eisenkr

The app has created custom keyboards for typing mathematical formulas. Makes for a much better experience.

I just posted an essay where Freeman Dyson explains his own views about the topic (http://news.ycombinator.com/item?id=532426 or, more directly, http://www.edge.org/3rd_culture/dysonf07/dysonf07_index.html )

The best first book on R I've read is Gelman and Hill's "Data Analysis Using Regression and Multilevel/Hierarchical Models" (http://www.stat.columbia.edu/~gelman/arm/). It's more focused on regression and Bayesian methods than basic probability, but it's a great book that's priced very competitively (only $39.99).

A few years ago I stumbled on a wonderful art project where people used a similar technique to record world class poets as they wrote for 15 minutes. The site is still live at http://quickmuse.com/ if you want to see a poet laureate or two write poems. Robert Pinsky's poems are particularly fascinating to watch because he never makes a single edit or mistake.

Although the article claims that delaying employment with grad school doesn't make you money, there's some good empirical evidence that a delay can create beneficial cohort effects which pay huge dividends over a career.

If you get a job during a nasty economic time, you'll probably end up saddled with a lower salary. This salary then becomes the foundation on which all future raises are calculated (e.g., a 5% raise).

If, however, you got a job during a robust economic time, you'll start with a higher salary and a higher foundation on which all future raises are calculated.

The difference between waiting a year for an extra bump in salary can pay off a lot because the difference will be compounded over the years. This creates an economically rational incentive to avoid the workforce by entering grad school for a little while (no PhD programs) or by finally giving the startup thing a try.

There's a negative stigma about "managers" in the startup world because the word manager is associated with pointy-haired bosses and TPS reports. A manager can, however, be a very positive force when the technical leads on a team are spending a lot of time on management related tasks like meeting with investors, doing financial projections, and revising the business plan. A manager can do these things to free up the owner-developers' time.

In general, I like to think about hiring a manager the same way most startups think about hiring a designer. While all programmers can do some UI design, many startups realize that they're not going to do UI design as well as they write code, so they hire designers. The same principal can apply to hiring management, hiring lawyers to handle patenting, etc: You can do it all on your own, but it might make your life a lot easier if you hire somebody who loves and excels at the specific challenges you're facing.

I don't think we'd ever get to see the data, but this outage created a beautifully structured natural experiment. It's almost impossible for companies to learn how well they would do if a market leader disappeared, but 30 minutes without Google probably told Microsoft, Yahoo, and even a place like Cuil what their best-case future looks like.

Many top business schools today have grade non-disclosure policies that stop employers from asking about grades and stop students from sharing their transcripts. The students love the policies because they claim it increases collaboration and teamwork, but my own experience suggests that the students also love not working hard. The undergrads at a place like Wharton (who have grade disclosure) often look for classes with loads of MBA students (who don't) because the MBA students pull the curve down.

When people are going to school to learn, I don't think removing grades will hurt performance, but learning can suffer when people go to school for credentials and a job. Career-oriented students seem to listen and learn more when there's a meaningful grade waiting for them at the end of the semester.