Instead, take it as input and context, and then form your own opinions (and biases). We knowingly ask for biased opinions of our friends, family, and coworkers all the time; apply the same process here.

http://en.wikipedia.org/wiki/List_of_cognitive_biases

That's a good point, and I definitely have this tendency.

On the other hand, I don't think a list like this is that helpful. With as much experience as Sam Altman has, I'd prefer to see him do deep dives into particular items on the list, with an attempt to establish causality between the item on the list and the success of a particular company within a particular context.

Why did this particular strength help this particular company at this particular time in its lifecycle, facing this particular environment? Has that strength proven a weakness in a different context?

That seems more useful to me than, effectively, "Be tough, be smart and care a lot", which is great advice, but not something anyone with passing familiarity with startups doesn't already know.

It's clear we mostly look at success as an example, but then again there could be a cause/effect confusion in the accusation of "Survival bias"

I think PG's advice "Don't Die" combined with some of these traits like They generate revenue very early on in their lives, They keep expenses low and They are focused on growth logically seems to lead towards a company biased towards survival.

It would be interesting to build a Bayesian model of these traits.

Also, I feel like akinator.com is using a Bayesian tree. Would you confirm this? What algorithm do you think they are using?

I thought PG's explanation in "A Plan for Spam" was pretty good: http://www.paulgraham.com/naivebayes.html

Also, I feel like akinator.com is using a Bayesian tree. Would you confirm this?

It's possible, but I doubt it.

What algorithm do you think they are using?

A plain binary search tree would work perfectly. I guess it could be broken by people entering deliberately wrong data, but..

It's clearly some kind of search tree, because the questions are influenced by previous answers.

You could do a Bayesian version, but I'm not sure it would perform any better.

I'm trying to list the usages of Bayesian inferences in industry. So, there is spam filtering, I bet there would be recommandation engines (ex: amazon's product recommandation, ad targeting). Would you know of other uses?

Do you know of some kind of tutorial of how one particular bayesian engine is applied to a particular dataset in a commercial venture?

Also, I have been looking at weka (http://www.cs.waikato.ac.nz/ml/weka/), and j48 in particular. j48 is an implementation of the C4.5 algorithm, (which is close to ID3 and C5). Here is how I learned about C4.5: http://www.cs.umd.edu/~samir/498/10Algorithms-08.pdf. Would you happen to know if one of these algorithms have commercial applications?

In a different subject, Judea Pearl has been awarded the Turing award for using Bayesian networks to calculate probability of causality. Which is different than traditional Bayesian analysis, where we calculate probabilities of a property being true. In his theory, he claims one can calculate the probability of a property being a necessary condition, the probability of a sufficient condition to another property (and the probability of both combined). I'm really having trouble understanding the ins and outs of this too. Have you heard about an usage of this theory? Do you think it has a viable shot at having commercial applications?

"They are obsessed with the quality of the product/experience"

Maybe "competitive advantage" is not so directly addressed, but I feel some of the other points touch on it, like "do things that don't scale".

And "obsessed with the quality of the product/experience" is BS wording. It means nothing. I could say the same thing about any subject without making any distinctive point. It's not about being "obsessed", it's about "making the right choices in terms of priority regarding the product/experience". That's more like it, and that says that decision/compromise is needed in everything you create.

I made the mistake in my first years. I made consultingware products which didn't grow but paid expenses and helped me keep my company alive and even grow it very slowly. Then, I dropped the non-growth products and focused on a single growing product.

Same goes for Jeff Bezos. Yet, incredibly successful. That's a few major counter-examples.

Bill Gates is certainly ruthless when it comes to business. You don't want to come to him if you work at microsoft and your project is failing.

However I'm sure on a personal level he is very nice, and he is certainly doing a lot of good with his foundation.

More on topic, this seems like the very definition of cargo culting to me. Why not just focus on what makes sense for your company? There generally is no simple catch-all solution, especially not to success.

It would surely help if the survivors themselves would acknowledge survivor bias more. Not seeing much of that.

Still, I think most of these points are either obvious or tautologies, and many of them can cause harm when they are focused on obsessively.

And, so I'm not just another critic, let me offer two points I'd make about startups that stand, somewhat, in opposition to the points made here:

1) Successful companies ship. This is often in direct opposition to obsessing over quality: you have to accept that your app, which is a piece of shit, needs to get out there, so you can get feedback, before you make it a gleaming cube of perfection.

2) Random chance is going to factor in your startups success far more than you'd like, so building a startup that will not die no matter what (rev/run > 1) is going to give you a better shot at making it over the hump. This often means putting up with uncertainty and not knowing exactly what you should be or will be doing. And it will be hard on you, mentally.

<Super-Succeess> is typically the outlier. You could have 100 companies with these traits, and 98 of them could fail. Its not clear with that type of failure rate you can really differentiate meaningful information, because the overlap of explanatory "data points" is just too noisy.

Super success is typically more of a black-swan. This is essay is less relevant in that endeavor. But it probably is not bad at all if it was merely aiming for "nice/er places to work". Or, "avoiding a bad bet" type of argument. That makes it closer to describing a "white swan", which again is not altogether useless. Its just not really what the title is proclaiming. Not that Sam would be unique in that regards, thoug. I think its fair to cut him some slack on that.

Of course, it begs the question: Do "super successful people" tend to eggagerate? Which is even itself an enlightening question. It seems many people believe in "fake it to you make it". But on the otherhand, many people who have "made it" tend to revert to understatement. And many more become successful merely with true confidence. So again...who knows.

> After any process that leaves behind survivors, the non-survivors are often destroyed or muted or removed from your view. If failures becomes invisible, then naturally you will pay more attention to successes.

another one:

> As the psychologist Daniel Kahneman writes in his book Thinking Fast and Slow, “A stupid decision that works out well becomes a brilliant decision in hindsight.”

I think it's pretty unlikely that he's not taking unsuccessful startups into account. But it's not a statistical paper, it's an essay. You can't attack the guy for not including likelihood ratios and p-values in his prose.

Sam Altman has experience with a particular group of companies. A small subset of companies out of a huge set of people who do business. In a wide variety of places.

This writeup certainly makes some good points but it is specific to companies of a certain type and Sam has not qualified that at all in his post.

"than pretty much anyone has seen companies at all."

I've been observing companies (as have many other HN readers) since before Sam was even born. (I started my first company right out of college and that a very long time ago.)

OK, fine. If you have lots of relevant experience, then write a counter-argument that's more substantial than "selection bias!"

I'm sure we'd all like to read a comment that outlined some specific, well-thought-out points of disagreement with Sam's arguments.

I see mostly opinions, not arguments. An argument is generally what follows if you ask someone "why?".

In the very first paragraph, Sam writes "It’s from personal experience and I’m sure there are plenty of exceptions."

What should be done to satisfy people like you?

Yet still one would expect that most quick decisions are bad decisions. I would conjecture there are far more failures that come from quick decision than successes.

But if the market strongly favors speed then it will select those few that just happen through chance to make the right quick decision.

http://youarenotsosmart.com/2013/05/23/survivorship-bias/

(Taken from reply below)

If lots of unsuccessful companies were frugal with money (say), then maybe being frugal with money doesn't help your chance of success.