Hacker News new | past | comments | ask | show | jobs | submit login
Nourishing a Culture of Failure (whattofix.com)
68 points by DanielBMarkham on Feb 9, 2011 | hide | past | favorite | 41 comments

Very good post, lots of good points.

One thing I remember reading that really stuck with me - most people externalize failure and internalize success. Meaning, when things go wrong, they point to external factors. When things go right, they credit their own skill and ability.

The opposite way is to externalize success and internalize failure - that means claiming responsibility for it whenever you don't get desired results, but being very skeptical of your successes and looking for where external factors broke your way so you don't get too high on yourself.

Turns out, externalizing failure/internalizing success makes you much happier but much less able to produce results.

Whereas internalizing failure/externalizing success makes you much less happy but much more able to deliver results.

Now note, this only applies in cases where your incentives come from producing results, as opposed to convincing other people you did well. That's the trouble with middle management and with politics - you win by convincing people you're doing a good job, more than actually doing a good job. That's very conducive to blaming external conditions for everything going wrong, and claiming personal success for everything that went right... which is much less conducive to results.

Nasty, huh? Actually that's why self-employment/entrepreneurship is so cool - we get paid for results, not for convincing anyone we're doing well in spite of the results. Thus, while it's still hard to claim every failure as your own and be skeptical of the successes, we do get compensated better for it.

Actually, I like your post better. Its a quick summary. The original was more of an essay, dense and long-winded, ending with "I guess you can't explain it". You just did explain it.

> One thing I remember reading that really stuck with me - most people externalize failure and internalize success.

I'm not sure of your definition of "most people" but how do you square this with the impostor syndrome that is reported to be widespread among highly skilled people? It's the exact opposite of what you claim as the default psychology.

One question pops up from your post: What happens when you internalize both success and failure?

Educated guess: mood swings.

When you succeed, you're on top of the world and can do no wrong. When you fail, you're a loser who can do nothing right.

There is a big difference between saying:

"Wow, I really screwed the pooch there. It's going to take a while for me to live this one down. I better get to work."


"Wow, I really screwed the pooch there. I'm worthless and I can't do anything right. I should go drink."

One of those, I think, is a healthy way of accepting responsibility for failure. The other is not.

The same, really, goes on the upside.


I was making a shoot-from-the-hip guess. The difference between the two approaches which you mentioned is a correct sense of proportion. Of course, the same applies to the other ways in which one internalizes/externalizes both success and failure.

It's got nothing to do with proportion. Assuming you don't consider suicide a rational choice, the first is always going to result in better outcomes for the individual who screwed up than the second, no matter how dramatic the failure.

I mean, proportion has something to do with how publicly and how strongly you repent for your mistakes, but I don't see any case where it is in your best interest to accept that you are worthless.

To be clear, I can see cases where it may be in other people's best interest to give up on you, but I don't see any case where it's in your own interest to give up on yourself.

Even when you fail one could argue that it's not all bad - at least you would feel in control of your life. What about when both success and failure are externalized? I think that kind of helplessness would be a much worse feeling.

Perhaps those suffering from paranoia externalize both success and failure. Failure is proof that They are out to get you. Success is proof that They are setting you up for a fall.

Which sounds like an exact description of the emotional life of a founder.

I see that you've posted about starting your venture: http://news.ycombinator.com/item?id=1519002

So you're speaking from experience. Have an upvote.

I've started my own business, too. I commiserate.

Implosion, followed by a singularity.

In general, I agree that if a culture is to grow it needs to be capable of taking risks and making experiments. When every experiment succeeds, either you weren't innovating enough or you're lying and now supporting a failure.

When he starts complaining about social protections though, I'd draw the opposite conclusion:

Need health insurance? No problem, we'll make sure that's provided for.

The article seems to me to imply that this sort of protection is a brake on innovation. I won't disagree that it has the potential to put a brake on innovation through additional taxation (though not as currently implemented in the USA, where total healthcare spending is far from efficient), but as a concept it enables failure and risk taking.

We talk about startups being ramen profitable, and that's fine. But my first employer was a startup who, when he started, had a wife and young kids - he needed 4x ramen + healthcare + education profitable. He's since made a success of the company and sold it, but how would he have been able to responsibly take the risk to start the company if its failure would have left his family not just bankrupt but without any form of medical provision? What if he also couldn't have afforded to educate his children? Would that have been good?

People are best enabled to innovate when they can be free to fail without a disastrous cost to their career and their lives - this is the lesson of the invention of the limited company and the prosperity that resulted from that. Social security safety nets such universal healthcare enable innovation over a wider pool of candidates because they reduce the harm of failure to the innovator and their dependents.

> (referring to healthcare) as a concept it enables failure and risk taking.

I cannot agree enough. Sorry for the sidetrack, but I have always thought about the scenario where say a person who is still far away from standard retirement age (maybe by about 20 years till medicare kicks in) has, say, just enough saved up to carry her through the rest of her life. If she doesnt have a working spouse with healthcare, she is forced to take a job. Whereas if healthcare was provided regardless of a job, it frees her from the necessity to work as well as opens up the job to somebody who needs the income as well as the care.

One consequence is that someone with specific life circumstances preventing her from leaving her job will now be able to do that to pursue something that might be better. Another consequence, and one that is unintended but virtually unpreventable for a government, is that someone with an idea he is less sure of and which is less likely to succeed will be pushed in the direction of quitting his job and pursuing his unproductive idea. My thought is that when a government does something like this, it pushes everyone further in the direction of risk, similar to when the US Government promised to back the banks in the case of impending failure. And like the banking situation, while it is possible that a lot of the risky decisions will pay off quite handsomely, a lot of them will end poorly, possibly enough to swamp the ones that paid off well. In the aggregate this could mean a lot of otherwise productive people working on unproductive projects, leaving less productive people doing the stable, productive work.

That's why I don't think the government is the best tool to encourage or provide incentives for people to quit their jobs and work on startups. The government can't filter well enough to cause more of the good consequence and less of the bad.

The argument is not that the government should provide incentives for people to quit their jobs and work on startups. The way I see it, healthcare in its current state with its unreasonable costs, is a problem. Even if the average earner chooses to finance her healthcare herself, she will not able to sustain it. So this is a case of the government removing a roadblock so that the individual can weigh other factors to decide whether or not a job/startup/retirement is right for her.

Single-payer healthcare has some selling points, but allowing workers to finance others' early retirements is not one of them.

The assumption is that the early-retiring worker has also paid in to the pool while accumulating the wealth.

I literally cannot up-vote this enough. An "innovation lab" has to be based on failure tolerance and the process to achieve that has to be customer driven. (i.e. the lean startup methodology)

Big companies are notorious for their low tolerance for failure, but I've been surprised to discover that many research institutions are maybe even worse. An academician needs a steady stream of hi-quality publications, i.e. stuff that people generally agree are worthwhile contributions. Truly new ideas are easily buried, because they don't provide a steady stream of results, and there is no time to pursue them. I wonder if there are any labs that actively try to fix this somehow.

He's right about giant corporations becoming stifling to innovation, but I disagree about Norway. There is no reason why Americans should need to pay over twice as much per capita to provide the same barely adequate health care for their employees!

Startups are even more vulnerable to high monthly startup costs than they are to lack of financing. You can get by on credit cards working out of your garage, but not if you rent a large unit in an office park and not if you need to shell out hundreds per month per employee for health insurance.

Right now, college students are creating most startups. That's not only because they know the newest technology. They have no responsibility to pay for or protect a family. The single greatest innovation that would prompt the formation of more mid-career American startups would be any reform that takes the cost of health insurance off of corporate balance sheets. It seems most likely that would be some kind of "Medicare-for-All" system, though other innovations may be possible.

By "barely adequate" health care, I mean that we should focus much more on nutrition and lifestyle and less on drugs and prosthetic organs. It doesn't matter how many surgical robots a country has if we make food choices that amount to poisoning ourselves and lifestyle choices like we're trying to turn ourselves into veal. Forcing people to pay cash on the barrel-head for health care won't improve that because humans make emotional, not rational decisions.

I liked this article a lot because I've just came to the same realization a few days ago. To encourage innovation you have to have a culture that encourages it ... that puts the emphasis on doing amazing things instead of not fucking up.

Big companies are all about not fucking up, and when there is a fuck up, bending someone over the barrel for it, so that it never happens again ... which is why they are places where innovation goes to die. I mean why would you attempt some crazy initiative when you'd just get dinged/fired if it didn't work out.

Google's masterstroke was is in empowering people to try new things without fear of failure with their 20% rule.

So whenever somebody asks how you/they can be innovative ... what they really should be asking is "how can we give people who work there, the freedom to try new things without fear of failure".

Any organizations that will do whatever it takes to possess that culture will generally succeed at this 'innovation' thing.

Failure also works well at company scales. When many hedge funds failed in 2008, some great people were released into the wild. As a result, NYC is now the #2 internet startup city.

The most limited, precious resource in the world is people and their time. So admitting failure and moving on is the best thing, continued death marches are the worst thing.

I was in NYC during that time period, and very few people (approaching zero) at the hot startups were from hedge funds.

You maybe had a DBA or sysadmin from a failed investment bank.

In fact, it seemed like a non-trivial number of startup engineers and product people in NYC were people who were sick of living in Silicon Valley.

Maybe it prevented certain smart people from going into hedge funds in the first place, but I wouldn't be so sure. Firms like 2 Sigma and Jane Street were still hiring, and actually had more startup-like attitudes towards workspace and technology than many of the startups.

> But to have more successes we must necessarily have more failures

I'm not sold on this. There's seems to be the popular belief that the ratio of failure:success is fixed, so to get more success, you have to get more fail too. Along with the "fail fast" technique that always reminds me of "the tortoise and the hare" fable. There seems to be an obsession with going as fast as you can, and not worrying if you fail.

Feynman's problem solving algorithm just seems more sensible: 1. write down the problem 2. think very hard 3. write down the answer

Do you have a reference to Feynman's problem solving algorithm? I've never heard of it.

One counter example is Feynman's Rule: "If the sign is wrong, change it." There is a long story behind that, about Feynman heading off to present a big paper, asking a grad student to "check the equations", grad student calls with terrible news that there is an incorrect sign in the middle section, Feynman says "I feel it is right." and delivers the paper, and grad student finds a correcting error a few pages (and a few days) further on. Basically, if you understand the physics, you know the final sign, and (after the first time) there is no real need to verify every sign in every equation along the way. That story is really far from "a miracle occurs" and the answer is written down.

Feynman also was extremely interested in how he came up with breakthroughs, almost obsessed about it, and did some meta-research about creativity. He told one story about a solution to a superfluidity problem he came up with while walking down the street. It was apparently bubbling up in his subconscious, and suddenly he stops in his tracks. He raises his hands like holding an imaginary picture frame, and makes a motion with one hand: "If I just tap it here - Yes! The answer fell into place." It drove him nuts that he could not ever do that again. He said many times after that he held his hands up, tapped the problem he was working on, and ... nothing.

> Do you have a reference to Feynman's problem solving algorithm? I've never heard of it.

Hmmm, just googling for it associates the algorithm with Feynman. You know, I'm not sure if I actually read it in a Feynman book or if it's random trash I read on the internet. Wikiquotes attributes it to "a fellow physicist".

Murray Gell-Mann came up with it.


Doesn't fit with the popular belief that "if you just apply yourself enough, you too are going to succeed".

It's a rather common delusion :)

Great article. A good example of this flaw at work is energy policy. Too much activity centered on activity to make people feel like something is happening. Not enough on tax policy that would incent people to innovate on their own. Politicians want to be standing next to the money as it is being handed out.

A lot of people keep parroting Google's famous "fail quickly" model of encouraging innovation. But they seem to forget another well known fact: Apart from their core search product, Google has not produced another successful product. I think the two issues are related and a blind acceptance of Google's philosophy might be more harmful than good. Especially if your company does not have the kind of cash reserves that Google has.

You can't use Google to prove anything. You are not privy to their evaluation of success and failure. But don't feel bad, very few are.

What about GMail?

For that matter, what about Android?

Android was an acquisition

it wasn't that successful before the acquisition

So success defines innovation?

Paul Buchheit was an acquisition in a sense

Your message is somewhat contradictory. You describe the need for failure, but what you ultimately describe is emotional failure, perhaps with some external, material loss. In much of the world and time up to this point, failure has meant a significant risk of irreversible loss to self. As in, for example, health that cannot be recovered. A social judgment or status that will not be forgotten and cannot be escaped.

So much of the current talk about innovation, entrepreneurship -- I'm thinking of the lastest U.S. Administration program. A basic reason for the existence of this board. It takes change as a basic premise, a fundamental good.

Here's the secret: Social structures develop and evolve to promote stability. It's that stability that allows a greater combination and coordination of effort in more advanced works. That enables and makes sense of deep and long term investments, material and intellectual.

Those patterns can become constraining. Society reaches a limited "set point". Then change eventually overcomes inertia and shifts society out of this local maximum and back onto a course of less limited growth, of renewed prospects.

And since society is (increasingly) complex, opportunities for change exist, in aggregate, constantly, although in greater or lesser degree in different segments at different times.

Failure becomes desirable, when the benefits of trying outweigh the costs. When it won't kill you, or your family, or irretrievably break you -- then maybe you'll give it a go. Or when there is nothing to be gained in not trying.

So, move past "Change!", and ask how to make change happen. Help its agents get unstuck from local maxima. And ask what the goal is: An endless, exhausting treadmill of ever increasing change? Or a happier medium, when people can create and feel useful and e.g. still have a family and kids before their 40's?

This may be one component in why you don't hear so much about "blockbuster, breakout" start ups in Norway, to cite your example. Because people can grow their business to the point of having a good quality of life, without feeling that they must be the biggest because it's "winner takes all".

(This is speculation on my part, but based upon some years' observation, reading, conversations, etc. And my speculation about Norway is quite limited and takes your citation of their startup rate as fact, because I've never been there.

Also, admittedly, they have a lot of oil revenue at the moment, which certainly doesn't seem to hurt their aggregate well being. Something which is due in part, again, to preventing a "winner takes all" structure sought by "big oil"; regulation that has been cited and discussed here, before.)

Applications are open for YC Summer 2021

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact