masters degree and stuff; so I can be completely independent. I thought joining a big company would be a good start because it would pay me good, I would be able to learn stuff and they could even help me out with my masters degree.

What hadn't hit me up until now is that, by doing that, I wasn't getting closer to my goal to be independent. I would just switch from the parent tree to the company tree; different pressures, but the same dependence. I start to think now that in order to be independent, to have the freedom of doing what pleases you most, you have to start your own company. You have to be your own means of survival. That's getting into the wild and finally living in a democracy.

I'm disgressing a little now, but it's because of this thing someone told me once. He made me realise that we actually live in a dictatorship most of the time. Your parents tell you what to do, then your boss tells you what to do; and you have very little say. In order to feel more like your opinions matter and that you can actually do something about the world, I guess you have to have the freedom to be able to say what you want, to code what you want, etc.

So, in the end, I was surprised. Very good article, as Paul usually writes.

Unfortunately, student loans constrain the freedom of action of many recent graduates. They have to have that income to make the payments, which can be quite high.

This of course favors upper middle class people as young company founders.

Yes, that is a real problem. I thought of mentioning this in the essay. The problem is the same as with visas: the government's definition of work is a job working for someone else.

I hope eventually they'll wise up and change the rules about student loans, because the intent of the rules about repayment is not simply to get the money back as soon as possible, but to discourage people from merely slacking after college. And on that scale founders are actually doing better than regular employees.

> The problem is the same as with visas: the government's definition of work is a job working for someone else.

It's not quite that. At least some governmental oratory states that entrepreneurship is what drives the American economy. Verification of 'work' is just a hell of a lot easier if it's for a known company.

I'm sure megacorp lobbyists in DC will get right on that, Paul.

Yes, this is far, far more severe than any credit card debt that that a student could possibly get into.

Related to this, I know several people who had to drop out of school recently because they hadn't built up credit and didn't have parents to co-sign on school loans.

Perhaps you should write on this some time, but I don't see how that could actually solve anything.

How much of a hindrance are student loans to founders? Or inversely, how much of an advantage is having no debt?

Well, in one sense, it's much easier having no debt. Imagine trying to found a company if you owned a house with no mortgage, had a vegetable garden out back, and owed nobody anything.

Realistically, what are your living expenses... how much money do you really need to buy 3- 6- or even 48-months of your time?

Now imagine that you own no property of any sort, and have large loan payments that you have to make every month. How much money do you really need to buy even 3- or 6- months of your time.

The opportunity cost of starting the startup is roughly the same (adjustments for interest between the riskfree rate and your credit rating notwithstanding), but the actual number of dollars that have to come from somewhere is quite different indeed.

Could YC somehow register as the corporation the founders work for? Then they'd have a "regular" job for loan repayment.

This made me think of something Peter Drucker wrote, and I think it came from the article summarized here:

http://knowledgeworks.wordpress.com/2007/06/05/article-the-c...

pg says that the modern firm is the product of modern technology, but this summary of Drucker's article points out that technology is also obviating the need for much of the "middle management" layers that pg criticizes as anathema to humankind's natural state.

I'm pretty sure somewhere Drucker also says that the only reason to keep activities within a single firm is because the transaction costs are lower than having activities performed by an outside firm. I think he also argued that as transaction costs between firms fall due to technology, it makes sense to "outsource" more and more, which of course we are seeing in today's economy. Which suggests that we should see more small firms over time.

Too bad Drucker is no longer with us. I'm sure he would have a lot more interesting things to say relevant to startups.

I wouldn't say the modern firm is directly a product of modern technology. The Roman army had the same tree structure. It's more that modern technology makes it possible to produce things on a very large scale, and that means it pays to have companies with thousands of employees, even though they are individually only a fraction as productive as they might be.

If I had to structure an organization of 1,000 people, here's how I'd do it: rather than tiers of management on the order of 1, 10, 100, and 1000, I'd just have two levels. At the top would be a small committee that controls allocation of money and from time to time publishes non-specific guidelines about where they'd like the company to go. One person on the committee is in charge of hiring salesmen, who once hired are basically unsupervised and work for commission. The rest of the company is a bunch of small engineering teams of up to 10. All functions other than sales and engineering are outsourced. The most senior member of each team has hiring/firing authority for that team. He gets to decide the size but 10 is the max. These generate their own ideas and speak to the controlling committee only for budget requests. Salary is not part of the budget request: everybody in the company gets the same salary. Bonuses are awarded to teams that implement profitable ideas. More money goes to bonuses than to salary. If a team shows a pattern of underperformance, the committee fires the entire team. If this happens, they get 30 days notice that they're going to be fired, and they can use that time to find jobs on other teams within the company.

If I then had to scale this to 10,000, I still wouldn't add another layer of management. Instead I'd just add more teams, and expand the controlling committee to a confederation of committees, which all draw money from the same pool but have specialized expertise. When teams request budget, they approach the committee with the best understanding of their proposal.

The 1,000 person model looks a lot like a bunch of founding teams and one VC firm. The 10,000 person model bears some resemblance to the federal government, which is perhaps why an organization the size of the government manages to function at all, however poorly.

One very serious problem with that idea. Team X isn't insulated from lawsuits that Team Y gets hit with (or contracts that have the potential for massive costs, for that matter).

Suppose Team X isn't doing well. Probability of getting fired is 50%. They decide (independently) to do Z, with a 90% chance of increasing their team's profits (and not getting fired) as well as a 10% chance of major lawsuit that will crush the company.

This nearly doubles Team X's survival probability, while exposing the company to a massive risk. This is a classical example of agency costs. And this is why big companies are usually pyramid shaped.

Could you structure the company (legally) such that each team was an independent company that the parent company holds a majority stake in? This way if an individual team got sued, the parent company's liability would be limited to it's investment in that team, and sibling teams would be isolated completely.

Of course, I have no idea if that is how liability works, and even if it was, the accounting situation for the overall company would be a nightmare.

You also just described the major problem with the government artificially allowing such a thing as a limited liability (I'm talking about legal, not debt) corporation.

This has little to do with limited liability. In fact, the problem here is that groups within the company do NOT have limited liability (from each other), thus necessitating some oversight.

The only role limited liability plays here is that investors in this company to lose only their investments (but not their house and car) should the company die.

If I had to structure an organization of 1,000 people

You might also want to read this: http://www.amazon.com/Maverick-Ricardo-Semler/dp/0712678867

This is why I love the HN community - not one but two people recommend the same book - that's a huge vote of confidence that forced me to take a better look at it.

Just based on the summary, it sounds like he has a lot of good ideas and one bad mistake: profit-sharing is socialized. So if the company does well then the employees do well, but individual contributors would have trouble getting rewarded.

It may not be as bad a mistake as you think. There's some experimental (if unintuitive) evidence that rewarding individual contributions in organizations is harmful, even to that individual's motivation. The extrinsic motivation extinguishes the individual's intrinsic motivation.

Many high achievers have an attitude of "achievement for the sake of achievement." Individual rewards crush that spirit.

Joel Spolsky wrote about it on more than one occasion. Here's one link: http://www.joelonsoftware.com/items/2006/08/09.html

See also:

A) Punished by Rewards (Alfie Kohn)

B) Herzberg's Motivation-Hygiene Theory (Two Factor Theory). The idea is that a very low salary can lead to dissatisfaction, but increasing the salary beyond the required minimum isn't one of the things that increases happiness. Link: http://www.netmba.com/mgmt/ob/motivation/herzberg/

There's a huge difference between the sense of ownership that comes automatically from small organizations and the 'incentives' that a paternalistic committee chooses to bestow on a few in its largesse, while bleeding motivation in all the other ways large organizations do that PG talks about. I don't interpret any of those studies (including Herzberg's in the sibling post) to mean 'socialism isn't so bad'. Yes it is so bad. But that doesn't mean you can't implement capitalism poorly.

Capitalism isn't about profits, it's about motivation.

That's not entirely true. Workers set their own salaries, subject to the review of their colleagues.

If you think you are awesome, you can give yourself a hefty raise. If your co-workers don't agree though, in a few months you'll be out on the street.

I think it makes sense to share profits equally, since the whole point is that the result was achieved collectively. Salaries can be all over the map.

Semco sounds just like Nucor and Gore (of Gore-tex fame). There's probably a few companies like this around.

Deleted comment

> The profitable ones can receive more funding, if needed.

Who decides if it's needed?

I was thinking your 10k person model sounds like how research funding works... which seems to be a moderately functional model.

With your 1000 person model, I'd do away with the boss (in the sense of hiring and firing). With that few people, you can all just work together fairly easily, with natural leaders emerging for different projects (and hiring/firing decisions can be made by the team as a whole). There is a factory in NC that makes aircraft engines that works this way, and it is quite successful. The teams are self organizing internally, and send representatives to deal with other teams and plant management.

In practice that's how I'd expect/hope it would work. But while ideally the groups should act by consensus, there has to be some means of making a decision if no consensus can be reached. For groups that small, a benevolent dictatorship tends to work better than a democracy.

All in all, not a bad framework, though I expect much tinkering would be needed to optimize it. It’s worth noting that almost every large company strives to be “entrepreneurial”, although some are better at implementing this than others. The world’s largest private company (Koch Industries, I believe) prides itself on “market based management”, though I’m not sure what this means in practice. Koch is also a big funder of libertarian causes, which gives some hint at his mindset.

I know of one huge company that does exactly this. Except instead of outsourcing "all functions other than sales and engineering", they outsource nothing but sales and engineering. :)

The idea that there is a certain "natural" size to groups of people reminds me of the Monkeysphere essay:

http://www.pointlesswasteoftime.com/monkeysphere.html

The size of a monkeysphere is estimated to be 150 people, which is quite a bit larger than the 10-person group Paul advocates, but I guess you need to know at least a few people outside of work. :)

I wonder if the problem with large corporations is that you don't care about the other people in them or that you feel the other people don't care about you? Certainly the people in the highest positions can't care about everyone under them -- assuming they can only care about 150 people -- so perhaps that is the limiting factor of effective size.

The "monkeysphere" is more commonly known as Dunbar's Number (http://en.wikipedia.org/wiki/Dunbar's_number)...though Wong's essay is much funnier than the Wikipedia article.

Dunbar's number is a limit on the size of a society, not on the size of a group that can work together to complete a task.

So one might say that an optimal size of a business is 150 - family members - external to business friends - significant acquaintances. With various employees being more or less numerically involved in a company.

150 is the maximum, not the optimum. It's just a really big safety margin.

I am a 27 year old programmer and have worked for two small companies ( < 20 employees ) and for one company of 200+ employees. The only thing I can say about working for the large company is that it felt like breathing through a straw. Go work for a small company or as Paul Graham suggests, go start up your own. Leave the big companies for the burnt-out folk with kids and debt.

yet another 27 year old programmer hear. i've worked for google for the past 3 years. after saving up a bunch of money and working with a lot of very smart people on two very different projects, I'd have to say I'm happy I chose to work for a company :) That said, there is no substitute for being able to do exactly what you want. I get my kicks in that regard programming on my own (reading SICP, toying with scala, etc). But guess what, at a good company, so does everyone else. My smart co-workers all have interests in and out of work, and just being around them and talking to them day in and day out (while eating really good free food) is a great way to learn a lot and get pointers to places to start learning more on your own. We're not all brain dead code monkeys, believe it or not.

Ultimately though, the money I've saved will afford me the opportunity to strike out on my own. I certainly don't knock small companies, or starting your own. I just think large companies (OK, at least google) can be a great way to start your career before moving on to other stuff.

I'm 26 and just passed my fifth year at a very large company (18000+), with four years in IT and the past year in Software Engineering. I've been looking for an oppurtunity at a smaller company, since my current one is stifling my ability to learn and be productive.

The issue I'm finding is that small companies are typically looking for someone with more experience than I could currently offer. Has anyone else come across this issue? It seems to me (and others where I work) that we have to somehow pay our dues and get experience at a large company before we gather enough experience to become valuable assets to smaller companies.

One route you could take would be to look for a startup with young founders. They won't insist on experience, because they don't have it themselves.

Another would be just to look for a good startup. All a good startup cares about is how good you are, not how many years it took you to get that way.

Its also possible I haven't been looking in the right places either. From what I've noticed, its hard to pinpoint jobs at small companies since they don't seem to advertise as much as the larger ones.

However, Jamie from snaptalent was nice enough to email me and pointed me to their job ads page. Unfortunately there's only one company on the east coast listed so far, but I'm hoping once they get rolling with more job ads that my chances for finding a small startup or company to work for will increase.

You can also try:

http://jobs.37signals.com/ (usually webapp jobs)

http://jobs.joelonsoftware.com/ (not always small companies)

We (over here at snaptalent) just put up joinstartups.com We'll keep this up to date with all the startup jobs (and only the startup jobs) from our network

Hey there, Sumon from Snaptalent here. Expect over the next few weeks for us to rampup the number of companies on here.

Who else on the east coast would love this? Some great startups there that we can help get for you

Email me: sumon [at] snaptalent [dot] com

Something that (stupidly) never occurred to me before I kicked of a startup - if you want to get a job with one, go along to networking meetups (e.g. opencoffee) and you'll soon find some opportunities.

Personally I'd much rather work with someone that is smart and never programmed before than someone that is a terrible coder with 5 years of experience in buzzword-of-the-month.

Last time I looked for a job I actually crossed out any place that wanted more than one year experience with anything. I guess I'll never be a Senior Java Engineer or a Senior .NET Architect or a Senior Oracle DBA, but is that really a loss? Asking for five years experience in some technology is like actively seeking out people that are content to actually do the same thing for five years. Any small company that wants that is going to get trounced by a competitor that learns new tricks.

I am also a 27 year old programmer working for a very large company well over 10K employees. Unfortunately, doing well career wise is more about politics than programming.

I have never worked for a small company so I cannot comment on that, but I hope for sure that there is something better than what I have experienced so far! Just have to pull my finger out and get out of there, or start some sort of startup ...

if you're serious about getting out check out some of the jobs we have up at http://snaptalent.com/ads.html There are some YC companies on there, but also some other really cool startups.

I can't see any ad's at that URL ... maybe something is broken?

One thing thats kept me in my big company job is that I am in a part of the UK where there isn't a massive choice of places to work - there are other big companies, but few if any startups :(

Where in the UK are you? You can always quit and move if theres nothing available

I was actually wondering if there was a page where I could just view the list of jobs being advertised on SnapTalent... seeing as it's a self-selecting group of advertisers, that has very high value to me.

Feature Request: Show me the company, job title, and city, please?

As another 27 year old programmer, I 100% agree. The only real benefit I think I derived from working at a company (both large and small) is getting a good idea of what works and what doesn't work for large scale projects (which I wasn't exposed to in school). And even that lesson was learned pretty quickly, within 3-6 months. All returns were diminishing exponentially from that point on.

That same lesson could be learned in the same time frame in a start up - you'd just end up with some throw away code after a pretty short period of time, which isn't a big deal at all when it comes down to it.

I'm having a very similar experience working for a large corporation and I've made plans to extricate myself and go back to grad school. I'd be careful about associating "kids" with people who are burnt-out or who just want to crawl into their little corner as a faceless cubicle dweller. I'm 26 and have a wife and an 11 month-old son, both by choice. There is joy and fulfillment to be had in a family that commercial success cannot replace. It's interesting to see how many people assume that it's not possible to have a family as well as a stimulating and challenging career. Do not make these assumptions. I will prove them wrong.

There are small companies like one of the startups our host sponsors, or like those Joel Spolsky runs, and then there are those which languish under the them of a boss who is living a good life while 1 to 5 or so programmers slave over ancient technology and insane customers.

So qualify your employer carefully before you go to a small company. In particular, if they are using old technology, run away, nothing they do or say can be trusted.

> boss who is living a good life while 1 to 5 or so programmers slave over ancient technology and insane customers

I find this is most often the case with small companies that aren't run by programmers or ex-programmers, although that isn't always a hard and fast rule.

After reading all the nasty comments about this article on reddit and other blogs, I can only come to one conclusion: none of them bothered to RTFA. I get the point totally - humans don't work well when in large organizations. I've gone from massive companies down to a organization of 1. I'm not as stable, I'm more stressed, my income is lumpy. But I have ideas. Buckets and buckets of ideas. I have so many I have to write them down in a book to stop them from clouding my mind. And I have learnt - so much. Not so much about programming - I already knew plenty on that subject. No, I have learnt about the 'way the world really works'. I have learnt how to solve people's needs, how to sell things, how to handle people who don't agree with you. I have learnt how to dig deep and find the courage to keep plugging away, just so I can keep that dark corporate cubicle and mindless 'team building' exercises at bay.

Has anyone checked out the reddit comments for this same article? It's like night and day.

http://reddit.com/info/6cu0t/comments/

Oddly the thing that stood out in those comments was the discussion of living expenses, and specifically health insurance.

Everyone is quoting things like $700/mo and $1100/mo. Even the guy with the high deductible plan alluded to paying $116/mo for a $5k deductible.

Am I the only one that pays $55/mo for a $1k deductible plan? I must be lucky or something. And that aside, seriously, you buy insurance for catastrophes... not to save $150 on a doctors visit. Why the heck would you spend $1100/mo on insurance?

No wonder these people think startups are hard. They'd need a VC round just to cover their insurance premiums.

My first major job out of school was with CitiBank. I hated the corporate culture there, and the immense inertia. When I left I had health insurance through the federal COBRA plan, but it cost me $400 monthly. I don't recall the deductible offhand. That was a double-whammy because not only did I no longer have the steady income from that job, but I now had an additional $400/mo expense. Fortunately I had some small savings, but I was almost to the point of going into debt for basic necessities before my next full-time job started. Beyond that, even though I had insurance through my next employer, the insurance company had a policy of denying coverage for any pre-existing conditions for three months. I was fortunate that it was only three months; I believe they can have such policies which last for up to a year.

The only thing worse than paying huge amounts of money for health insurance is not having insurance. My out-of-pocket expenses were about $800-$900 monthly the few months I did not have insurance covering things like prescription medications.

I'm living frugally now and saving funds to begin my own startup. But I now know that my true expenses are much greater than I had imagined as a student due to the high cost of health care.

You should compare your health care expenses to what people in countries with state-paid health care pay. In Norway, the government or local government end up with about 60% (probably a bit more) of the money a regular employee earns. I think something in the realm of 10% of that money goes to health care (stitching together these numbers is a bit tricky, because our health care doesn't pay for exactly the same things your health insurance does)

If you live in Washington state, you can apply for Basic Health. It's state subsidized health insurance. I pay only $17 per month right now.

Before I found Basic Health, I was paying $200 per month.

It seems to vary a lot state-by-state. I remember reading posts about $55/month or $79/month health insurance before quitting my day job, but the cheapest I could find here in Massachusetts was that $116/month high-deductible plan. Most high-deductible plans were in the $120-130/month range, with some of the ones that didn't discriminate based on age ranging up to $230. COBRA would've been upwards of $400, and I think plain old independent HMOs would've been comparable.

Reddit believes that it is impossible to be happy or succeed in America because we do not have Universal Health Insurance paid for by taxing rich people. All in all, they have a very pessimistic view of the world in general and the free market in particular. The startup crowd is long-gone from reddit.

My hypothesis is that this effect comes from the international nature of reddit's audience. I hope this hasn't become the normal thinking of people in the US.

A lot of people these days (myself included) have chronic conditions that require ongoing regular doctor visits and/or medication. Even at a high premium level, insurance is cheaper than paying out of pocket (my maintenance medication costs $48k annually!).

I guess you could argue that that makes me not a good candidate for a startup... which is why I still work for BigCo, and work on my startup at night (N.B not the one currently listed in my profile).

I've got a friend that remarked on that once...he has Marfan's syndrome, which means that he has pricey routine medical treatments and the possibility of some very expensive emergency operations. He's working at Microsoft (despite being a big Mac person), and he said that one of the reasons why he chose to work for "the evil empire" is they have great health benefits, and no small company would be able to cover his health insurance.

Ahhh, very true. I imagine it would be nearly impossible to get a plan outside of a company with $48k guaranteed outlay, since you wouldn't have coworkers helping to even out the cost for the insurance company. I guess there are always government backed plans like COBRA for cases like that.

I think the best candidates for startups are the ones who actually make the leap... everything else is accounting... best of luck to you!

COBRA is pricey as hell! My COBRA premiums (if I had taken it when I left my last job) would have been around $850/month.

This is why the company has to pay for this stuff. I'm pretty useless if I get sick...

$55/mo? Here in NY, you'd be lucky to get a Band-Aid for that. The cheapest plan I've seen here is the state-subsidized Healthy NY, which is ~$200/mo for crap coverage.

My girlfriend is going back to school in the fall, and she's facing the decision of what to do about insurance. Seemingly, no reasonable option exists.

I was paying $550 a month in NY for United PPO while I was doing freelance/contract work. I ended up in the hospital for a week. Bill came out to something like $45,000, I didnt pay a cent. Lesson here is it's not worth getting cheap junk insurance.

I don't think I would get coverage from healthy new york based solely on their website appearance. I'm shocked they aren't using the blink tag, seems so appropriate.

Unsurprisingly and understandably so. PG's essays resonate most strongly with a particular type of person. There are enough assertions that do not generalize across diverse populations and nobody who reads his essays should actually expect them to. The population in this site are not are self selected to the extreme. I for one am happy to see multiple opinions.

There's something about anonymous commenters bashing PG and YC on reddit, of all forums, that makes me smile.

Well, here's one way to think about those commentators: http://www.youtube.com/watch?v=ZxxPLDZnqwA

Interesting the comment about 'living a human-scale life' was worth the read.

Bridgewater Associates is a large (~400 employees, $36 billion assets) hedge fund that tries to avoid hierarchy:

http://www.bwater.com/home/philosophy.aspx

"Conflict in the pursuit of excellence is a terrific thing and is strongly encouraged, in fact demanded. There should be no (or as little as possible) hierarchy. Certainly there are organizational "superior-subordinate" relationships; however, every "subordinate" is encouraged to argue with his or her "superior" if he or she thinks they know the better way, and every "superior" is required to encourage this."

I've got a friend that works for Bridgewater, and it's amazing the extent they go to foster this independent culture. My friend can be kinda abrasive at times - not really arrogant, but more an "I'm right, damnit, and I'm going to be pedantic about it" personality. His first assignment was copy-editing some client reports. He came back with a long list of grammatical mistakes in the report, along with a lesson on proper grammar, for the fund manager. The fund manager said, "This is excellent - keep up the good work," and made sure my friend got his hands on all the reports going out...

That's excellent! How long has your friend been there (one issue with Bridgewater is that I have met many people who worked there, and very few who worked there for more than a year -- I think it's pretty jarring for most people).

He graduated in 06 and started there soon after, so about a year and a half so far. I think he really likes it...started out in reporting and I believe he's now on the investment associate track, so obviously he's hoping to settle in there. (He's also a foreigner though, and Bridgewater is sponsoring his visa, so I don't think he could jump ship if he wanted to...)

Let's assume humans are indeed "designed" to work in small groups of only 10 people, then at the bottom level of the company tree groups should only consist of 10 people each. So far so good.

But why has a company of 1000 a three-level tree structure with 10 groups of 10 groups of 10 each rather than a two-level structure with 100 groups of 10 each? You assert the reason is the 100 bosses couldn't work together as that group size would be too big. But the bosses don't form a "workgroup" in this sense anyway. They may meet occasionally but spend most of their time with their own groups. The reason for the additional tree level seems more that the supreme leader of the company wouldn't be able to oversee all 100 groups himself.

The distinction matters because for the purpose of being overseen, groups can be represented by their bosses without having to act as one person (as they might have to if their bosses formed a workgroup subject to size limitation due to humans' innate design), so the inverse relationship between freedom of the leaves and company size can't be deduced.

One of the things that drove me to leave Deutsche Bank was that I had much less choice in who I got to work with. This is actually a big deal. By and large, in a startup you choose who you work with (either by hiring them, or being picky about which startup to join). In a big company, you have no real say.

And it can be a real bummer not to work with dynamic, ambitious people.

I work for a large company and I also feel the way described in the article. What baffles me the most is the constant need of approval in case I'm about to do something that goes a little beyond the "company policies", like setting up a public repository for code that's going to be released anyway.

Often these mysterious company policies are not explained in the first place, yet I'm supposed to adhere to them. So I find myself pondering, "Is it ok to do this?" Engineering-wise (and by common sense) it would be a no-brainer. Then I waste time and energy asking approval. Sometimes I get denied and so I wasted my whole train of thoughts related to an otherwise obvious choice. What bothers me the most is that the majority of people seem numb to these problems.

Why did you hire me if basically you're questioning my trust, and why are you blocking me from doing what I'm supposed to do? Then why you invite me to "fun" events afterwards?

There are a couple of questions (otherwise great article). One small is right only as long as it is economically viable. Do you believe Google would still be viable as a search engine if they had remained at 50-100 people size (or whatever you think is the optimum size?)

Second, most YC companies don't yet have business models. That is not a knock on them - I am simply making a statement of fact. Given that many (or most!) are hoping for a nice "exit" from a large company, doesn't their success pre-suppose the existence of large companies to take the other side of their exit trade? In effect, does the YC model (mostly) require the continued existence of large inefficient players? What happens when the (software & internet) world were to be full of small, ultra-efficient start-ups? No exits? (the second question is totally hypothetical - large firms do exist, but just asking to clarify the argument myself).

I wasn't saying big companies don't work, just that it's bad for one's brain to work for them.

I find this to be a major thesis in many of your essays, but I've got to know... do you feel it would be bad for your brain to work in google research or (gasp) microsoft research or academia? I mean, I would bet working for Norvig is a more significant mental workout than founding many of the yc companies.

In earlier versions of the essay I had a sentence saying that in a sufficiently pure research job, you're effectively working for yourself. I don't know about Google specifically, but I get the impression Bell Labs at its height was like that.

I imagine Bell Labs would have been an incredible place if only based on people that worked there. I wonder how close IBM and HP were at their research peaks. Also, I should mention that I wasn't intentionally picking on yc companies and perhaps that was the wrong choice. Some of the technology coming out of a few of them seems to be more interesting than some of the best labs. I know that isn't the goal, but it certainly isn't bad side effect! I really only meant to pick on the droves of obscure, uninteresting <insert social aspect here> startups.

I think the main constraint is the lack of creativity and the variety of tasks one has to perform in a big company. In a startup there are so many challenges and so many things to learn its always changing,

From personal experience, MS Research (and reportedly, Yahoo Research) is an awesome place to work for researchers; lots of freedom to do what you want and push the boundaries of what's possible. What you learn there is very different from what you learn doing a startup, of course, but the point about lack of creativity or freedom aint valid.

Interestingly, Google research is not very good in this respect due to their focus on products. Restricts creativity; if you really want to do products and be creative, get into a startup instead.

I roll my eyes whenever I see something like this ascribed to human nature and evolution. Passing by the post hoc ergo propter hoc nature of the argument, it doesn't allow for human diversity. Humans are very diverse, some of thrive in small bands, and some of us thrive in huge hierarchical organizations. There is no "one" human nature. Look at reproductive strategies: harems, polygamy, polyandry, monogamy, serial monogamy, celibacy, all touted as by one group or another as the one true, natural, reproductive strategy for humans. You could argue that the evolution has best fit us to live in tropical savannas, but the Inuit, Lapps, and Siberians would take it very much amiss if you told them they should abandon the arctic because humans had not evolved to live there. Evolutionary biology does not consist of telling "just so" stories to explain the virtues of your tribe.

What I would tell Siberians is to wear coats because humans had not evolved to live there.

In the context of "The people who come to us from big companies often seem kind of conservative." --- how do founders coming out of academia fair? Specifically, founders with, at least, a few years towards a phd.

They seem smart but sometimes impractical. Fortunately smart people can learn to be practical.

Can you give examples of impractical things they do? (Or practical things they don't do).

I think that this analysis only focuses on one half of the forces that direct survival. Humans may work more efficiently as small groups, but we have also formed into large hierarchical structures for as long as we have had history. Egypt and Summaria were both hierarchically governed. Organizations survive by becoming more efficient, but they also survive by have more labor available. Napoleon would not have won a single battle if his soldiers had fought in groups of 8 to 10.

I think that it depends more on the specifics of the situation and what you are trying to accomplish. Large organizations have advantages that no small organization can match. You gain your freedom at the price of security.

So what? Egypt and Sumeria were organized like North Korea today, worship of the head of state. And Napoleon was a tyrant trying to impose his will on an entire continent. If freedom is incompatible with Napoleonic dreams, that is not a bug.

Freedom fighters have often chosen the guerilla way.

A better example for your point would be the Americans in WWII. But even then, soldiers are not robots and the best commanders give their underlings the freedom to get the job done.

You're thoroughly missing my point. PG's argument is that man was designed, via evolution, to live relatively autonomously. My point is that even a casual glance at history disproves that. The German and Swiss states, for all their autonomy, couldn't stand up to Napoleon's hierarchical tyranny. In fact, it took a far more tyrannical society, Russia, to put Napoleon in his place. Hierarchy is sometimes necessary for survival, and is thus favored by evolution.

When we seek freedom, it is not because it is the evolutionarily superior state. It is because their is a conflict between individual advantage and collective advantage. When evaluating life decisions, you have to account for both.

Because German and Swiss states couldn't stand up to a larger force organized hierarchically is not proof that humans have _not_ evolved to work in small groups, it is only very poor empirical evidence that a larger force with better weapons can defeat a smaller, disorganized force.

As for why people seek freedom, I doubt most consciously think about it in those terms. You seem to think that individuals ponder about the pros and cons of going it alone or being part of a collective, but it just doesn't happen that way. When you're brought up in a culture that is already one way or the other, and that is all you know, you rarely choose because it's not even a question that enters your mind.

I'd suggest reading Beyond Civilization by Daniel Quinn.

The essential flaw in your logic is this: "evolved to" Evolution has neither purpose nor end. That which is more 'fit' (a very vague term to be sure) is what evolution favors. If a larger force with better weapons can defeat a smaller, disorganized force, then evolution clearly favors the former.

Napoleon is a case example. There are many, many more I could call upon, but I think a lesson in world history is beyond the scope of this debate.

Hierarchy is probably necessary for any society. Tree structures are efficient.

But tyranny is inefficient. It's all about ruining the bonds of trust that hold society and the economy together, just to benefit a small number of people.

All your examples are tyrannies. There is a word for people who invoke naive Darwinism to justify tyranny, but it escapes me at the moment.

Ah, but startup founders living in the US can still rely on the US Army to protect them from invasions.

You're already offtopic -- though I think not as offtopic as you might seem to be.

You are correct, which is one reason why the US has had such a prolific history. On the other hand, as the US government (and army) grow, we are now going in the opposite direction of lacking security from said army.

And yet founders live almost exclusively on pizza...

And Mt. Dew (or, when you're lower on money Citrus Drop), which is the epitome of refined sugar, high fructose corn syrup, and hydrogenated vegetable oil

>Credit card debt is a bad idea, period. It is a trap set by evil companies for the desperate and the foolish.

Best line of the whole article.

Credit card debt is foolish... credit cards themselves are great. Definitely prefer to swipe plastic rather than carry around pieces of paper.

Thus the debit card was invented.

The danger of the debit card is that if there is an unexpected charge or some other banking error, your cash is tied up until the problem is resolved. You can't access your money. With a credit card the tied up money is the credit card company's . My policy is to use credit for anything over $20, and just pay off my balance every month. It's a good compromise between safety and convenience.

In the UK at least, debit cards dont offer the same levels of protection as credit cards if your details are stolen

Yeah, but I still use the credit card so I can keep most of my money in savings accounts and time my card payments to my paychecks. Having a balance sitting in my non-interest-bearing checking account seems inefficient.

Yes, but you try canceling a purchase on a debit card. Plus, I get points for free flights on my cc.

credit cards are great for building credit, which is useful for other things, like buying a house

yeah, provided you pay it off. My brother called me while he was in college and asked if he should get a credit card. I told him "if you can charge $20 a month to it, and pay it off every month on time, then yes, get one. If you can't, don't".

He was nervous about his organizational ability, and decided not to get one. Four years later, he had a tough time getting an apartment in SF because he had no credit.

It's stupid, and credit card companies definitely target college students knowing that many of them will end up paying severe penalties and interest charges. But the thing is, they'll give you a credit card because you're a college student without the same credit history they'd want to see after college. So if you have the discipline to charge very small amounts and always pay it off on time, it is wise to get one.

I agree with the sentiment of the article, although the title is a bit misleading. It's not like you were meant to be a boss either, which is running a company is about once you get past the startup stage. I'd say the message is that if you can't be a boss yourself, try to make sure that your boss doesn't have a boss (i.e. join an early-stage startup).

Large companies have many other drawbacks besides the hierarchical structure. I've worked for companies of all sizes, and these are the things I particularly disliked about large ones:

- Information travels slowly up and back down. Decisions take time. For this reason many projects, especially those having to do with rapidly changing technologies, just cannot be done at large companies.

- Sometimes your work gets lost in the noise. Your entire project may be canceled and most of the company won't notice it. One of the perks of working for a startup is the feeling of being in charge of things and knowing that your work is immediately visible.

- Human relationships do not respect the hierarchy, so there may be conflicts between random pairs or groups of people. The larger the group, the higher the complexity of politics. This non-linearity is perhaps the worst obstacle when it comes to getting things done.

Some interesting problems require more than 8-10 people, no matter how insanely great they are. I work at a highly interdisciplinary biotech startup. My instrument SW team is eight guys, the tools-for-scientists team four, mathematicians are about five, etc.; then you have a bunch of physicists, chemists, biochemists, etc. etc. I'd rather suffer the problems of a large company than work on a more pure software problem.

What specifically did the programmers in the cafe say or do to lead you to believe that there was "something odd about them"? Perhaps you remember one or two details that actually got you thinking about this.

It was mainly their body language and facial expressions.

As a victim of several such team building ordeals I totally got that.

I imagine a sort of confused purposelessness with a touch of embarassment, all concealed with forced cheer.

Mr. Graham,

You write in your "cliff notes" follow-up to this essay,

"My guess is that this is an instance of a fairly common Internet phenomenon. People are reacting to what they imagine I'd say in an essay on this subject—that an essay comparing startup founders to corporate employees would say that founders are great and corporate employees suck."

Is it wise to waste your precious time explaining yourself to any "reader" who hopelessly injects his own peeves and anxieties into what others say, muting intelligent discussion on any topic that Mr. reader happens to have strong opinions about? I guess it is easier to thoughtlessly fire back than take any of your suggestions under consideration.

You know this essay touches more than a few hot buttons. When you chisel away at the social creed surrounding the safe job in our society, you're asking many people to call into question a lifetime of decisions. Ones, I suspect, that were made mostly out of fear of veering too far from a well-paved super social highway promising happiness and fulfillment through Conformity rather than Individual Desire. You're calling the bluff their life is literally based upon, stripping away some of that veneer of dignity. If they fire back, it's because they have nothing left to say and need to fill the emptiness inside.

If people worship the mavericks and company founders, it's because such individuals exemplify attributes very far removed from the cowardice that most people face in thesmelves even before their feet touch the floor out of bed each morning. They may feel the energy of their own great potential, but have painted themselves into a corner with what they regard as "acceptable" behaviour and risk. So they stand in awe of the achievers and gawk at those who try to be achievers, secretly praying for the day when it might someday be their turn.

Thank you for sharing your thoughtful essays.

Back in the 70’s I started a burglar alarm company - I had the knowledge and there was a need in my geographic area for another provider.

I’ve been self employed on and off ever since. At 63 I’m having trouble finding a new business to start, not because I lack the drive, or ability to start a company but in this location and economic times I’m having trouble finding a way to leverage my knowledge and abilities into a new endeavor.

I suspect that a lot of your programers just haven’t found their “big idea” yet. In computer related fields, the cost of entry is now so low that a huge percentage of programers have some idea they are playing with on the side. Like many actors, they will be overnight successes.

in the notes of this essay there is a strong statement about not funding startups with credit card debt. Just wondering if there is an acceptable limit of cc debt to use to get from concept to prototype? Or if its all the same and just a bad idea. beyong YC and friends & fam, there are limited options for the small pop of cash that gives you the time required to work in your startup. So far at my company we have been very tightly spending on a cc, not incurred a lot of debt but still don't have the small cash cushion to ditch the day job and jump in full time. sorry for the tangent.

One way to think about CC debt: if you fund your startup with credit cards, you need to guarantee that you'll grow earnings by at least 20%/year, just to keep up with interest charges. If you can't do that, you're personally on the hook for any shortfall, and have to make it up yourself out of future job income (after compounding, no less).

There're very few startups where revenues are certain enough to take this risk. Most likely, your startup will never make anything, and then credit card debt for a startup is no different than credit card debt for a spending spree. That's why startup founders are wary of debt financing. If you can guarantee that level of earnings growth (perhaps you have customers already lined up, and just need to deliver on a well-understood technical problem), then it can make sense.

It's very different if you have a day job, income, etc. I think he's saying "Don't work full time on a startup by living off of credit cards." I have credit cards with limits more than a YC investment, but if I lived off that and the startup failed, instead of starting fresh, I'd have a lot of money+interest to pay back.

I would suggest working part-time as a freelancer as another option. (i.e. if you can't get into YC)

I keep reading negative comments about this posting, but I think I'm reading a lot of defensiveness. I think it is important to keep in mind that, as I read it at least, Paul is not trying to slam people who work for others. He's really trying to inspire people, to get them to give themselves permission to work for themselves instead.

If what you hear when you read this essay is a personal criticism of the choices you've made in life, perhaps you should take a moment to evaluate whether those negative thoughts are coming from the essay, or from within?

Have not the last 50,000 years of evolution primarily been of social structures and culture?

Consequently, is being 'wild' necessarily a positive attribute?

Doesn't this (and comparisons to animals) to some degree reflect de-evolution?

Multicellular life is faster, smarter, and more successful than unicellular life.

The corporation is usually way slower, dumber, and typically does not outlast any of its component organisms.

The difference is that the components of the corporation are individually intelligent and self-aware. The corporation's "thoughts" are comparatively slow and stupid. Maybe one day there will be some Borg-like way of making humans act together, which truly makes something smarter and more nimble than any individual. But I don't think you can argue corporations represent the next step in evolution.

Argh! I hate these casual invocations of evolutionary biology that are really self-serving fables. By numbers of individuals, diversity, and by mass, bacteria and blue-green algae make up most of the biological world. Multi-cellular life only appears dominant to you because that's who you eat and who you mate with. Evolution cares not a whit for smart, dumb, slow, fast, it only cares who has the most babies who live to make other babies. Making arguments for the virtues of entrepreneurs is fine, just don't try to dress them up as scientific truths, unless you actually can muster the evidence.

I do too... I thought I was arguing against one. But I see your point.

Has anyone else been to a Safari or seen wild life (read: Lions or Cheetahs) in their natural habitat?

If you have, that metaphor of a caged Lion vs. a wild Lion is (almost) all you need to extract out of this essay.

If nothing else, should offer a big confidence and morale booster to envision yourself as a lion prowling in the wild as you furiously hack away at your startup.

I have been on African Safari as well and that point instantly resonated with me. The 100s of 1000s of miles were theirs to freely roam around and do what they wanted - and the humans who came to see them are the ones in the cages (except that the cages moved around - in the form of a jeep with a guide).

I also used to wonder what is going on in the Lion's mind - some 10 jeeps surrounding it while it is resting in the shade and some oohs-and-aahs (and even though it is 1 pm, some idiot always has the flash-on by mistake).

I'll just say this.

When i was on the safari, our vehicle got stuck in the mud. There was a pride of lions within 50 meters of our jeep. We had to get outside and push. One member of our party was the look-out, and at one point yelled "the lion is coming" at which point we all moved as fast as we had ever in our lives.

All the lioness did was get up and start pacing.

While I was at the zoo, there was a 3-year-old boy taunting the Lion in his 15 by 45 foot cage.

See the difference?

"That might be worth exploring. I suspect there are already some highly partitionable businesses that lean this way. But I don't know any technology companies that have done it."

I believe Gore-Tex may work this way

I don't disagree with the sentiment there's a small thing that I think is slightly incorrect.

Baboons, particularly Hamadryas Baboons, can be found in troops ranging from 5 to 250 animals (although a troop of 400 has been reported). These troops have a very strict hierarchy with about 4 "levels": a harem - containing one male "follower" and up to ten females, two to three harems unite repeatedly to form a clan, and two to four clans form a band.

Very large hierarchies do and can work in the wild, although they are rare or last a limited period of time.

Hmmm. I don't think it takes brain surgery to figure out people are happier when they own a piece of the pie (landowners vs owning land, Voltaire, etc) - and obviously this means more when you are working in a small group that can see results or achieve goals creatively and together. This article really only address three divisions of labor - internet software service, defense providers and investment banks. Which, at last check, combined employeed only 20% of the entire US workforce. To get into a larger, meaninigful discussion on a subject that is studied across the world, you would have to understand the dynamics of laborers, unions, operations and management. It's not as simple as saying "If you build it, they will come". And, it's real hard to feel sorry for white collar employees working for eBay, Google and Microsoft when 10% of our country struggles to stay relevant at the poverty line. It should be noted, that I've been both a successful start-up founder and a big business employee.

I cite the comment above mine, haven't we all written a term paper like this in college, titled "Utopia in the Workplace?". Paul, you should tour the John Deere manufacturing plant in Illinois to see a large company, with the "old school" top down management style - and try to figure out why that place is thriving. Or the Kitchen Works in Elmira. Or the Soy Convertors in Iowa. On and on.

A thoughtful essay, I agree with its spirit. Nicely written.

The arm-waving stuff about evolution disappointed me, tho. Supposedly, the genes humans have are in aggregate woefully ill-suited for success in modernity, at least this is what I hear from people who make similar arguments relating to evolutionary psychology. Programming computers post dates by at least 100K years the environment in which our genes were supposedly selected for reproductive success. On theory, everyone of us ought to be clumsy at best around computers, tho the evidence is all to the contrary.

Even granting your evolutionary premise tho, it has some problems. For one, I think you overlook a number of notably spectacular successes in eusocial species where size matters. Instead of lions in Africa, for instance, spend some time observing an ant mound. I also wouldn't expect that reproductive success and professional success would necessarily track each other. Or maybe I misunderstood you.

My $0.02: I think fulfillment of the individual and the success of the organization is more the result of the nature and motivations of the actual individuals involved with it than, per se, the size of the organization. Personally, I've worked as a programmer for big and small, but my best experience was as a private contractor. The evolutionary equivalent of the lone hawk if you'd like. (I'm now doing something ENTIRELY different).

I think most young programmers are well advised to spend some time in the box before they can learn and appreciate thinking and working outside it.

I've thought about how we could structure a business in such a way that no team became too large, and all functioned relatively independently across the country or the globe. I called it a "federated" business, which had a Constitution at its core that served the same purpose as the Constitution of the United States: to hold together these "federated" business tribes as one company. Each tribe did whatever work they wanted -- they had to find the work and do it, but could always request assistance from other tribes in the company who had expertise in areas they might need to complete a job. Any individual or tribe would be free to leave at any time without any restrictions. The only purpose of the "federal" part of the business would be to manage office-type things like legal, payroll etc. and possibly maintain a core tool kit that could be used by all of the others in their work. I figured out how to map it to the current corporate structure: each tribe would form as its own LLC; the core federated part would be an LLC as well, but would not own the others. There would be an agreement signed between the "tribe" LLCs and the federal LLC, but as I said, any of the tribal LLCs could "secede" at any time, as long as they fulfilled any open obligations to other tribes' customers. Also, all bylaws would restrict any tribe and the federal part of the organization from going public, even if they seceded. That would be the one restriction. I would only be interested in this kind of business if it were for the long-term, and not with the goal of flipping or cashing out. Not sure it would work as advertised, but it sure would be a cool experiment.

/s.

The first thing you notice working at big companies is that they take away your admin rights on your workstation. You can not be creative with a restrictive environment. You are forced to use the tools everyone at the company uses and force to think like everyone else.

For those who are working at small company, stay where you are. It is true you don’t learn much at big companies. It is not much about how well you do things, but it is how well you are connected and how well you can exaggerate.

Not everyone is meant to be or want to be a startup founder either.

If you look at the history, most substantial technology innovations are not from startups but universities and research labs of large companies.

Startups are more about the right idea (in business sense), the right time and the right place than innovation and quality work. It's more like a monopoly game than truly substantial and creative technical challenges. It attracts more people who think they're good than they're actually good.

Been there, done that, got the t-shirts. For real hackers who're incited/excited by the article to be a "founder", here're a few notes based on real experience:

1. Many top hackers are not interested in and bored by the financial aspect of the business. But you have to learn the basics in order not to be taken advantage of. It's not that hard to crack once you decide to bite the bullet.

2. Pick your business partners carefully. Not just someone who's smart but someone you can trust. To use a cliche, it's a jungle out there, which I could only appreciate after experienced it the first-hand.

3. If you want to go through the Y route, talk to the previous founders (esp. not so successful ones). You learn the most from failures.

4. Make sure you understand and make copies of every thing you sign.

5. Beware of business people who boast photoshop chops.

I suspect that most good hackers can do well with the financial and legal aspects of a business once they decide that it's an important system to understand. Hacking a contract and hacking code are not all that different in the skills required.

Still, having partners you can trust seems very important to me, too. Money can either unite people or tear them apart. You have to know that your partner has got your back.

The point is every good hacker doesn't necessarily want to bother with financial/legal aspects of a business, as they're not interesting to them. I know a top hacker who's not only good at system abstractions but also knows bare metal like the back of his hand, debugging live applications with heavy OS interactions by patching running binary on the fly without needing to lookup a reference. Everyone around him is in awe. However, he's not that interested in money. He thinks he's rich enough with a few millions in options and 300k in annual salary & bonuses. And he's happy with his job and his boss, who, like everyone else, respects and pampers him. The big company can give him resources a startup or smaller companies simply cannot afford. Why would such a person do a startup?

The original article generalized too much, missing the most significant contributions to the progress of technology from people who _like_ to work in academia and large companies.

Oh, there are absolutely people who are not interested in founding a startup, for many reasons. I'm just trying to say that hackers interested in becoming founders have the skills and hopefully the drive to master the financial and legal aspects. At least enough to avoid some of the major pitfalls. If they aren't willing to face these aspects of founding a business then they should definitely seek some other avenue of fulfillment. I think that hacking is a necessary but not sufficient skill for becoming a successful founder.

The essay reminds me that autonomy is a glorious thing - we are ends in ourselves, not someone else's means to an end. It takes courage to be autonomous, both from the individual and from the group around that individual. Teddy Roosevelt said it:

It is not the critic who counts, nor the man who points out how the strong man stumbles, or where the doer of deeds could have done better. The credit belongs to the man actually in the arena, whose face is marred by dust and sweat and blood; who strives valiantly ... who spends himself in a worthy cause; who at the best knows in the end the triumph of high achievement, and who at the worst, if he fails, at least fails while daring greatly, so that his place shall never be with those cold and timid souls who know neither victory nor defeat."

Life itself requires audacity to be truly alive. Audacity, as you pointed out needs no boss - no coach, no manager. It needs more than anything, freedom and respect from others. This essay is one of enormous hope to me - young people I work with are timid and aspire to the imprisoned safety of the cage not the free ferocity of the wilderness. Too many aspire to be either the sheep or the shepherd - to few aspire to be neither.

Excellent article. I am a lifelong corporate IT guy who has a life long fantasy of working in a startup company. In my most recent job in which I have been on for 13 years, the IT shop consisted of about 30 people when I started. We were very entrepreneurial when I started and used to be rock stars who would crank out solutions for the business. Now we are 200 people moving at snail speed spending most of our time on compliance projects, infrastructure upgrades, and mindless enhancements to monolithic systems. It is hard to promote change because the status quo is rewarded. As I walk the halls, many people look like they are serving time. Management (which includes me) is always trying to foster innovation and new ideas. But only the newer employees participate. They have not assimilated yet. I am sure after they hit the wall a few times, they to will walk with slumped shoulders.

So your article hit home and was a great read. The next time I go to the zoo and look at the lion in the cage, I will realize that I am looking in the mirror.

http://blogs.ittoolbox.com/eai/madgreek

Your article inspired this post on my blog.

http://blogs.ittoolbox.com/eai/madgreek/archives/programmers...

Thanks for this Paul. I've spent time on both sides of the side spectrum in addition to my own failed startup (but I'm not beaten yet!) and what you say is at this point in my life glaringly obvious.

I'm making the jump this month from a (safe, established) monolith back to a (exciting, dynamic, risky) small company and hope to be back on my own startup path soon, and this post was a bit of inspiration precisely when I needed it.

This article really touches on something I've thought a lot about. People are supposed to be indepependent. The freedom-loving spark is in all of us, and when we work for a behemoth organization or spend all day long taking orders from some arbitrary boss, we lose something. We change somehow.

Not that following a great leader is a bad thing, because it's not. Following a great leader is actually one of the best ways to learn (think getting mentoring by a successful entrepreneur... or reading and trying to live by the great works of Aristotle, Ghandi, Christ, etc.)

But there is a big difference between voluntarily choosing to follow a great leader, and being told that you must follow a leader. And if you miss work for 3 days, "we'll find somebody else."

120 years ago, 90% of Americans owned their own land and their own companies (usually farms).

Today, the opposite is true. 10% of Americans own their own land (sorry, a $500,000 mortgage on a $501,000 house doesn't count. Try not paying your mortgage for 3 months and see who owns that land)... and 10% of Americans have their own companies.

Maybe Americans are mostly living below their potential today...

Nice essay, as usual. I've always found these to be a source of hope and inspiration. At the end of the month, when I close the books on my company, I like to sift through these writings for inspiration. Because he's right: there is no guarantee you will be successful. I'm the partner in a struggling chemical company. We actually make physical items, which makes us a rarity as far as start-ups go.

You weren't meant to program computers either...

... nor, I daresay, were any of us meant for great wealth; which, as Graham may yet demonstrate, is just a much a corruptor of character and insight as poverty (if not more so...) Just as large corporations tend to stifle, indeed put a halt to, the generation of ideas, so to does the imposition of large sums of money. If that were not true, then the startups trajectory wouldn't bend towards acquisition... (or, despite Grahams elisions, shall we call it 'surrender'? )

The whole point of evolution is to adapt and mutate and so there is no such thing as 'normal' or 'meant to.' An interesting aspect of evolution is that we get to thumb our noses at the very idea of 'normal' and that we get to shrug off the very idea of 'meant to'.

Paul Graham has argued eloquently and succinctly this very line of reasoning heretofore. It comes, thus, as a shock to read this essay, coming, as it does, from the programmer who advocates bottom up design in opposition to top down mandates. Of course, the point of evolution often contrasts harshly with the point of cynicism: which is simply to justify any childishly satisfying, but nonetheless, foul mood or thought. Or, as Mencken put it, ``The cynic is the man who, when he smells flowers, looks around for a coffin.''

So much of this essay strikes me as the worst form of 'sophistry at a distance' that I scarce can credit Graham as the author. Surely some philosophy undergrad, still pink and slightly damp behind the ears, has stolen into Graham's study and slyly substituted his/her unfinished term paper for some valuable slice of Graham wisdom. Perhaps a wily hacker is having sport of Graham, and us, by subtly crafting an HTML parser that cleverly distorts, ever so slightly, essays fed to it... The world waits, with held breath, for the un-masking of the villain.

Yes. You weren't meant to have a boss, but get the feeling you think founders are somehow better than the run of the mill coders, but that is just not the case. We are all part of a system and need to work with each other. Hell, if I had a 9 to 5, I'm sure I could be a better husband and father and golfer, but I have worked an average of 12 hours a day for the last 10 years trying to be an innovator and I'm probably not going to stop anytime soon.

Founders may have freedom of thought, but they don't have freedom of time because they devote 90% of their time to their craft. It would seem that founders are the ones missing out, unless you are talking about the MIT and Stanford geniuses who's brilliance attracts funding.

I like what I do, but I don't think it is for everyone. If given a choice, I'd be the VC who gets to pick which young guns have the best ideas and make them share the wealth. The founders, who actually have to come up with the great ideas and actually execute...they have the hard job.

When I talk about humans being meant or designed to live a certain way, I mean by evolution.

Evolution is the change of a species over time--it does not have a conciousness in order to "mean" to do things. The related reasoning in the article appears to infer the superiority of the last 10,000+ years of "hunter-gatherers" over the last few thousand years of larger organizations with managers for a specific purpose (building pyramids, aquaducts, etc.)--but these large public works projects are in fact the very hallmarks of the advancement of evolution. I'm not saying that we were "meant" to work in large organizations, but it is hard for your article to that conclusion using evolution as a reason. Indeed, the concept of evolution would seem to contradict your point. By definition, your ancient bands of hunter-gatherers were less evolved than we technologists are today, and today the large organizations (from government to google) reign supreme...therefore, large organizations are more "evolved"...

For large public works projects to reflect the advancement of actual biological (rather than cultural) evolution, they'd have to affect people's chance of having descendants.

There may have been some of that. Maybe people who could tolerate organization into armies have been exterminating people who couldn't. But the time during which that has been taking place is pretty short by evolutionary standards.

I don't think the evolutionary analysis works here. Small human groups on the savanna only directly competed for resources with other small local groups. Imagine if a single group, only 10% "better" than others, were able to catch all the antelopes worldwide. Everyone else would have to find another niche. Similarly, I might be confident I could be the best spreadsheet writer in Peoria. I could "own" that market. But the entire spreadsheet market, worldwide, is supplied by a few startups plus a few bigcorps plus a few open source projects.

Corporations arose to enable and respond to global markets made possible by industrialization. The Internet enables global marketing by small firms, but doesn't protect them from global competition. Yes, individuals should be encouraged to take a swing at forming a startup if they have some brilliant new idea. But don't fault them for rationally choosing safety in a larger group if they realize their ideas are not competitive on a global scale.

Paul, you should read some Stafford Beer - Brain of the Firm would be a good start. He replaces your sentiment:

"each person gets freedom of action in inverse proportion to the size of the entire tree"

With a theorem that states that the amount of freedom of each person is in principle a computable function of the purpose of the organization.

Beer describes how organizations could be run using a model called the Viable System Model, which he hypothesises is not only sufficient but also necessary.

Large organizations do in fact seem to inhibit freedom to their own detriment, but in Beer's view this is not an inevitable outcome of large organizations, it is simply a failure to properly design them.

Anyway, I highly recommend his stuff. Brain of the Firm describes his main theory, while Decision and Control is a fantastic voyage through his work in cybernetics in the 50s and 60s. It's all highly thought-provoking stuff. I tend to re-read his work every 3-4 years and always achieve new insight from it.

If you are interested in this subject, then a much more interesting reading about this subject can be read in John Viney's book Drive. I wish I had completed reading that book many years ago, yet I stalled and failed to read the final chapters that mention all that Paul has written but with a lot more explanation as to why things are the way they are. Groups are 7, 7x7 are tribes and beyond that you have hierarchies. Anyone who has worked for or come into contact with these different types of companies, eg. accenture will immediately spot the difference. All software companies function well up to tribe level, beyond that it gets very complicated and unattractive. A succesfull company should always strive to build well functioning tribes and keep hierarcial aspects of a company as transparent and non-intrusive as possible. IT shops that try to reorganise into hierarchies are destined to destroy themselves and if your's is, leave now.

Thanks for the article. It bears out the experiences I have had working for small and large companies. Engineers in startups are closer to the business and (sometimes) to the customer. They are solving real problems and inventing new things. Though it can be stressful, it also gives you a sense of worth. The other thing is does is allow you to "play" in the sense of being creating - making new things and having fun doing it.

Large companies tend to be political. Folks want position, title, etc. As one person put it - "I handle 250 off-shore engineers". My instant reaction was "but what have you, personally, built today."

There is the notion that hierarchies are good. That as you move away from coding you are "rising" in the hierarchy. This substitution of image for ability while gratifying to the ego, kills creativity and promotes mediocrity.

As you state, and experience bears it out: software works best with core (small) teams with creative innovate people working for a common goal.

"I can imagine for larger groups to avoid tree structure would be to have no structure: to have each group actually be independent, and to work together the way components of a market economy do. (...) But I don't know any technology companies that have done it."

I was thinking about this part and I realize Y Combinator doesn't look too different from that.

Great article. a) I liked the HFC / food comparisons. b) I've been thinking a ton about the relationship between maslow's hierarchy of needs and the needs of groups (and, potentially, their parallelism and that parallelism's relationship to maturity and intelligence / technological singularity). anyway... i dug (one g, not two) it.

I'm not sure if you have to go back as far as the savanna for a comparison. I'd regard the lowest level in corporates as servants of the boss, the bosses job is to fight off the other bosses and acquire resources for his people, so that he can acquire more stuff and convince his bosses that he can use the resources more efficiently.

Corporates are more a feudal hierarchy, whereas startups I liken to the old guilds of Europe that evolved out of the feudal system, it has been said that programming is one of the few crafts left, and so the comparison may be appropriate. The guilds occurred because the craftsmen had skills that weren't available to all, and so could demand a premium for there services, much like programming (and writing, and graphic design - witness the recent writers strike)

Maybe there is a new republic on the way all we need is to lop off the heads of a few managers :-). Let them eat cake!

Paul,

Good essay. I apologize for seeming nitpicky, but worry you may be encouraging misunderstanding about the type of person who succeeds at sales inside a large company:

<blockquote>The restrictiveness of big company jobs is particularly hard on programmers, because the essence of programming is to build new things. Sales people make much the same pitches every day; support people answer much the same questions; but once you've written a piece of code you don't need to write it again. </blockquote>

Sales people actually do build things. The things they build are called client relationships. Each one is new and different in some way. Once the relationship is created, upselling is pretty easy to do, as are yearly renewals. Often, these renewals are sent over to the account management team and taken out of the hands of sales.

There's actually quite a bit of strategy involved in creating these relationships, especially for a complex sale -- complex in terms of the organizational structure of the customer and the technology of the product. There's an awful lot of time spent thinking about what questions to ask different people in the organization and when in order to find their specific problems and elicit the specific needs. Only after you know what you need to know about the organization and the people in it do you pitch. A lot of these problems and needs are similar, but the pitch has to be tailored to what they have told you. This is how you show the customer respect, and build trust. A lot of it is subconscious but a lot of it is learned behavior.

Furthermore, sales at a large company can be quite entrepreneurial. You either make your numbers or you don't. If you don't, you get fired. If you exceed them by a large enough margin, you get to skip sales meetings and show up whenever you feel like it.

This is the first time I've commented on one of your essays, but I've bought about 5 copies of your collection for friends.

Brilliant article and well documented. I have always believed that nature always has answers to very complex human queries and you brought this out well in comparing organisation structures with animals that thrive in the jungle and wilderness!! ... i like!

You should read about eckart wintzen -edit, with you I mean Paul Graham. He founded BSO - it-company- in the Netherlands, now known as part of Atos Origin, but lost it old strategy of staying small while growing big, but that old strategy is very interesting.

Another appeal of big companies that wasn't touched on in the essay is that 'everybody's doing it'. I would argue that in addition to the economies of scale of mass production a large reason why you see people in corporations, eating crappy foods, and even wearing the same clothes is because 'everybody's doing it'.

There is some sound evolutionary reasoning in doing what lots of other people are doing. With the ease of living today and the massive amount of information available, the thinking is anachronistic. Too bad there are so many things that we're hard wired to do.

Both Google and Oracle try to use the "market approach" with almost no managers cited as a possibility in the article. At Google, it seems to lead to a lot of people with no direction and a lot of projects with inadequate resources. At Oracle, it led to outright warfare between departments and a total lack of cooperation. The difference is that at Google, failing didn't cost you your job. At Oracle, it did. So Googler's were independent but didn't have to compete for scarce resources the way Oracle people did. People seemed like happy lions at Google. They seemed miserable at Oracle, like bands of warring chimps always afraid of being annihilated by those in the next cubicle over.

reposted from http://www.theoildrum.com/node/3721 --- [-] [new] aangel on March 21, 2008 - 11:07am | Permalink ... Also, personally I am uninterested in eliminating hierarchy. Hierarchies are very efficient ways to organize to produce results. In particular, they allow for accountability, that magical quality of something being "count-able" and thus able to be quantified. Quantification and lines of responsibility for those quantities are what make managing large endeavors possible. There is on the opposite spectrum the "open source" movement or "hive" approach to accomplishing large projects, but even in those systems often you'll find a person or persons at the top who help direct the system as a whole and frequently they have the final call for a proscribed set of decisions.

When done right (and it is done right in many places), individuals perform better when accountability is present. They are clear on their tasks and given the resources (or not, as the case may be) to accomplish them they are able to use their commitment and ingenuity and drive to accomplish them. People who are unclear on their tasks report that they are unsatisfied and frustrated with the system; anyone who has had conflicting directives from an employer will appreciate that.

It's possible to spread accountability across multiple individuals and strictly speaking accountability isn't by definition an individual phenomenon. I can be accountable for the condition of the environment — but so can anyone else and everyone else if they accept by the group the role of being accountable.

But it does take a great deal more effort to manage an endeavor when accountabilities are spread across more people simply because the communication overhead increases to make sure omissions don't occur because each accountable persons thinks that "the other person was handling that." For that reason I will always set up projects (and advise my clients to do the same) where it is quite clear who is accountable for what and for whom — in other words, a hierarchical system.

One counterexample to your theory is that if bosses limit autonomy, then large universities should limit autonomy significantly; however, they're generous in autonomy.

Now on the theory underlying your essays.

If it's barbaric to work, then one should note that it's feasible to stop working immediately, without significant wealth. It can be done piecewise: many of my friends save up and travel for a year or two, or get schools to pay for their traveling. Assuming one is investing, and not increasing spending, it will get easier rather than harder to take more and more time off from jobs over the years. Of course most people shoot themselves in the foot by taking out loans or increasing spending, but that's their problem.

If autonomy is important, then why assume startups and develop lots of minor contradictions with autonomy? Why not assume autonomy, and see where this leads?

Given your value set, which seems centered around autonomy and creating/hacking, I can't see why you promote startups as a lone solution rather than promoting a variety of solutions. For example, living cheaply, moving to cheaper nations, taking years off to travel, only reproducing when one's children need no longer work (so as to make the problem of work globally and locally decrease, rather than increase), working to bring high tech to areas without it (which might give different kinds of autonomy, as well as cheap living), working contract type jobs where taking say six months off isn't a problem, working in journalism in the areas of high tech or art (more traveling, less pay), or working in academia, where the problems with autonomy go away and hence work becomes non-pernicious.

Your work with startups is really great, but I think there are lots of other options that are equally consistent with the values of long-term autonomy and creating/hacking. I don't see that you have any particular duty to present these options, but I do think that hackers should be trying to increase the sphere of jobs known to be consistent with their values, by discussing them publicly.

I love the analogy about lions in the wild. When you are living by your wits, there is a sense of "apprehension" (not sure I would call it fear) but that is where the adrenaline comes from and also the sense of freedom. You are (more or less) in control of your own destiny.

I am a veteran of the tech industry. Started a few companies and consulted for a twenty year period. I haven't had boss (the wife excluded) during the entire time except for a stint with Shell Oil out of school. Just recently graduated from law school (went back as an "old man") and am in the process of launching a "digital law practice" that feels in many ways like a tech startup. We are doing it completing "on the cloud" with virtual offices in a couple of states.It has gotten a lot easier (still not "easy") to do!

I think I found a middle ground before I started my startup. I spent about 10 yrs of early career in so called big companies. And I agree with most of what you said in article. That place is no good for creative programmers.

My middle ground was to "toy with my own ideas" in my own (and sometimes on company's ;)) time. While the big company job paid me well, my sharpness was maintained by own small projects.

10 yrs later, I couldn't take it anymore. I also had some cash and started my own startup. And needless to say, I think I am doing well. I also do not think growing big is goal of my startup.

You really should read up on the work of Clare Graves. Whether people work better independently or with a boss is dependant on their dominant 'values' system - this can change but does so slowly. As people grow through life we develop through psychological states that oscillate between 'we' centred and 'I' centred. There are four main values systems that a programmer is likely to be inhabiting. In the first(DQ) a boss is essential. The second(ER) will work very well as their own boss. The third(FS) craves peer support - but not a boss. The fourth(GT) can work in either way - as long as they don't get board because they are out of there if they do.

(in reality people are complicated and can be inhabiting multiple values systems - but most have a dominant one.)

"In the group one level up from yours, your boss represents your entire group. A group of 10 managers is not merely a group of 10 people working together in the usual way. It's really a group of groups."

one fundamental difference between managers one level up and beyond and programmers is that peer collaboration amongst managers is much-much-much less common than peer collaboration between programmers. This is inversely proportional the higher you go up the hierarchy. it seams to be due to the fact that managers have to deal with that one extra dimension of that programmers don't have to. Apart from managing their own bosses and having peers, they also have the extra dimension of managing their own team or group.

I think this is brilliant, because I have a friend who was recently fired because he is basically smarter and more brilliant than anyone else in his "programming" group. We work for a Fortune 100 "technology" company in which the technical manager above him does not know a single programming language. If you can imagine how demoralizing that is. I am convinced that they got rid of him so they could outsource the actual work to India, while maintaining their cushy managerial jobs doing nothing. In the meantime, the morale of the rest of the real programming team is plummeting, and the managers are leading blindly, because they do not know how to offer real technical solutions. Talk about DUMB large companies!

Your point about being more worried and yet happier at the same time is totally possible. According to the Big Five Personality trait theory from personality psychology, good emotions and negative emotions are two seperate scales, not one scale. Good emotion is Extraversion and bad emotion is Neurotocism. link here: http://en.wikipedia.org/wiki/Big_Five_personality_traits

Also, the two biggest things that correlate with success (three overall, two presonality) are openness to new experiences (creativity) and conscientiousness (discipline and willingness to work hard). The other one is intelligence. I reckon YC weeds for those kinds of people.

That was a great post, and I recognized themes from the books: "The Tipping Point", and "Freakonomics". What I like the most about Y combinator, from an observers standpoint, is that there is an offered solution to the entrepreneur's inability to raise capital with no track record. Other micro-investment and advisory groups are popping up all over the place, and this will stimulate more entrepreneurs to take a chance on themselves.

Couple comments here and what's written in Paul's post allude to our capitalistic consumption-based society. I think Annie sums the state of our world's affairs up pretty well, and you can see the relation to different aspects of your lives and big business at, http://storyofstuff.com/.

My problem is the lack of solutions that I see offered by those that raise objections to the status quo. Its good to make others aware, but always go the extra mile and offer a suggestion to change what you feel is so messed up.

That is what is missing in our 2008 elections. Just look at Obama and Hillary sputter about what they intend to do, with great passion. However, they never actually say how America will get out of Iraq and what will become of the Iraqis, how to fund the health care and education programs they propose, or what solution they’ll put in place to get the economy rolling again. All I hear is why race and sex matter so much in this election, and how historic it will be when a woman, or black man becomes president.

Those are not reasons to be America’s president! I don’t care what color, size, sexual preference, or country of origin our next president has. I want to know how they’re going to fix this mess.

As a startup founder, I keep confidence by following several principles:

1) "Success is merely others recognizing what you knew you had inside you the whole time." - Will Smith

2) "Plan your solutions for the worst, and leave hope to others." - Craig Benson, founder of Cabletron

3) "Champions take chances and pressure is a privilege." - Sharapova speech after winning the last Australian Open

The challenge I see is how to do a really big project with small independent groups of people. Open source is one concept that helps. In that paradigm, not everyone has to start from scratch. You can grab someone else's code and add functionality to it.

My pet project of the moment is to develop an "Industry wide SOA" with a defined interface between the services. This would allow entry into (in this case the medical) industry by smaller groups that can't do it all. See the beginning of the idea at: http://blogs.ittoolbox.com/km/med-research/archives/open-soa...

I think Jeff Atwood has a good point, in that your essay does have this sort of lofty view of startup founders, and implied condescension to the poor sods who didn't do that. That may not have been what you meant, but that's certainly how it comes across to someone with a more neutral view of startup culture. Your addendum doesn't really change that, because you're still positing a dichotomy of what software engineers can be: EITHER we're go-getter startup founders, XOR we're caged animals suffering as our souls slowly die, depending on which environment we end up in. In drawing those lines, I think you're missing a pretty wide range of human experience.

Great essay.

I work at university (as assistant prof. in CS, not in US), and it feels a lot like working in a big corporation.

The difference is that we are free to do what we want. But this is because nobody cares about what we do, so it might be even worse in the long term.

Paul

I couldn't agree more, although my background in advertising was inverse as well. From the world's largest agency (3 years at JWT), I spent 10 years at a mid-sized outfit, and then I finally put my name on the door when I was in my mid-thirties. What may make this different than programming, is that in agencies we all tend to be given as much room to create as possible (the exception being at JWT). However, to live is to create. Humans are in their soul creative, as that is part of the life force, the source. Let that atrophy and you become a waking/walking part of a whole. I love both your analogy to animals and food. Great Post! ieclectic.com

Hi Paul:

   I like what you said about lions.  Here's a little essay I did on cats a while ago.  It should ring a bell or two.

Eric

...... I’ve discovered that the way people react to cats is a good indicator of how they will react to people with vision. A certain number of people are totally freaked out by cats. These same people are freaked out by visionaries. This is really unfortunate. In reality, cats have at least seven admirable qualities that every person who would desire vision would do well to emulate. First: Cats are nocturnal creatures. When everyone else is asleep, the cat is alert and aware. Likewise, visionaries do their best work when everyone else is sleeping. You snooze, you lose. Second: Cats have astonishing eyesight. They see things nobody else sees. They are particularly sensitive to things that change; their eyes are custom-made to detect the slightest movement. On the other hand, they are totally colorblind. They are not distracted by the showy, the colorful, and the flashy. They only see what is important to see. Third: Cats are totally oblivious to public opinion. A cat neither knows nor cares what anyone thinks about him. He simply goes about his business. He doesn’t surround himself with yes-men (or yes-cats). You will never see a committee of cats. Cats never ask permission. Prophets and visionaries, likewise, have a gift for thinking on their own. Fourth: Cats slip into and out of tight spaces with ease. They are lean, mean, and adaptable. They are totally unencumbered by entourages and bureaucracy. Fifth: Cats have insatiable curiosity. If something in their universe changes in the least, they want to know why. They know there’s always something out there to explore. They take nothing for granted. Sixth: Cats are quiet (for the most part). They don’t waste a lot of energy bragging about their exploits before the fact, or trying to explain them afterwards. They know that nobody would understand them, anyway. And finally and possibly most importantly: Cats tithe. In fact, a cat will usually leave considerably more than a tenth of his latest hunt on your doorstep. You may not feel you have any pressing need for a third of a gopher, but that’s entirely beside the point. A cat understands a tithe is not for your benefit, but for his.

Hi Paul: As always your essay is right on the money. I've worked for big companies and I've worked for small ones, and let me tell you, small is better. I've heard it described like this. A big company is like an engine with a monstrous flywheel. If a couple of cylinders are misfiring (or not firing at all) you really don't even notice. In fact EVERY cylinder can loaf along at pretty mediocre performance, and the flywheel will still spin. The problem is, most companies interpret a spinning flywheel as a useful function in itself. They don't realize there's no useful load attached to the engine.

Paul, I think you need to take the next step here which is getting out of the venture business. The whole venture industry is based on your startup founders becoming bosses or being acquired by companies with bosses. Even if they end up rich afterward, the availability of the money is directly tied to other programmers getting duped into working unnaturally (with bosses). Have you considered leading your startups toward a small business path? More here: http://www.stubbleblog.com/index.php/2008/03/take-the-next-s...

We actively encourage startups to stay as small as possible for as long as possible. And like all investors we would prefer the startups we fund to go public rather than get acquired. I bet if you outsourced all the inessential stuff you could grow a company to that stage without needing more than 150 people, which you could do with only two layers.

[updated]: for some reason wuth all this login/acct creation, i lost first half of the comment. here it is:

I liked the essay even tho toying with the word "natural" is pretty funny, considering Paul's own later article on debating and arguments.

However, to talk about the main point: show me any startup/group that has succesfully REMAINED small as their service/business started to succeed and expand.

There is an upper limit of task switching and competence one can achieve in diverse areas.8-10 people? Fair enough until one after another they start doing something completely unrelated to their core professions just because the idea is to keep it small.

I usually take PG with a grain of salt but he hit the nail on the head with this one. Another phenomenon I've seen when working in larger groups/organizations is the predatory ways developers will pursue the core/api/library work so as to expand their freedom and limit their dependence on others. The essay's comments about the programmer at Google reminded me of this as I've seen it time and time again in my career. Whereas in the smaller groups/companies I've worked, all the developers were responsible for the entire stack. It felt more communal and I think the groups were healthier and more productive.

Whew! Brilliant article. I've been having a lot of conversations about this lately. I'm a freelancer. I became one because I wanted to get out of the cubicle. Now I'm doing all sorts of stuff. I'm working on amazing projects, things I never planned on (or could even conceive of.)

The hard part is that I'm working a ton. But I love it! I'm happier than I've ever been. I feel stronger. I feel like I can do anything--and I will, without hesitation. I am worried about the future at times. But I guess that comes with being in the wild, like the lions you're talking about. Great analogy! That really pumps me up.

This is now one of my favorites. Could not agree more. You sum up very well my reasons for leaving IBM to do a startup.

I keep trying to give my friends who're still there and working for other companies this message. They don't get it.

Do they not get it, or are they unwilling to? Admitting to yourself that you want to change is a hard step to take.

I think the people to whom this advice applies already know that something is wrong, and that there must be something better than the corporate monotony. All they need is some encouragement, and a bit of a support network.

Great essay thanks Paul. There are some who prefer to have a boss. I have managed people who just want to come to work and be told what to do, to function and know exactly what they are to do and that they will be paid for it under defined terms. These people don't like risk (maybe they have enough in other parts of their lives) or don't have the creative gumption (passion) to want to do their own thing. Either way, bosses have a place. Sad thing is for clients, you end up paying a lot for the infrastructure required to move these people collectively to some productive end.

Thanks for the article. I found your articles professional quality and for free. :-)

My father worked for a chemical company(10000 workers in Asturias, Spain) as chemist . He realize that he could buy a conventional ceramic oven for doing the same that another specific one that costed 100 times more. He tried to convince management. They denied the permission.

He tried again with different ideas, it never worked.

PD: Well, it worked better in another (later) smaller(100) company. The trick if he had a good idea, was to say it was an american method (Seems they didn't trust in local intelligence, this was in Madrid, Spain)

Footnote [1] ("When I talk about humans being meant or designed to live a certain way, I mean by evolution.") makes me wonder...

If the most prevalent state of affairs at present is that people go on to work at large companies, and by most accounts these people are reasonably successful (in fact, in a reproductive sense they may be even more successful than the average startup founder, since they have more time to pursue relationships), wouldn't one logically assume that over time the footnote, and the point of the article that it isn't how humans evolved to work, no longer hold true?

M0 theory intro http://www.buildfreedom.com/content/reciprocality/r1/intro.h... http://www.datamodel.co.uk/Reciprocality/www.reciprocality.o...

The idea that folks become addicted to boredom or ritual via producing dopamine. People have suggested that startups are run by "mappers" (associative thinkers) then are taken over by packers (people who like ritual, rote-work).

Many techies are "natural immunes".

I don't agree with Paul's argument as to why others won't feel the pinch: "The restrictiveness of big company jobs is particularly hard on programmers, because the essence of programming is to build new things. Sales people make much the same pitches every day; support people answer much the same questions; but once you've written a piece of code you don't need to write it again. So a programmer working as programmers are meant to is always making new things. And when you're part of an organization whose structure gives each person freedom in inverse proportion to the size of the tree, you're going to face resistance when you do something new." I don't agree with this characterization of programmers as more creative than others. It is true that you never <span style="font-style:italic;">have</span> to write the same code again (though how many times have you written authentication or a wrapper around an authentication class) but that doesn't mean that everything you write flows out of a brand new non-linear creative endeavour any more than a sales person's pitch is entirely new. Sales people sell services that have never existed (and then they get engineers to code this selling process into an online flow and move on to selling new stuff). Support folks get entirely new questions all the time (and then get engineers to code a system for responding to the easy stuff in an automated way). These folks that Paul singles out as being less restricted (or less "meant" to create) do all sorts of breathtaking new things too. I've seen support folks handle a very angry person with an issue with a brand new type of product by creating solution to the user's problem and explaining it in a way that is brilliant. I've seen sales folk come up with entirely new types of business arrangements or finding an elegant "in" to a relationship. It is certainly true that both of these groups do a bunch of work that is less creative, but so do engineers, even at startups (gasp!). These are the things that follow the creative move, things like debugging, unit testing, perfecting UI, etc. A good startup as well as a good big company, will value creation (and the stuff after the creation) in all of its people, not just the ones that are from Paul's chosen tribe. More at: http://www.bricoleur.org/2008/03/graham-dont-work-for-compan...

Ripping off Clay Shirky's book Here Comes Everybody, he points out that sharing sites like flickr allow groups of large size to collaborate effortlessly around some tasks (his example is people taking pictures of the Mermaid Parade in NYC) without having a boss or even knowing each other, on Flickr through tags. More tools like Flickr could supplant some of the hierarchy in large or medium size companies. Why shouldn't a company be able to be a large group of individuals that self-organize like the people working on Linux have?

The assumption seems to be that only smart programmers will read these essays. That's maybe 10% of programmers or less (maybe a hit count would help estimate?).

Big companies are all about efficiency through division of labor. They want guys who will write uninteresting Java applications for 10 years without complaining. But for a few thousand dollars a year more (especially if you're talking entry-level), they can hire a top-10% person overqualified for the position, just in case they need the talent.

Brilliant! For sometime now, I have been contemplating working for a big company or starting something myself. This is if not anything more, a trigger. Thanks for all that u write!

I work for a Large Auto Rental Company. We need some unbridled programmers. We have a HUGE IT dept, but I don't know what they do. "Managers" sit for hours transcribing data from one program in the computer into another. Hours, days, careers, lives wasted doing what a patch of VBA should be burping onto screens whenever desired. The veterans have never seen what a computer can do, THEY do it. They even use floppies twenty times a day. Please send me a cowboy programmer.

You don't need a cowboy programmer! You need to fire the IT boss. Also you should fire the big-boss because he hired and kept the one who was in charge of IT. The mediocrity comes ALWAYS from the top. Luckily for you the market in which you are is more juicy than your company can suck. So it survives even being inefficient!

It's the same with schools. My kids are at an eco school with a maximum of 12 in a class, and 3 classes in the whole school, by choice. They will not 'expand' but encourage other schools to start up and follow their model. It is all about sustainability. Like the Transition Towns initiative - small communities leading the way, because governments don't initiate. They don't lead. http://www.transitiontowns.org/

As I read this article, I realized that if you scale its message up to include not just companies, but the way modern society is structured, you will see exactly the arguments presented by Theodore Kaczynski: http://en.wikisource.org/wiki/Industrial_Society_and_Its_Fut.... If you read his words (ignore his actions for the moment), he is saying the same thing about society in general.

> The restrictiveness of big company jobs is particularly hard on programmers

It depends on how you define 'restriveness'. Sure in a large company it's difficult to make the types of decisions that will affect the whole organisation. On the other hand in a small company you are typically very restricted in terms of the numbers of users who you can make a positive difference on their lives.

Startups are typically experiments to see if there is any mass appeal. Large companies actually change the world.

I agree, mostly... Anytime you work for anyone else you give up a part of yourself. The more they pay you, the more of your soul you have to sell. The only way to be free is to do your own thing. For some that's starting their own business. For others it's walking in, working for awhile, learning what you want, and walking out the door when you're done (not when the company is done with you).

Oh, and what's the deal with note #1? That was a silly and unnecessary note to add.

Clear! After CS.EE Eng school I completed comissioning as an officer in the British Army. The hieracy, groups, rules, and orders were perfect to fight other Armys. "Peace keeping" in Ireland was not nearly as well suited. I left and joined a US start with 10 poeple (long story). The 10 person start became 3000, (10-60-120-240-480,900 now 3000 people). Relecting I feel like I'm back in the Army. Small groups 10-100 people can run rings around 3000.

You are the greatest. Lonely for clear, precise thinking, and actual content here in the midwest, and finding this site, this essay included, has been a joy. Plus, for all my own start-up endeavors, these encouragements to persevere are very helpful. True things do need to be said again and again and in different ways all the time. Thanks. Thanks for the careful thinking, and for advancing the idea of careful thinking in general. Thanks.

Well, since no one has mentioned the self serving nature of the article, I guess I will. Seems like someone who founds startups and needs a good pool of people to start those would need to convince people of all this.

Bravo, they seem to be buying it!

I worked at a few small companies and there are certainly downsides to small companies. Likewise, there are downsides to starting your own company as well, like friggin receivables.

I work in a 100ish person company now and it is awesome.

Here is what Malcolm Gladwell wrote about just this topic in a 2002 New Yorker article: "Institutions are not just the best environment for success; they are also the safest environment for failure--and, much of the time, failure is what lies in store for innovators and visionaries. Philo Farnsworth should have gone to work for RCA. He would still have been the father of television, and he might have died a happy man."

For programmers this was explained by Brookes in Mythical Man Month back in 1975 - a timeless classic. The return to smaller units is a bigger trend I think. In 1955, Fortune 500 controlled 1/3 of US GDP. In 2000 it was 2/3. I think we are at an early stage of a massive trend that will reverse this. Not just tech/media but shops, farmers etc. Phew, it was all getting kinda white bread boring out there!

Damn you, Paul. You had to post this just as I was trying to make the decision whether to accept a relatively high-paying corporate job or keep struggling to "live the life" running my own business. I'd almost convinced myself to go back into the world of commuting, bosses, dark windowless cubicles, and soul-crushing boredom in exchange for certain financial stability.

As someone whose startup was acquired by IBM, I whole heartedly agree. Every day in IBM is rot. Can't wait to do another startup.

Personally I think I saw what you were talking about - and by your notes I certainly interpreted this essay differently then some folk.

Yours was a good read - and given other 'coincidences' made me think.

Thanks, Joe

http://ponderousprog.blogspot.com/2008/03/to-tree-or-not-to-...

>Or rather, a large organization could only avoid slowing down if they avoided tree structure.

Valve, the makers of Half-Life and Portal, work like that. No one has a dedicated role, there is no hierarchy, there's just a guy with a vision and a large team of widely-talented people who work on what they want to work on. Obviously it works, too :)

Nice essay, only scratching the surface but catching some important content in the process.

Similar approach with more detail in 'The Future of Management' by Gary Hamel (ISBN-13: 978-1-4221-0250-3), at Amazon

http://www.amazon.com/Future-Management-Gary-Hamel/dp/142210...

GE has a jet engine plant in Durham, North Carolina that has 170 employees and one boss. It seems to work well. Although I have not heard that the concept has propagated to other parts of the company

http://www.fastcompany.com/magazine/28/ge.html

"groups of 8 work well; by 20 they're getting hard to manage; and a group of 50 is really unwieldy."

Amen. I like to state here my 2nd law of corporate's immutable entropy: individual intelligence times somewhere between 50 to 150 leads to collective stupidity. It's a rather sharp s-curve, really.

Great stuff Mr. Graham, thanks. Jurgen

An interesting essay, but some misleading assumptions.Consider these points: Animal groups do have bosses and hierarchies; Society at large is required to be hierarchical in order to create enough cash to allow high risk ventures to fund your 'wild' or 'natural' programmers.

I made a similar point, albeit more briefly, in this post: "Pardon Me, But Would You Have Any Grey Poupon?": http://buzzpal.wordpress.com/2008/02/23/would-you-have-any-g...

i think this is a universal challenge. i currently work fo a global oil company with a worldwide staff strength well over 50. i am based in the Nigerian operations. there is intense competition. But i really dont see what the competition is all about. we tend 2 push solutions 2 users rather than do what they need. there is a funny ranking system which forces u do workk and not share. I have been here for a year and a few months and i look forward to move on in the next 2 months. i think life was better even when i thought i was earning 1/5 of what i earn now. it is miserable working for huge companies. some people sure have to do it. But count me out soon. Life is beautiful and free. I agree.

I'll have to ask my Birkenstock-wearing living-in-Berkeley friends what they eat now... :-)

http://en.wikipedia.org/wiki/Paleolithic-style_diet

I work for a really small company. But I am feeling unconformable since a year. Now reading this essay I understand the actual reason. We have become a typical outsourced branch of big giant tree since last couple of years! Freedom has reduced!

I've had a note to do a post based on the idea that no matter who I work for, a company, or my own company, that I "always work for myself". I'll include a link to this essay for sure. Right stuff. www.WildWildEastDailies.blogspot.com

A very successful "free intraprise" system that creates very free small groups is underway in the Forest Service. The focus is not on external products and services, but on doing the work that is needed by the line officers. Biological surveys, Oracle database management, trail building, planning, etc. "Enterprise teams compete for these jobs. Line officers with budgets can hire outside contractors, one of several competing enterprise teams in that business, full time employees, etc. The enterprise teams have most of the freedoms of a start-up. They have internal "bank accounts" that they can carry earnings over from year to year. This provides security which allows risk taking. They chose their own products and services and set their own prices. They choose who to hire and have limited ability to fire (from the team, not the Forest Service) team members who do not work out. This is an area of weakness, but not enough to crater the system.

The result is very good. The intrapreneurs (people who behave like entrepreneurs inside a larger organization) are 1.8 times as productive as the average government employee and far happier. They tend to keep working after becoming eligible for retirement. Most say they would never go back to the old bureaucratic way.

There don't seem to be any fist fights. One team got too big and is breaking up into smaller enterprises.

The system was designed based on the book "The End of Bureaucracy and he Rise of the Intelligent Organization" by Gifford and Elizabeth Pinchot. Note: I am Gifford) It has also been applied successfully in a large semiconductor firm.

The largest organizations (nations or the entire planet) bring about order using self organizing systems. They do so because, for all the reasons in the article, the chain of command has limited ability to use the intelligence of all the members of a large system. While decision making is concentrated, intelligence is widely distributed, approximately one brain per human.

Very large commercial organizations are possible, but they must use self organizing systems, which involve freedom of the parts, as a major element of organizational design. After all, nations gave up on hierarchy as the method of organizing their economies with the fall of the Soviet Union. Even though it is large, the United States allows entrepreneurs considerable freedom. Large companies can too.

Re: Size of enterprises see: The Dunbar Number at Life with Alacrity:

http://www.lifewithalacrity.com/2004/03/the_dunbar_numb.html

Great observation and well presented.

Try this if you want to get to the source (even though it will most likely make no sense to you on first blush)

"The desire for guilt produces the fear of love"

or to say the same thing in Ken Kessy's words

"The desire for the mystery is greater than the desire for the answer"

TheD

wonderful analysis...

I can so much relate to this one. I work for a medium size business (~500 employees). I have been termed cowboy, SWAT and such - it took me around 5-6 years to realize I am more of an entrepreneur than a big team player. I have an idea and I want to execute it to full extent.

Problem is, after all, software is for a business and to run a business you need non-technical business mind, one that understands customers, their needs, marketing, sales & support. Thus a need for that other half. I am still trying to find one :)

Kapil Tundwal http://ktundwal.blogspot.com

I'm just skimming the article right now, but I hit a couple tiny errors that stopped my reading flow. For example:

"The only people who eat what humans were actually designed to eat are [a?] few Birkenstock-wearing weirdos in Berkeley."

commented on jeff atwood's blog in regards to this - i think he misinterpreted the point of your article to some minor extent (and so did I since I read the quotes only at that stage). he seems to have homed in on a few particular points (which may have admitedly been too much of a generalization on your part - everything is open to interpretation), but I loved your discussion of the tree concept.

hoping this doesn't put you off (I'd imagine it bought a lot more attention to your site and let you tighten up anything that could have been mis-interpreted), keep up the great writing.

Very interesting article. Just last year I left a big (for New Zealand) insurance co to work in a small start up internet co. I have no security, a small salary but a much more fulfilling job.

There's pizza, and there's junk pizza. I think you mean the latter. A good pizza can be pretty healthy -- certainly it doesn't seem to do the Italians any harm in terms of life expectancy.

I'd say working for a company is more like being in a circus than a zoo. You have to perform silly tricks and act like you wouldn't rather be napping.

What a load of bullocks. Humans are a social species. They are supposed to work in groups, and any group needs leaders and followers - one is not more important than the other. Comparing the life of a "company man" to that of a lion in a cage is a straw man argument.

Some people, myself include, are more than happy to put in our 40 hours at our jobs (I happen to enjoy mine), and than get to go home and define ourselves in ways not related to our careers.

Your insistence that being a startup founder somehow makes you more human is arrogant and offensive.

I never said anything like that.

But you did make the analogy of lions to innovators. I confess I prefer the notion of hunter-gatherers and "traveling light" as opposed to Greek Phalanxes requiring entire cities to eat at night. So I give you that analogy.

But the biggest obstacle to me in becoming a startup founder is health care. The difference between a W-2 job with benefits and self-insurance medical is huge. Like I-could-lease-a-Laborghini-with-the-difference huge. That's not all. Medical self-insurance goes up a lot: The costs went up double-digits-percent every year when I was a contractor. And I have no medical problems! PG, you should write about this as a barrier to startups.

Brilliant essay!

Rethinking the prevailing view on work is key to future success. Being driven by interestingness is more important than title and money.

My experience indicates the ideal group size to be five. The same holds for board of directors. I will not serve on boards larger than five,

Nice to see the first footnote in the article. However, unfortunate that it needs to be there at all. Keep up the excellent writing.

My employer, a large pharmaceutical company, even uses the word "ONE" in it's intranet portal; oneabc.abc.com

> Birkenstock-wearing weirdos of Berkeley

Did anyone else find this as funny as i did?

Could you pick a more mainstream brand of shoe Paul?

I've made this mandatory reading for the entire leadership team at my company. Great writing.

Rick Bullotta/CTO/Wonderware

Can you make it so that recent comments are at the top rather than the bottom?

I agree with your observations on the size of an ideal (human) group. So does Jesus.

I am curious to hear you relate this to Jesus. Please give me details.

If it wasn't implicit, I was just pointing out that he led a group of twelve (excluding himself of course), no more.

Bravo man ... bravo... I am now going to consume your entire blog from now on.

My comments relate to Graham’s Cliff Notes summary of his essay, rather than his underlying essay.

Every profession has a pecking order. The fact is some people are better at doing X than other people are. There are at least four factors that could determine where one sits in the hacker pecking order:

1. How smart you are

2. What you have contributed

3. Do you have the guts to start a start-up?

4. If you do, were you successful?

On that score, Graham does very well in the pecking order. He’s clearly brilliant. He has made numerous contributions -- spam filters, author of the two leading books on LIPS, ARC, just to name a few. He had the guts to start a start-up and he was successful, having sold it to Yahoo. The only thing he is missing is being co-founder of, say, Google or Microsoft or Apple or Adobe.

Graham says people think his group are elitists, implying the opposite, that they are not. Of course they are elitists! But why is that bad? Graham does not like Java is that it was designed with training wheels, to prevent programmers from doing stupid things. That may be necessary for some programmers, but it is not necessary for Graham and his target market -- the very top programmers, who do not need training wheels, who do not need to have the language prevent them from doing stupid things. (Have you ever met a LISP programmer who was not very very smart?) His primary partner, Robert Morris, is a professor at MIT. These are two uber elitists.

Graham likes to write essays that challenge orthodoxy and he is very good at doing so. See, for example, his essay “Mind the Gap,” in which he argues that a higher rate of income inequality may be a good thing rather than a bad thing (www.paulgraham.com/gap.html). It’s hard to think of a more controversial proposition than that. So me thinks that Graham’s next challenge-conventional-thinking essay should be why elitism is good, not bad. And he does not have to start from scratch, he can read “In Defense of Elitism” by William Henry (http://www.amazon.com/Defense-Elitism-William-Henry/dp/03854...).

So why does Graham want to portray his group as non-elitists even though they clearly are? Because elitism has a bad rap. Some elitists are smug and arrogant but I do not think Graham is. I am pretty smart, or so I have been told. But I do not think I am as smart as Graham. When I read Graham’s essay, I feel, “OK, this guy is smarter than me, but he is not arrogant about it, he is laying out his arguments in a clear concise logical manner so that others, almost all of whom are not as smart as he is, can follow his logic.” That is not arrogant.

Some people criticize elitists if they do not give something back. Graham is clearly giving something back. His essays are free. If you don’t want to buy his book, you can read all of them (I think) online for free. He is giving ARC away for free. Although Ycombinator is hardly non-profit, I am still shocked at little equity they take (an average of 6 percent) in exchange for what they provide.

Graham argues that Ycombinator would not exist if all it did is fund “rare geniuses.” That is true but misleading at the same. I have been an entrepreneur all my life, so far successfully. The thought of working for the man, I would just jump off a bridge and die. But starting and running a company is a bitch, even for someone like me who is genetically programmed for it (and for nothing else). Our society glorifies entrepreneurship, which is a sign that our society is more advanced than others. In one of his State of the Union addresses, President Reagan urged every schoolchild to consider starting a business. The fact of the matter is that few people are suited to starting a company. You have to be good at a whole lot of different things; most people at best are good at one thing. You work all the time. There is currently a best selling book, The Four Hour Workweek. Yeah, right. Try the 14 hour work day.

Few people are suited for this, and that includes gifted hackers. Few people have the balls to start a company and that is a good thing -- perhaps 2 percent of the population is suited to being an entrepreneur, I doubt it is as high as 5 percent. Perhaps among hackers, or gifted hackers, the percentage is higher, but I can’t imagine it is 10 percent. Most people are better off working for someone else. That fact, by the way, creates a lot of opportunity for the few who are suited to start companies.

Ycombinator’s thesis is that there is an untapped market of young entrepreneurs who could start companies but do not and will if they are funded by Ycombinator. Yes, there is such an untapped market, but it is not as large as Graham implies. Let’s assume Ycombinator funds thirty companies a year, twice a year, and each company has three hackers. That’s 180 people. A rounding error of a rounding error, since every year more than 4 million people reach whatever age is the minimum age for starting a company.

As far as I can tell, Ycombinator has been successful. There are several reasons for this. There were the first to market, that always helps, people think of them first. Graham’s awesome reputation and his Web site are extraordinary marketing machines, so they get the pick of the litter. On the other side are several brilliant people with a variety of backgrounds who are not virgins, they actually have done it before; in short, they have the perfect background to pick the right people to back. So Ycombinator may not limit its investments to ideas proposed by “rare geniuses,” but the caliber of the people they back is so close to genius it may be hard to ascertain the difference. More than anything, Ycombinator has been successful because they are able to attract an extraordinarily talented group of applicants from which they choose a few to back. If Ycombinator started funding the “average” hacker rather than rare geniuses or close to it, I guarantee Ycombinator would be a flop. Graham’s statement that they are not limited to funding “rare geniuses” is a misleading as the Dean of Admissions of Harvard or Stanford stating that everyone and anyone should apply for admission to those universities. The fact of the matter is that it is damm hard nowadays to get admitted to Harvard and it is damm hard to get funded by Ycombinator, even if you are really really smart.

James Mitchell jmitchell@kensingtonllc.com

in response to Notes: 1 why then always talk about evolution like its some mysterious magical force able to "design".

"I'm sure if I studied it more intensively it would all make sense"

Holy shit this is inspiring.

My comments relates to Graham’s Cliff Notes summary of his essay, rather than his underlying essay.

1. How smart you are

2. What you have contributed

3. Do you have the guts to start a start-up?

4. If you do, were you successful?

James Mitchell jmitchell@kensingtonllc.com

What about Ken and Dennis.

An enlightening article with equally interesting comments.

pg is an enlightened man

> producing fresh vegetables doesn't [scale]

The reason starchy and hydrogenated foods dominate is agricultural subsidies, not so much natural economies. If you killed corn and soy subsidies a whole lot of junk food would disappear. Fritos are only cheaper than sauerkraut because you paid part of the frito bill when you filed taxes. Healthy fresh vegetables are most definitely mass produced.

Running with the metaphor any way, government has a major role in perpetuating huge corporations and preventing the emergence of smaller ones. SarbOx is an obvious example. OSHA makes it very hard for a small group of guys to set up a manufacturing shop, even if they know how to make it safe. Read this guy on how it's basically illegal for him to run a small, traditional farm: http://www.mindfully.org/Farm/2003/Everything-Is-Illegal1esp...

It's true that farmers of commodities like corn and wheat have the government wrapped around their fingers. But the reason they're big enough to do that is that their market scales. No one has invented a fresh carrot that can sit on a shelf for six months the way a candy bar can.

I heard a radio show (I think it was NPR) about this recently. "Speciality Crops" don't get much in the way of government subsidy.

What's a "specialty crop"? The general rule appears to bet that if it is a recognizable vegetable that would appear in it's natural form on your dinner table, it's a specialty crop. Interestingly, many of the nation's (and world's) specialty crops are either locally grown, or are shipped in from California.

If it gets processed into something no longer recognizable as food, it's not a specialty crop, and is most likely getting subsidized.

Ironically, the government recommended food pyramid is the inverse of government subsidies.

Only 1/3 of US ag is subsidized, but its a disgrace. The programs damage the environment, cheat consumers, and pay the wealthy. Off the top of my head, the major program crops are corn, wheat, soy beans, cotton (!), milk, peanuts, and sugar.

More here: http://www.freetrade.org/node/609

"in 2005, farms with average household incomes of $200,000 per year accounted for 9 percent of all farms but received 54 percent of government payments."

The subsidies follow the disproportionate voting power and political influence of the relevant mid-west states, not the inherent marketability of corn.

Veggies and nuts are mostly grown at massive industrial operations in California with desert sunshine and irrigation systems. It's definitely a scalable operation not much different from corn and soy.

Not to nit-pick, but carrots keep really well even at room temperature. That was the whole point of root cellars. Veggies store fine. Lots of "fresh" stuff at the store routinely comes out of 9 months in cold storage, and that's without the complex chemical processing and packaging applied to the junk food. Then there's freezing and pickling.

The prevalence of crap in the US diet comes from a deliberate government plan to pump out more cheap calories. It might have made some sense in a Marshall plan world, but we're still stuck with it. That's the story as I understand it.

>The prevalence of crap in the US diet comes from a deliberate government plan to pump out more cheap calories.

It was deliberately done by the government? I probably shouldn't be surprised... I always assumed it was the corn farmers pushing for subsidies that did it. Do you have any more information on this?

http://en.wikipedia.org/wiki/Earl_Butz#Career

> Butz revolutionized federal agricultural policy and reengineered many New Deal era farm support programs. His mantra to farmers was "get big or get out," and he urged farmers to plant commodity crops like corn "from fencerow to fencerow."

movie recommendation (and accidental pun on parent poster): King Corn (http://www.imdb.com/title/tt1112115). Enlightening.

Another related movie (was on Sundance this past weekend): http://www.we-feed-the-world.at/en/film.htm

I agree with the assay, and I have personal experiences regarding workgroups (mostly hackathons) that match precisely the (1) small group requirement -7 to 10 people-; (2) the avoidance of any tree structure: horizontal and unguided development; and (3) the avoidance of mediocre people.

Yet I have a concern nagging my mind about the reduction of freedom in hierachies. Ender's Game book seem to argue the opposite: that giving general orders to a group while leaving the implementation details to each member is the best of both worlds.

[Remember the school master Graff describing to Ender the hierarchical army he had built with his toon leaders: moving in synchrony, yet each piece showing initiative and creativity -as opposed to the single-minded, single-track insector army lead by a single queen.]

There was a book I read recently...I can't remember the title, but it was one of those generic "How to manage for innovation and creativity" books, by a famous author.

Anyway, one of the examples used was that most-hierarchical-of-hierarchies, the U.S. Army. It talked about how in Vietnam, the U.S. got its butt kicked because basically "no plan survives contact with the enemy." The army would have these detailed marching orders, but then Vietcong would come out of nowhere and blow the plans to shambles.

The army's solution was to include a "Commander's Intention" section at the top of every order, which was a one-sentence description of the overall objective that the military hoped to achieve. At high levels, it would be very general, like "Establish air superiority over the region between the XYZ and ABC parallels". At low levels, it might be very specific, like "Capture the hill overlooking the southernmost runway at XY Airbase, to allow the 3rd mechanized division to enter the base without fear of artillery attack."

The crucial point was that if conditions on the battlefield ever made it impossible to carry out their orders as written, officers were authorized to use whatever creativity and judgment necessary to achieve the Commander's Intention, including giving new orders to their own subordinates. So you had plans, but each subunit knew where they stood in relationship to the whole and could adapt on the fly to take into account changing battlefield realities.

Anyone here in the military that could confirm or deny this? Anyone know what book it's from?

Thank you Mr. Graham.

Your essays never cease to inspire, delight, and amaze.

I feel VINDICATED upon reading them.

Always precise, yet loosely coupled.

Ideas and concepts that seem to exponentially expand within the mind of the reader.

I have personally experienced feeling like an entirely new being after going to work for myself . . . and I am going on my sixth year.

Why do we have to compare programmers with wild animals?

Because they smell like them.

I haven't read this yet but I know its a damn good essay just by the title.

Reply on news.ycombinator.com