In Palo Alto in the heart of Silicon Valley, hedge fund manager Joon Yun is doing a back-of-the-envelope calculation. According to US social security data, he says, the probability of a 25-year-old dying before their 26th birthday is 0.1%. If we could keep that risk constant throughout life instead of it rising due to age-related disease, the average person would – statistically speaking – live 1,000 years.

I get ~ 1,400 years, as the "half life," but I see what Yun did. And that takes into account that 25 year olds have already gotten most of their risky behavior behind them. Your auto insurance rates dropped when you turned 25 for good empirical reasons.

Of course, these calculations don't mean anything because we don't have anywhere near enough of a baseline yet to make that sort of extrapolation plausible.

Keep your work desk productive.

Agree.

De-clutter your work desk regularly,

Disagree this follows. Works different for me. I'm with apocryphal Einstein, "If a cluttered desk is a sign of a cluttered mind, of what, then, is an empty desk a sign?" If my desk is clear, it's a sure sign I am being real unproductive and procrastinating hard on something. When I am on a roll, the papers pile up.

One place I worked had a clear desk policy. Desk had to be totally empty every night. I hated it. Got nothing done. Quit that gig to go work somewhere they cared more what I got done than how my desk looked.

Other people vary!

Sharp knives are safer than dull ones.

I don't think that overfitting is a good metaphor for your problem. Overfitting involves building a model that is more complicated than an optimal model would be. What exactly is the model here, and why do you think that learning just a subset of the course's material leads to building a more complicated model?

Instead, your example looks like a case of sampling bias. Think of the material of whole course as the whole distribution, and of the exam topics as a subset of that distribution. "Training" your brain with samples just from that subset is going to produce a learning outcome that is not likely to work well for the whole distribution.

Memorizing disconnected bits of knowledge without understanding the material - that would be a case of overfitting.

Keep your work desk productive.

De-clutter your work desk regularly, getting rid of things you don't actually use. This includes equipment, paper, plants and even furniture that's doing nothing. Put misplaced items back to their designated space. Designate spaces for supplies and references if you haven't already. Free nearby spaces which are cluttered with things you don't actually use. Put those things out of reach, fill the space with other things.

A good idea is to remove every single item on your desk and think about what you actually need. Repeat this monthly. Put everything back to its place at the end of the day. Repeat this daily. If you find that you need to fetch something daily, put it closer.

Summary of best comments on the original repository

The best advice posted (best comments) in the original repository included (I blatantly pirate-copied it over from their various authors):

• Avoid commuting, or failing that, commute effectively (i.e. by train or bicycle and not by car, so you can do some useful work or exercise).

• Start your posts with a summary if it's more than 3-5 paragraphs. Use paragraphs.

• Treat craigslist as a free storage. You don't need to physically own all the tools if you can pick them up for <(0.1 paychecks). Treat those things as if you'd already own them. Pick 'em up when you actually need them.

• Spend more effort (money, time) on optimizing things you regularly use, such as clothing, matresses, hygiene products, kitchen accessories, and ergonomic computer hardware.

• If you are trying to do X, surround yourself with people who do X. E.g. if you want to read, go visit a (university) library.

• If you are looking for a job, tell everyone you know. Many jobs are gained through personal connections. Post on facebook.

• If a complete stranger or an acquaintance can do something useful for you, ask. (Politely. At a convenient time. With an appropriate amount of honest flattery.) If they say no, don't press them. Always remember to thank them twice: After they agree to help you and again after they've actually helped you.

• Obtain a smartphone. Use it to look data (wikipedia), directions, prices (amazon), and places (yelp), record memos and various data, and read emails. Also, set alarms.

• Learn to cook at least a handful of simple, cheap, fast meals.

• Never post a web link that requires readers to click on it to find out if they want to click on it.

• Observe the 80/20 rule for cleaning. Better clean twice as often less efficiently than waste a day cleaning stuff thoroughly.

• When in need of a conversation topic, ask a question about the other person's life. Anything about their life. People like to talk about themselves.

• Always negotiate the salary when being accepted for a job.

Lots and lots more, from habit formation to winter clothing. Just take a look.

Please describe the mug.

Please please add this to the http://lesswrong.com/lw/gx5/boring_advice_repository/

Relevant in this context:

Boring Advice Repository

Spend Money on Ergonomics, Ergonomics Revisted