dmvaldman

That may be his point. My point is that he stitches together falsehoods to evoke emotional responses in his readers.

example? that quote is from the book btw, not the article.

Yes it is. Because Sandberg never says she admires her brother because of that fact. She says she admires her brother, and then states that fact about the brother. Greenspun is the one who makes the false connection. Here is the full quote:

I should have understood that this kind of self-doubt was more common from females from growing up with my brother. David is two years younger than I am and one of the people in the world whom I respect and love the most. At home, he splits child care duties with his wife fifty-fifty; at work, he's a pediatric neurosurgeon whose days are filled with heart wrenching life-and-death decisions…

With Greenspun's logic, you could equivalently claim Sandberg admires her brother because he is a neurosurgeon.

ah. so being methodically retarded is all part of his plan. how clever!

that's not the point. saying left-handed people make 20% more than right-handed people is equally (if not much more) dubious.

you can't argue that someone's references are possibly fabricated, and then support your own point with your own "scientific" reference.

really it comes down to: if you ever find yourself arguing that someone's references are likely wrong because they are in the humanities... well...that's a conversation ender.

and i have a phd in math. believe me sometimes humanities people rub me the wrong way. but i would never make a general statement like that, and especially not to "prove a point."

it's the refuge of a scoundrel.

This is so embarrassingly bad...it starts bad, and then gets worse... examples:

He discredits one of Sandberg's scientifically verified claims that women take criticism worse than men, by referencing "as John Ioannidis notes, “Most Published Research Findings are False.”

Yet he begins his review with a US News report:

"This US News article describes research that found that, among college graduates, left-handed men earned 20 percent more than right-handed men."

Talk about a bayesian paradox here. It's easy to prove people wrong (and right) when you believe A and ~A.

I could go on, it gets much worse. Does he really think Harvard uses the same internet bandwidth for their particle physics dept and the dorms? Ugh...

this is awesome. honestly people. shut up.

Three logicians walk into a bar...

Bartender: can I get you all a drink?

1st logician: maybe

2nd logician: maybe

3rd logician: yes

Really nice! Some things are broken on my iPhone, like the spacing when deleting and typing letters when they overflow the input area

Isn't the pdf of a uniform distribution just f(x) = 1. What he gave was an exponential distribution with rate parameter lambda. The rest of the math seems correct. But the dependence on e is not surprising as it is relevant to the initial pdf. Am I missing something?

How does this compare to Leap Motion?

You use this argument a lot, but the ranter suggests that the reason many turn to Backbone is because of a self-selecting bias of it being the most popular.

the orange falling on the knife was pretty cool....but..yeah

I'm tangentially interested in full text search. What libraries are you using?

I'd consider working at famo.us.

http://www.davidvaldman.com.s3-website-us-east-1.amazonaws.c...

Laziness? But if that will get Light Table to work, then I can overcome it.

Crashes on open for my Mac OS 10.7.2 Anyone else?

This is something I've been thinking about myself. Sometimes when writing large Backbone apps you may find yourself binding the same event twice to a model, or creating memory leaks by not deleting objects efficiently. It would be great if this tool could pick up on code mistakes like that.

Personally I find the force-directed graph animations distracting. I would render the graph statically and have better animations for the propagation of information in your code.

deleted. thanks

Thanks! I like your idea of summary/blurb. The point of the widget is to parse video, so clarity is key. However we're trying to make a general purpose tool with a crowd-sourcing element, so we also need to keep the structure down to a minimum.

Thanks for the feedback! Scrolling works best by using a mousewheel or trackpad. I agree it's not ergonomic for the click and drag approach, and adding width would help that.

We ask for a username only to give a unique identifier for a comment, and so that you can see which comments are yours. We don't ask for an email or social network link because frankly, I dislike demos that do :-P

Developer here. Feedback would be much appreciated. I should also shout out to Firebase for the realtime backend, and Backbone.js, for making my javascript a little less of spaghetti code.

any desire to open source your video UI? it's terrific

This explains the equation, but it doesn't explain the "philosophy." Being Bayesian is in direct conflict with the "scientific method." The scientific method says: make a theory, then test it against observation. To a Bayesian, first comes the observation, and they pick the theory that best explains it.

If T = theory, and O = observation

P(T|O) = P(O|T) * P(T) / P(O)

tells you: given my observation what is the probability of my theory, P(T|O). And the equation gives you how this quantity depends on how your theory predicts the observation, P(O|T).

That is correct. I tried to allude to that in the end when I said how a Bayesian mysteriously gives a certain confidence to the space of models. In the case of a uniform prior I believe my example is correct.

Usually when Bayesian probability is explained, it often comes across as "duh, that's what anyone would do" except not rigorously with equations. I tried to give an example contrasting a Frequentist and Bayesian approach.

more like a verbose explanation of Bayes' Theorem...

I'll give my succinct spin: Throw a coin in the air 10 times. Say 6 times the coin lands heads up, and 4 times tails up.

A Bayesian would say: the most likely explanation is that the coin is biased in favor of heads with Prob(heads) = .6 and Prob(tails) = .4. Because given such a model, the chance of this outcome is 25% (10 choose 6 * .6^6 * .4^4). Less likely is the explanation that the coin is fair.

A Frequentist would begin by assuming a model, likely that the coin is fair. She would then argue that the probability of getting 6 heads in 10 tosses is about 21% (10 choose 6 * 1/2^10), and she would have tools (number of STDs from the mean, etc) to test how much of an outlier this observation was, in order to validate her model.

In summary, the Bayesian considers many models at once, and asks how likely are they given the observations. A Frequentist starts with one model, and asks how likely is the observation.

The Frequentist uses the scientific method: develop a hypothesis, then test it. The Bayesian does something else, which I think is better, but ill-defined. She somehow tests all hypotheses at once, and puts different confidences in each.

This is excellent news. Does anyone know if this has to do with the recent Icelandic court ruling in favor of Wikileaks [1]?

Also, anyone who hasn't already checked out assange.rt.com should! JA has been producing high quality interviews with some of the most interesting world leaders while under house arrest.

[1] http://www.guardian.co.uk/media/2012/jul/12/wikileaks-court-...

This makes me start thinking what other problems can be handled more efficiently if everyone agreed to be satisfied with an "expectation" of fairness in the long term.

You gain efficiency, at the loss of acquiring short term risk. Might be a useful exchange in some cases.

One good example: in a traffic jam everyone decides to maintain a minimum of 50ft between them and the car in front of them.

What algorithm? I think these problems are fun.

That's Joe Incandela speaking, he taught me graduate quantum physics at UCSB. I feel awesome now :-P