RichardGibbons

Not for me. I'm interviewing for technical positions. That said, I do a tech-variant of this approach by asking people some of the websites they use the most, and how they'd architect them. Definitely the answer is impacted by both domain and the candidate's ability to concentrate under pressure. So, I have to factor in the candidate's background (e.g. people with graduate degrees in math typically come up with the right answer in under five seconds.) In terms of concentrating under pressure, if I'm not looking for the right answer, that's less relevant, and I do expect people to be able to answer challenging questions in an interview, because that's kind of the definition of an interview. People who are amazing, but have no way of proving it to me will never be hired by me. I would actually read quite a bit into this depending on the tone. Most likely, I'd interpret it as, "I don't know the answer to the question or even how to approach it. But I think I'm very smart, so my ego won't allow me to say 'I don't know', so I'll bluff by pretending that it isn't worth my time." So, you'd have to be pretty good on most other questions to survive that answer, since I don't like working with people who think highly of their own abilities, yet can't admit when they don't know something and believe in bluffing. I want to collaborate with my colleagues towards a bigger goal, not constantly engage in pissing contests and trying to determine the difference between when my colleagues are bluffing vs. when they actually know something. (All that said, I might have taken that sort of bluff approach 15 years ago. Since then, I've decided that I know how smart I am, and I generally don't feel the need prove it, though sometimes when I'm uncomfortable or feel attacked I'm an ass and forget it. :P )

The answer might be good or bad, depending on how the questions are asked. If it's with the intent of actually defining the problem, then that's good. If it's with the intent of being a smartass, that bad, and usually the difference is obvious when you're speaking with someone. I think some of these questions fall into the latter category, other than the the fact that each ant is walking along exactly one edge and ending up in a corner, which is definitely unclear. If that's clear, most of your other questions are irrelevant of course (and asking them would be pretty revealing.) The other thing that such questions reveal about you is that you maybe incapable of working independently, using Occam's Razor and knowledge that the problem most likely has a solution to find the most likely problem that I was asking. I tend to not like defining unimportant requirements for people but prefer to work with people who can figure out stuff their own. So, if you really asked all these questions, I'd probably think that you were totally incompatible to work with me. In most cases, the requirements are clear from the way it's expressed in the actual interview (though I admit that they weren't as clear here). For instance, in the real interview, there are pictures with cute little ants. :) Also, if you did ask how many ants were on each vertex (which you probably wouldn't since it would be obvious from the picture and how I phrased it while drawing it), I might say something like, "assume N ants on each vertex", since you asking the question clearly means that you think there's a good solution to the N problem, and that it's something that's reasonable for me to ask in an interview. :) (Particularly the ones about the definition of collision (unless you're ESL), since that term's relatively unambiguous in real life. Is it a collision if you have 3 cars smash together simultaneously? Is it a collision if you have a car smash into a stationary car?) Yes, matjone, the easiest solution is that that there are 2 paths where they don't collide (all clockwise or all counterclockwise), and there are 23 different paths, so 2/23 is 1/4. That said, the first time I saw it, I solved it this way: 1. It doesn't matter which way the first ant goes, because it's a mirror image, so suppose it walks towards the second ant. 2. For the second ant, it can either collide with the first, or go the other way, so 50% chance of collision. 3. For the third ant, it can either collide with the second, or go the other way, so 50% chance of collision. 4. Then 50% * 50% = 25%. Your fun comment is kind of telling too. People typically find it fun or torture, and if the candidate finds it fun, I'm more likely to hire them.

That's pretty funny. I wonder if for the last question, someone's ever said, "Because my adviser said that if I solved it, it would get me Ph.D." :) One question I've been experimenting with over the last few years is the ant problem: Imagine a triangle with ants at each corner. Simultaneously, each ant will randomly choose one of the two edges, and walk along that edge to another corner. What's the probability that no two ants collide? Of course, the interesting thing is the approach to solving the problem. It tells you a lot about whether they are the type of person who sees an unusual problem and gives up, or approaches it methodically etc. Getting the right answer is mostly irrelevant to me, but the process of arriving at the answer is interesting. Of course, it's not that useful for people who majored in statistics, since it's a trivial problem if you're well versed in probability. But in that case, you can just make it a tetrahedron rather than a triangle. :) I think some of the other Google/Microsoft questions are interesting too. Things like "How many windows are there in Seattle?" and "How many tennis balls can you fit inside a city bus?". Beyond that, I also test the ability to create algorithms to solve simple problems where there's a multitude of solutions ranging from trivial "correct but bad" solutions to "complex but good" solutions. For instance, creating an algorithm for factoring a number. Again, it's interesting hearing people's thoughts about why any algorithm is good or bad and ways of improving it.

LOL, I assumed you were talking about someone with a graduate degree. For me, the degree isn't what matters, it's just a filter, as ScottHall says. I have limited time to interview and the frequency of intelligent people is higher in those with graduate degrees than those with no degrees. I will hire people without degrees who are smart, but it's harder for them to pass through my filter to get the opportunity to sit in front of me and have a conversation. The "defining intelligence" question is an interesting one. For me in an interview process, it's a matter of how they answer questions that are slanted towards synthesis rather than than recall. I weight creativity and analysis very highly relative to skills. As such, I often disagree with other interviewers who focus much more on skills. My attitude is that if someone's smart, they can quickly pick up most skills, whereas the opposite isn't true. Of course, my approach has the problem that it excludes people who think well, but slowly, or who can't focus under the pressure of a job interview. Zippy, I'd agree that being smart and hard-working aren't mutually exclusive. Crip's comment, that the ideal is both, seems self-evident, so, beyond him correctly pointing that out, it didn't seem a very interesting thing to talk about. I'd also agree with your point of view, that smart, hard workers come up with more ideas than smart, lazy people. I think there's some sort of compounding thing going on there, too. Maybe an internal networking effect, where the more things you know, the better the chance that something apparently unrelated to a problem can be re-imagined to create a solution. If you work longer hours, you have more chances to put new nodes and connections into your mental network which increases the overall capability of that network.

Yes, I can see why you might prefer the hard worker. There could well be something to the industry as well. I'm in tech. Some of the factors that impact my 10 times multiplier estimate are: A smart person will come up with a good design rather than a bad one. As a result, all subsequent work will be an order of magnitude faster. e.g. imagine if youtube, in its early days, didn't have an architecture that easily allowed the identification of bottlenecks impacting scalability. If that had been the case, you would not have heard of youtube today. There are types of problems that only smart people can perceive before they happen. Avoiding these problems is incredibly important because of the leverage you get in solving the problem early The cost of an error can be huge, potentially in the millions of dollars. Smart people make far fewer errors. It's another leverage thing. A smart person will come up with new ideas that can grow into billion dollar businesses. Hard workers don't. Smart people tend to think more probabilisticly, perceive risk better, and understand things like sunk costs. This tends to lead them to make better business decisions and gives you a fighting chance in recognizing and responding to new competitors. When smart people need to figure something out, they'll generally come up with the right solution far more often than average people. Generally, with smart people, black swans are to the upside, while with average people, they're more likely to be evenly distributed or perhaps more to the downside. Smart people drive others to be their best. Average people, even if they're hard workers, slow everything down. I don't believe your point that hard workers figure out ways to do things better. Smart people are bored easily, which leads them to look for optimizations to make work more entertaining, whereas average intelligence people are much more willing to do repetitive jobs because they believe that nose-to-the-grindstone is the path to success. It's not totally on topic, but it feels like one of my favorite Bill Gates quotes applies, "I choose a lazy person to do a hard job. Because a lazy person will find an easy way to do it." All that said, when it comes to managing and retaining someone, it's much easier to have a hard worker. The discussion makes me wonder whether there's any tech companies who believe that hard workers are better than smart workers. Hmm, maybe Electronic Arts (putting the same sports video game out year after year.)

Wow, you and I should do job interviews together. You can take all the people who work their butts off and I'll take all the brilliant people, and we'll both be delighted with the deal. In my experience, someone who's really smart and works 8 hours a day will be ten times more productive than someone of average intelligence who works 12 hours a day. It's good thing for the world that we think differently, providing work for both types of people. Now we just need to find someone who likes to hire lazy, average or low intelligence people, and we'll be set... :)

The problem with this issue in Vancouver, at least, isn't that the crime isn't worth investigating. It's that the punishment for minor crimes is irrelevant. Apparently, there are something like 30 people who are serial offenders, with something like 50 or 60 break and enter convictions. According to the police, they can see the effects of these small number of people in the crime statistics. If a large proportion of them are out, the crime rate will go up significantly. If they happen to be in jail, then the rate will fall. Seems like you have to address that issue before expending more resources in fighting minor crime.

That's a neat world. Then, since I'm much bigger than you, I can beat you up and take your food. Hmm, actually, now that I think of it, I may as well enslave you, too. After all, it's not my fault if you lost the ovarian lottery by being smaller than me, or older than me, or having fewer friends. Totally fair, everyone can have what they can take and hold on to. What a great place to live!

Yeah, you could be right -- I could be totally clueless. If we put my cluelessness aside, I think Buffett's one of the clearest thinkers of our times, and also one of the clearest communicators (when he actually wants to communicate something). Furthermore, I think in his last few decades of his life (I'm optimistic :) ), he's doing what he can to make the world a better place. He's giving away huge sums of money, and working out the right strategies to put that into effect. Furthermore, he seems to be maintaining his integrity. And I also believe that he's talking about the things that he believes will benefit the country the most. Of course, in these sorts of situations, you may choose to toss Occam's razor out the window, Palantir. But it doesn't really make Buffett seem less credible, and not everyone will agree that your decision to ignore Occam's razor is an astute one. That said, they could be called clueless too, if you like. It does make things simpler. All that said, you are right that entitlements are a big issue as well that needs to be dealt with. As far as I can see, the numbers are indisputable. A combo makes sense. It's not about fixing government inefficiency, as many seem to say. That approach can't come close to fixing the problem. It's about demographic math. That said, feel free to call me clueless. I understand where you're coming from.

Or, maybe he just has a better understanding of the issues and what's important to discuss than you do, and you don't realize it.

Buying options seems very reasonable to me -- you have the potential for a big win but a small loss (100% is small when you consider the potential upside). The thing that I find interesting is that options are priced on volatility, not the value of the underlying, which seems to assume an efficient market. I don't believe in the efficient market -- I believe if I find a stock that is undervalued, then it will typically return to its intrinsic value over a reasonable time period. What's more, to me, the volatility actually isn't that relevant over a long time, because I believe that in a couple years, both a non-volatile stock and a volatile stock each have a good chance of trading at their intrinsic value. Thus, I'm happy to buy options on undervalued stocks with low volatility. All that said, the money management is the key. If you're betting on a the roll of a die, and get paid 8 times your bet when a 6 is rolled but lose your bet when a 1-5 is rolled, then it's a bad idea to put 100% of your money on a single bet. So, my rules are: 1. Bet small on any given option. You will lose frequently. 2. Estimate how often you'll lose (say two-thirds, of the time), and figure out how much you need to make from your winners to cancel out that loss and add a decent profit (say 400%). That should be your target. 3. Don't make the bet if you can't reasonably envision a profit greater than your target. 4. Only close out below your target if you're close to expiry, where "close" is defined as "the market's random gyrations are likely to have a bigger effect on the stock price than any return to fair value". 5. Don't necessarily sell out when you hit your target. If the strategy works, it works because of big winners. 6. If you like, after you've got a decent profit, you can sell in thirds or quarters, to take your risk of the table and play with the house's money. (You're fooling yourself if you think it's the house's money -- at that point it's your money -- but if this illusion helps you avoid taking small profits, it's probably fine.) The other thing that could be helpful is spending some time with a die recording the results and looking for losing strings of numbers. I think people in general believe random numbers are far more evenly distributed than they actually are -- the long term takes a long time to arrive. Just like you'll have a big series of wins, you'll have a big series of losses, and will feel the losses more than the wins. So, thinking through your feelings about a long strings of losses can be useful.

I found The Spirit Level a good description with evidence of why liberalism works. http://www.amazon.com/Spirit-Level-Equality-Societies-Stronger/dp/1608193411 The interesting thing is that it has reasonable evidence and explanations for the argument. That said, I generally also believe that fairness is good, social mobility is good, and that people generally grossly underestimate the effects of random chance. Typically, liberal strategies result in increase social mobility and reduce the impact of chance on people's lifestyles. Plus, killing people by, say, not providing them with affordable medical care, seems unethical to me. What's more, I don't think taxes and social programs are nearly the disincentive that conservatives make them out to be. Generally, people like to be perceived as successful, and wealth generators always seem to be perceived that way, while people on social programs are never perceived that way. And, from a purely practical point of view, I think that a low-regulation society leads to corruption and system gaming that leads to instability (see the 2008 crash, more the latter than the former). Plus, I think without liberal safety nets, society is unstable, since wealth will naturally gravitate to extremes, and eventually the people without wealth will decide to kill the people with it. So, a reasonable degree of regulation and liberal programs is a good thing. That said, I can understand the conservative point of view. It's hard to believe that a significant component of your success is luck, and it's hard supporting social systems that would cause your taxes to increase and reduce your relative social standing, even if you know that it's a good thing for society as a whole.

Yeah, I totally agree. The rational thing to do is to trust the scientist who have spent decades of their lives examining the issue, 99% of which agree that global warming is an issue.

Yes, I bought some flying out of Dulles Wednesday last week, and others from a west coast (non-airport) store this week. Our assessment was that ones from Dulles were OK, not up to the usual standards. We were guessing they were old. The ones from the west coast were great. My wife suggested that it wasn't worthwhile buying the Dulles ones again. Interestingly, the good ones cost about $17, while the inferior Dulles ones were $30.

I kind of think you don't get it if you think that your statement is some sort of refutation of what Liberty said. Also, there's no arrogance in saying something is unnecessary. Is it arrogant for me to say it's unnecessary for me to wear a pig on my head to survive? It's just unnecessary, no more and no less. I'm not saying that I won't wear a pig on my head, nor that the pig will probably get annoyed and squeal and kick at me. I'm not even saying that I'm not wearing a pig on my head now. But wearing that pig is unnecessary. They do. I think there's very few academics who would claim that they have proof of the absence of a a god. All they're claiming is that they have hypotheses that in combination suggest that a universe without a god is possible. That said, if you want to pretend otherwise because it doesn't fit your world views and ideas of what science is, that fine. Just don't be surprised when people don't take your argument seriously. (That might seem arrogant, but it isn't really. If you know calculus, is it arrogant to point out a math error of someone who, when they add 14 and 17, always gets 21?)

This is a bizarre thing to say.... I'm pretty sure that science has never claimed certainty that a higher being doesn't exist. Science doesn't claim anything with certainty. It's ok, you probably just don't understand what science is. Here's a primer to help you get up to speed: http://en.wikipedia.org/wiki/Scientific_method

After reading Sanjeev's writings for a decade, it would surprise me if he's timing the portfolio. I think he's doing the strategy that he thinks will provide the best returns. He recognizes the value of dry powder at the bottom of the market (and really, anyone who kept their eyes open during 2008 should also). Furthermore, saying, "nothing's really changed since January" makes no sense. I'd say more has changed in the last 5 months than most other 5 month periods I can think of over the last couple of decades. In fact, if this 5 month period doesn't meet your criteria for a change in investment thinking, then I think there's probably only a few that do. In that case, you should basically never have a change in your investment thinking. Which I guess is fine, but it's probably pretty non-optimal if you assume the economy will not affect any businesses you own. Moore's trying to help Sanjeev learn, but I'd guess that there's a fair amount that Moore could learn from Sanjeev. Nevertheless, Sanjeev will almost certainly learn more from Moore than vice versa. :)

The data says "there is enough supply -- the vacancy rate hasn't moved and the rents aren't going up. But the costs of home ownership have skyrocketed. Thus, home ownership is in a bubble." It seems odd to me that you want to interpret it as something else. It's largely explained by Ericopoly's point, that there's been a massive increases in borrowing. When rates are low, People are willing to borrow huge amounts of money to buy houses, but, for some reason seem to be reluctant to borrow money to pay their rent.

Real estate is a producing asset. That's how you know that it's overvalued. I think it has fewer influencing factors than most stocks.

This nicely sums up the conversation, from the third reply on this thread, by hardincap: What a nice statement -- totally summarizes and refutes all the opposing arguments, so that really, you have no choice but to understand. But, Hmm. I'm not sure if it's deliberately not reading the other persons said, or just simply misunderstanding. It's hard to imagine that it's the latter, since hardincap explained it so clearly and concisely. My working hypothesis is that the right wing media, with its one-sided view and demagoguery, has made people believe that they don't need to make sensible arguments when discussing an issue, but that shouting down is sufficient. Hence, the Christie's comments and the response of some people on this thread. The really unfortunate thing is that there are actually good arguments against Buffett's position. It's just that nobody's making them. I think Munger summarizes the situation very well:

Thanks Myth. I'd agree that I'll never persuade Hawk, just as it's pretty unlikely he'd persuade me. I kind of view it as having a similar purpose as a debate. Gingrich's goal isn't to persuade Romney of his ideas. Very good summary by Howard Mark -- it's pretty rare you see such a nice summary given of both sides with such minimal bias. I think the place you want to end up in is not necessarily the Buffett rule, but the ovarian lottery that Buffett talked about years ago (though I'm pretty sure that was someone else's idea). Basically, you want to create a society such that, if you were born randomly into someone in that society, you'd maximize your chance to have a good outcome.

Yes, that's what I mean. 100% tax would obviously be bad. 0% tax would also obviously be bad. There already is a little redistribution of wealth as of 1862. The government determines what is sensible. Most western countries, with the exception of Britain and a few others, have better opportunities to succeed. Basically, this is what economic mobility is all about. (That said, if you want to argue about which countries are the best at creating really rich people and really poor people with huge gulfs between them, then US is pretty good. Though Russia and China seem to be catching up in that respect, too.)

Nor, at reasonable levels, does it significantly disincentivize the anyone from trying to move from one to the other.

The whole idea of taxes are that they redistribute assets for the benefit of society. For instance, defense and policing are to the benefit of society, so we redistribute assets to ensure that we can pay for these things and society benefits. It's not about ensuring that the person or corporation from whom assets were taken receives benefits proportional to the assets that were removed. I would guess that almost every person and corporation receives benefits disproportionate to what they pay (most more, and some less). Though the right to live in a relatively free country with laws and low corruption is pretty valuable to me (and presumably to most corporations).

I assume you are being facetious since this is the Corner of Berkshire & Fairfax Message Board and not Mother Jones. Yes, I was.