tpeng

See my reply to waterlesscloud for an explanation. There are some big problems in the defaults, particularly around political topics, however, these are not issues of scale. They are issues of diversity. Reddit will need to address the diversity issue in order to succeed.

It's basically a mathematical guarantee. Reddit shows 200 comments by default, so if there are 20k comments, then you are roughly getting the top 1% of comments. If there are fewer than 200 comments, you are getting ALL the comments, basically unfiltered. I've also noticed that the best users make a greater effort on comments that may be read by a large number of users (granted, the average user probably does not)

Voting quality clearly goes up with scale. On a small subreddit, you can make a brilliant comment, and only one person votes on it and you are at score 0. On default subs, the voting tends to be more accurate (though with a populist/groupthink bias)

I think some are confusing comment quality with comment value. A really funny, but silly joke would be a high quality, low value comment. The default subs tend to be lowest common denominator subjects, like "cute" stuff and politics -- low value. Whereas niche subreddits are highly focused on your particular passions and interests. Niche subreddits have greater value to users, which is why they are also more valuable for advertising and monetization purposes.

At sitewide scale, the quality of existing niche subreddits would improve dramatically, and smaller niches that do not exist today as subreddits would gain sufficient scale to support independent subreddits. I think that these niche subreddits could support almost search keyword-like targeting (albeit with lower CPMs because no immediate indication of intent). Imagine native feed advertising volume with keyword-like CPMs.

Not that it's highly relevant, but I'm a developer of a popular reddit app for iOS.

reddit has the potential to unlock a lot of value if niche subreddits can achieve critical mass. There are quite a few subreddits that could support highly targeted and valuable advertising. Comment and voting quality also appears to improve dramatically with scale; compare the quality of comments on default subs to niche subs. If niche subs can get that dramatic improvement in quality, reddit would also be able to attract much better demographics for commercial purposes

Institutional investors are some of the most conservative investors you will ever meet. They don't have much downside from losing money in a fund that looks "normal", and they don't have much upside in outperforming in a fund that looks "weird". They'll look past personality differences for a proven winner but a track record as short as 500 startups' can easily be attributed to luck (even if it is not). It's a principal-agent issue

The creation/redemption process is intended for market makers known as Authorized Participants (APs) to arbitrage differences in the ETF price and the underlying price. While, yes, you can sell bitcoins to the trust in exchange for shares of the ETF, you would then need to sell those ETF shares in the market. So in other words, you would only have liquidity if there were demand for the ETF. It's possible that there will be more demand for the ETF than for bitcoins simply because ETFs are more accessible.

The index fund is the greatest investing technology ever invented. The under-adoption of this technology is a behavioral problem. The solution definitely isn't additional layers of asset allocators taking fees on top of this.

I'm kind of confused by what Sam is talking about in the RFS about enabling lower cost index fund investing. VFINX has a minimum initial investment of $3,000 and minimum additional investment of $100. At 17 bps that's literally $5/yr. on the Vanguard S&P 500 fund. You are basically talking about a nominal cost to service accounts (send statements, support, backoffice, etc). It would be interesting to think about how technology can lower these costs, but cost isn't preventing anyone from participating.

Many problems in AI, such as NLP, are AI-complete. While it's possible to solve subsets of the problem without creating a human level intelligence (and many companies have done so), these solutions do not "seem" very intelligent. Based on Sam's blog post it sounds like he does mean human-level intelligence. Which unfortunately does seem out of reach within our lifetimes though I hope to be proven wrong.

Very few people even bother to track calories. Some people do have trouble sticking with it, but for most, the eureka moment happens sometime in the first week when they realize that they are consuming far more calories than they had believed. It's not a problem of perseverance when most people aren't even bothering to get started.

Calorie logging is actually probably the best solution to the psychological problem. Your stomach is saying to your brain that it is hungry and that you must eat. Sometimes you might be so hungry that you feel you are actually hurting yourself. But if you have been logging your calories and macronutrients, your brain knows that you will be okay. You're not gonna die even if your stomach says that you will. It's a triumph of cognition over biology.

I've met a lot of people who say that calorie counting doesn't work, but I haven't met a single person who consistently and meticulously counted calories over a long period of time, eating at a caloric deficit, who failed to lose weight.

I think special situations and value are both enduring strategies that can exhibit high return / low risk potential.

The cliche among hedge fund guys is to be "long term greedy" not "short term greedy", which explains why some funds try to be more investor friendly with fees. You also have flukes like Berkshire that seem to have undercharged simply out of benevolence. I don't really discuss my investments much, but I mostly do not invest in hedge funds for reasons of tax efficiency--a restriction that doesn't apply to many institutional investors. The one fund I am invested in has significantly outperformed with low risk during my holding period. The funds that I would invest in are small funds that probably no one here has heard of.

Funds frequently return capital or close to new money in order to stay below their investment capacity. Funds also sometimes charge less than the maximum they can get away with for various reasons. The premise of your argument is simply wrong. Moreover, you are ignoring the very real excess profits that are made on the way to this hypothetical equilibrium. Berkshire Hathaway shares aren't going to make you rich if you buy them today, but I'm guessing the people who bought them in 1970 don't care.

I didn't say hedge funds shouldn't be benchmarked to their appropriate index, I said that hedge funds as a whole shouldn't be benchmarked to the S&P500 because they're not all equity funds. In fact, the majority of hedge fund allocated dollars are NOT invested in directional equity (not including strategies like merger arb that technically invest in equities but have totally different risk/reward from the market). It's not "the other side of the coin", it's my point exactly.

Investing in stocks is more labor-intensive than investing in funds. I can identify a few great managers that will do very well over time, write them a check, and be done. By your logic, why should anyone give YC or a16z money? I mean, if I know that YC is better at picking startups, why don't I just go pick my own startups? It's two different skillsets.

It's clearly not the case that I cannot select a set of hedge funds that is less risky than the S&P500. Fixed income funds, for example are much less risky (ignoring for argument's sake some details like inflation risk). I don't know if the "average" hedge fund is more or less risky than the S&P500. Depends how you define average. But no one is investing in the average, you couldn't do so even if you wanted to.

It's mostly irrelevant. No one is investing in the hedge fund average, they are investing in the hedge funds that they think will do well. It's actually the same if you look at VC funds -- the overall returns of venture funds is poor, but the returns of the top quartile of funds is outstanding. In any case, the S&P 500 is not an appropriate benchmark for the hedge fund industry as a whole, because many hedge funds do not invest (solely) in equities and those that do theoretically have different risk characteristics from the market.

The Werther effect could be another explanation for clusters of airplane crashes:

https://fc.deltasd.bc.ca/~dmatthews/FOV2-00074762/S02DB0598....

$1 billion is only about 3% of the deal price. It's not that high considering the regulatory risk.

Interestingly, this only appears to be true in Manhattan. In New York City, single men aged 20-34 outnumber single women. And according to this site, single men aged 20-34 outnumber single women 20-34 in almost every city in the US: http://jonathansoma.com/singles/

I have several hypotheses for Manhattan, but my best guess is that a major contributing factor is that the average age of marriage (and proportion of never-marrieds) is higher in Manhattan than outside. http://nypost.com/2010/01/05/all-the-single-people-in-manhat...

Accomplishment is a key component of happiness, and while specific strength goals may seem arbitrary, so are countless other goals and hobbies that people have. Overcoming a difficult challenge through hard work and determination can be its own reward.

Fastmail should really invest in some better domains. While I can use google to find a company's website, I still have to communicate my email address literally on a regular basis. I'd rather not have to spell out a weird domain.

Businesses in local have two-sided network effects. The most famous two-sided network effect is eBay, where the buyers attract the sellers and the sellers attract the buyers. Here you have local businesses and consumers as the two sides. Ideally, you would have long-term, exclusive relationships with the local businesses and offer some loyalty rewards to the consumers.

There are also some interesting scale economics unique to logistics. It doesn't cost much more to deliver to two houses on the same street vs. one house. So as the density of your userbase goes up, your cost per delivery drops. That being said, scale advantages of local delivery do not approach the scale advantages of a wider network like UPS/FedEx. If delivery were the only aspect of the business, I agree that it would be very competitive

Although I do believe local delivery companies are defensible, they are not necessarily a winner-take-all business in every market like eBay or craigslist. Multiple delivery companies can coexist in a local market, each serving a distinct set of restaurants

WSJ confirms talks -- "early stage" "deal isn't imminent"

http://online.wsj.com/news/articles/SB1000142405270230442270...

“most people don't have the time to go throw 45's around a gym for an hour a few times a week”

“For people that are overweight, what we want “

“cutting out the extra calories won't help once insulin resistance happens.”

It sounds like you are overweight, sedentary, and insulin-resistant.

I mean this in all sincerity and without any mean spirit:

Please find a way to get healthy. If what you are doing is working, great. If not, reconsider your approach. Please do not give nutrition advice to others until you are actually healthy.

I was also being quite generous in that 200 estimate. While many people do anecdotally report that their maintenance level seems to be lower when cutting, I suspect that it's largely due to changes in physical activity as individuals are more easily exhausted on caloric deficits.

I haven't changed my exercise routine, so my total energy expenditure has not changed. Cutting calories is just as difficult for me as it is for anyone else

I am not sure how common insulin resistance is, but given that it is a metabolic disorder, I don't think you can draw general conclusions about metabolism from those individuals.

I recently went from bulking to cutting and went from 3500 calories per day to 2000 per day. You are massively underestimating caloric intake shifts that are possible or necessary.

The stomach is not a good judge of whether you should be eating more or less food. For whatever reason, some individuals may be prone to overeating or undereating. If your natural eating tendencies have caused you to be over- or underweight, common sense says that you should stop listening to those natural urges.

The key to successful dieting is to transform eating from a biologically-mediated process into a cognitively-mediated process based on counting calories. Given this cognitive feedback, your stomach will actually adapt to a new calorie intake level, lessening hunger. Hunger can also be managed by eating more satiating foods and drinking water.

Caloric deficits (or surpluses) work like clockwork. The metabolic slowdown discussed in the article has marginal effects if any -- perhaps a couple hundred calories worth per day at most. A simple thought experiment shows us that the idea of a "set point" body weight is ludicrous: at starvation, of course, the body cannot possibly maintain weight.

The reasons why many people fail at dieting is that people underestimate the amount of effort/pain required:

http://www.cracked.com/article_18544_how-the-karate-kid-ruin...

Nerds are unpopular for a lot of different reasons. This essay seems incomplete.

Probably the biggest reason is that nerds don't make any attempt to fit in. While this may seem like the same as popularity, it's not. Popularity is the goal, fitting in is a prerequisite. Many nerds don't even realize this. Some realize it but value their independence over popularity.

Another reason is that nerds are simply lacking in social skills. Social skills are learned, not innate, and they take a great deal of effort, intelligence, and awareness. The essay to some extent acknowledges this but I think it makes excuses that sound a lot like sour grapes. "Sure, we're bad at social skills -- but only because we don't care so much about them!" I think if we are all honest with ourselves, most of us would love to be good at both social skills and our intellectual/scientific/engineering pursuits.

Another big reason is physical attractiveness. If I could make the rules, being good looking wouldn't matter as much as it does. But I don't make the rules and it matters a great deal to most people. Good looks help with social skills too because good looking people get more practice. Over time that advantage builds up. This is why you don't see a lot of really good looking nerds, even though many good looking people are in fact quite intelligent.

Another reason is movies/TV. Hollywood stereotypes really affect people's perceptions of outgroups a great deal. What movies gets wrong about nerds is this: they correctly display the negatives about nerds (extreme social awkwardness) while completely glossing over the positives (brilliance). This is because most actors and screenwriters are not brilliant (other than perhaps in an artistic sense). So to the average person who learns about people from movies, nerds seem like average intelligence people who are really weird. Again, if I made the rules, people wouldn't be affected by media. Nobody thinks they are, but nobody thinks they are affected by advertising either, yet it's a half trillion dollar industry.

TL;DR: There are a lot of reasons why nerds are unpopular. And "because we don't want popularity" is not one of them.

The most glaring error in this article is that the author is talking about the hedge fund industry as a whole, but he uses the S&P 500 as a benchmark. Hedge funds that don't invest in equities are not, and should not be, benchmarked to the S&P 500.