Having implemented several projects in the world of analyzing data to make predictions, I think a bigger issue exists: the philosophical (existential?) difference between frequentists and Bayesians. Essentially these two approaches have 'irreconcilable differences'. This is where the 'not enough evidence' objection (objection 3 in the article) usually rears its ugly head. In my observation, the reason silicon valley gets a deserved reputation for using data to make decisions well is that they are (mostly) Bayesian. Startups have to make decisions with very small data sets. If they were to wait until there was a large sample size of users, they would never get over the chicken-egg problem of scale. So they all out of necessity go talk to their small set of users, find out what they like, then make prior assumptions about what a slightly larger group of users might like.
There were many people working on the Manhattan project who then later became nonproliferation advocates. I personally would rather that people feel like they can work for companies that have made mistakes and voice their opinions about where things should go. It would be pretty hard to find out what is happening at companies if there weren't former employees talking about it.
Does anyone know if FB uses deleted photos in their face recognition feature? Or if there is any detail around what happens when data is deleted? I think it probably is like running chkdsk: the data still exists it's just not addressed at the user level any more.
No, the numerator is current assets and the denominator is liabilities. This way if you have half of what you are about to spend in a year, your ratio is .5. If you have double what you are about to spend in a year, your ratio is 2. In your example, your current ratio is .5. Still you are doing way better than most people I know and the stats on national savings and lifestyle requirements bear this out as well.
Your scenario meets the definition of 'using margin as a financing option'. It is the same as the overdraft option provided by credit cards, only it is more dangerous since in the margin call scenario they actually liquidate your stocks without asking you (if I recall from reading up on how this works at most discount brokerages a few years ago). The fact that you rarely do it helps some but as we know the thing about rare events is they happen every day :). I have held myself to a "current ratio" of over two, which is considered 'healthy' by most finance people when evaluating businesses. I would encourage everyone to calculate their current ratio and make sure it is ALWAYS over two as well. It's literally the current assets (cash you could get access to in a year) divided by the current liabilities (all expenses due in a year). Easy to calculate and immensely freeing once you decide it's a rule you'll never break.
This is all interesting to me. Probably around 40 years ago someone in management decided that certain people were winning the economic game because they were better (meritocracy) and then when we went through a series of technological disruptions we came upon a funny variant which I will attempt to paraphrase "those who are successful are the best at increasing their at-bats and therefore their opportunity to have 'lucky' events." This system seems OK, it selects on people who aren't afraid of failure, and that can't be bad. However, the new problem we have is a situation where there are basically monopolies whose heads still have the 'fail fast' mantra stuck in their heads....
We owe him, Sagan, Feynman and a few others a debt of gratitude for writing and speaking about their discoveries to the general public. Without their realization that more people should be aware of this stuff, no popular science!
If anything he is probably wishing there were a codecademy back then, as that probably would have accelerated even his path up the learning curve. I'm his age and back then we had to connect to the wall or go to the library and get a book with coffee stains on it. Also back then surprisingly few people believed studying CS was worth the investment.
Because there is a surprising number of well-funded startups founded by people who can't/won't build anything themselves. While this seems OK on the face of it (maybe they are good at other things like business plans or marketing), in practice what it means is that they end up burning large sums of money unnecessarily on hiring and cloud hosting because they start with a massive technical debt. Also, since they aren't building anything themselves, the odds that they are solving their own problem (a problem you know exists) goes down. The alternative for someone who finds that they can't build anything is to learn how to build something. While this seems like it will take longer than just hiring people and telling them what to do, in fact it is a good investment of time because it eliminates all the headaches I described above.
He came to speak at MIT while I was there, and I remember two things that stuck out. He's a really unassuming guy and smiled at everyone he met in the hallways(have to kind of enjoy when the market rewards humble people), and he also fielded the question "where do I find a technical cofounder" to which he had a great one word answer "codecademy".
The problem is that the ads model works for a while, then they realize it doesn't anymore and they change the free tier. This seemed to have happened recently with gmail. Impossible to know what the eventual monetization model will be when signing up. I think this is the justified cause of the mistrust. Also there are the observations 'free as in beer' and 'if you're not the customer you're the product.' Both of which have merit.
I've been looking for an (optionally OCR enabled) spaced repetition software package that allows me to take pictures of flashcards, and then load those into a standard spaced repetition algorithm, then render those either in a browser or an ios app. Anki supposedly does this but in my experience it's been hard to get up and running. Maybe something that just works with something like evernote or onenote would do the trick.
I second the motion on firecalc because it really helps you visualize in how many 30 year periods historically you would have gone to zero (which is obviously the thing everyone most needs to avoid). The 100 year averages are useless as they don't encapsulate realistic investing lifetime chunks of around 30 years, the firecalc approach gets around this with the 30 year increment parameter (which you can tweak as well). It's a better way to approach the problem.
Becoming 'fat adapted' reduces this cost significantly as you can get approximately 120 fat calories in a tablespoon of high quality coconut or olive oil for about 10 cents in bulk based on my last trip to costco, and as long as you supplement that with an inexpensive protein source you can get by on a very cheap diet. I splurge on high quality veggies simply because we don't know enough about phytonutrients' effect on health to rule them out. This in combination with skipping meals (intermittent fasting) causes the food bill to drop significantly and has some potential health benefits.
Noticed your new project is a keto recipe webapp - this will have to compete with some good ones: https://ketodietapp.com/, and https://www.ruled.me/keto-recipes/ . Fascinated by how ruled.me in particular seems to have won the 'indexed by google' game for 'keto recipes.' Also fascinated by how much ruled.me might be making with adsense. Also generally fascinated by special interest communities and how to monetize them, any content around strategy here would be greatly appreciated!
It's cool because it can train on very small data sets and recognize multiple images in one frame. For example, draw a chair, then draw a lamp next to it. It will recognize both the chair and the lamp not having been fed millions of images of chairs and lamps together.
This comment reminds me of Feinman describing how he out-computed an abacus expert just extrapolating from a few facts he had memorized. I think mental working space is limited and should be optimized on (as in, don't load your mind up with too much low level stuff), sure, but it's the ability to synthesize known tool sets and their capabilities rapidly that helps more. Learning 'another way of thinking about it' is generally the value of doing these first principles style tutorials although I think one can take it too far.
How much do founders typically get when their startups sell for less than what investors put in? Not a troll question, genuinely interested. Any data on how this averages?
I've played against dan players who describe this decision (tenuki, like the tenuki suit in mario, or "to deceptively play elsewhere") as mainly an intuition about "balance" which they somewhat vaguely describe as deciding that although the shape may not be good, the context of the shape is balanced, so it is OK to leave it be. According to Michael Redmond's commentary during the alphago matches (paraphrasing) often times the ability to calculate value is one of the last skills professionals actually master, far behind the concepts of shape and balance, and even among them some rely more on that skill than others. Lee Sedol, as an example, is relatively weak in this regard, Lee Chang-ho, on the other hand, seemed to be a prodigy just at this skill alone, allowing him to play a style which otherwise most observers called 'rigid'. So it is in fact common to find relatively high level amateur players who aren't good at calculating value.
Since alphago plays the game differently (optimizing on odds to win, even if by a little, instead of the human method which involves intuition about shapes), it may be that its explanation if it were programmed to provide one would not provide any human readable 'insight' which an expert could then take to improve his or her game. Imagine a simpler example, some optimization solution using multiple dimensions and linear algebra. A description of how this works would be unreadable to most people. Since most readable language operates in the world of at most four dimensions without introducing jargon, this is the real problem, not that the program cannot explain itself.
I never said the word 'ignorant' nor did I say anyone was a fool. I said 'not understanding'. If someone doesn't understand chess, I don't call them a fool, I just point out they are not trained in chess. Also to your point about grains of rice, I think the world would probably benefit from everyone purchasing the optimal number of grains of rice, and eating that number of grains of rice. Food waste is a problem, as is the belief that 2 dollars here and there doesn't add up.
Is that annual? If it is anything like any other vice, basically it follows the pareto principle, causing massive damage to a small subset of the population which does most of the overuse. "Don't bet the rent money" very much applies here.
Related: I wonder if any analysis on the life outcomes of another jackpot system, "sold company for a windfall", have been done. I appreciate the bad lotto result stories because they confirm my correct bias that lotto should not be played because it is a tax on not understanding probability (except at about 350M for powerball, in which case your enemy is not expected value but variance). However, no such analysis seems to have been done on the startup ecosystem, which generally relies on big exit stories for 'free marketing'.
Basically all social networks with real names are an extension of one's professional brand. This is not good for facebook in its efforts to attract kids because 1) what parent wants to sign their kid up for a permanent professional brand and 2) what kid is thinking about their professional brand anyway?