aneesh

You don't have to have a random sample to estimate a probability. In the Bayesian interpretation, you can have a reasonable prior that you update as you collect more data. For example, a meteorologist can look at the clouds and "guesstimate" that there's a 50% chance of rain tomorrow. A frequentist would object and say that you can't take lots of random samples of tomorrow to estimate that probability. The Bayesian is willing to take the meteorologist's guess as a prior, and update it as & when more data becomes available. In this case, I think Sachin's statement is reasonable in the absence of any real data.

The 5-year gap doesn't make sense. If you're at your peak at age 22, wouldn't you, by definition, begin to decline immediately after? :-)

I suppose you could level off for 5 years, but then it's misleading to call 22 a "peak".

Long answer: See the following discussions

http://news.ycombinator.com/item?id=470128

http://news.ycombinator.com/item?id=389692

http://news.ycombinator.com/item?id=115723

http://news.ycombinator.com/item?id=12010

Also, if you store your customers' credit card information, you have to do implement all sorts of access & privacy controls, a ritual otherwise known as PCI compliance. You probably don't want to go through all that trouble.

But it's a good question to ask. Maybe Tony & Co can tell us whether workers make up for this loss over the course of the week.

It gives you much more flexibility: you can block certain numbers, forward to calls to voicemail, forward to other phones, record calls, make calls & listen to voicemails online, etc. There's plenty of utility even if you have a single phone.

Obviously change the algorithm used to generate the codes for a start. And even though the codes themselves are indistinguishable from real codes, you can probably detect patterns in their use (ie, someone from a town in China who's never had an iTunes account before suddenly buys $100 of music) and prevent a subset those codes from being redeemed (with some very small amount of false positives).

What else would you do?

It's important not to misinterpret that fact and conclude that preventing metastasis will reduce cancer deaths by 90%.

Says who? In my experience, "career experts" want to make sure their advisees get a decent job. If 90% of people who follow your advice fail at their career, I'm betting you don't stay a "career expert" for very long.

People often argue that corporations that require employees to use IE are boosting the browser's market share, but from these numbers, that effect seems small. The weekend market share should be a reasonably decent estimate of browser share without the "corporate effect".

"No, the marginal income (in dollars) is the smallest amount of dollars that you'd take to work for someone else for those 6 minutes."

Theoretically, yes. But try to find someone who will employ you for just 6 minutes. There's not that much liquidity in the job market (yet?).

For me, the marginal income from an extra 6 minutes of time is zero.

Only if you think the "game" is to accumulate karma. For me, the submissions are merely seeds around which insightful discussions grow.

One nitpick: intuitively, I'm not comfortable with the idea that a 20% chance for each nominated movie is a reasonable prior. There is a ranking of the 5 nominations implied by the voting process (http://tr.im/gEYg). However, not being very familiar with Academy Award voting, I'm not sure if this data is ever publicly released. Anyone?