"But isn't there something a bit creepy about all this?"

No. There's something creepy about a culture which seems to believe you should only befriend people your own age, and that anything more than 15% either way makes you a pervert.

I do think, however, that you need to be careful what you say. I remember playing Half Life online about 8 years ago, and I caught myself using the same profanities I would use amongst friends at our weekly LAN party. What might be a perfectly innocent profanity to me ;) might raise a few questions if other players were particularly young, and I had no way to know, so I reigned it in a bit. Social networks could easily lead to similar misunderstandings.

The comments here are interesting. I think some people get upset as soon as they hear the name of an illegal drug, and instantly get ultra-defensive. You'd think people that claim to be among the most intelligent wouldn't immediately make such an emotional judgment. After all, when is the government ever right about anything? What they consider legal or illegal should not affect your judgment; 50 years ago it was essentially illegal to be black.

Anyway, if you read the article and substitute "coffee" for "marijuanna", I don't think anyone would be upset. If you are against "chemical help", why would you care about the current US legal status of the drug?

FWIW, I don't smoke pot, but I do drink a lot of caffeine. You could say that I'm a drug addict, and that's fine with me.

"How is it possible for someone to get a Ph.D. without picking up even a trace of scientific literacy?"

One word: humanities.

Wow, an entire article about the benefits of pot and working and no mention from this "Ph.D" about the difference between Indica and Sativa marijuana.

For those too lazy to Google it, Sativa (usually grown outdoors) is great for working (think euphoria, and optimism), Indica (usually indoor) is for relaxing at the end of the day and "zoning out". Think 'body high'

Those FUD commercials you see with the kids deflated on the couch? Indica. The commercials you don't see? Well those would be showing motivated Ph.D students working their ass off after smoking a little Sativa (apparently unaware of the chemistry behind the scenes if this article is any judge).

Lesson for kids: Do your homework (this applies to your hobbies too) If you think just smoking weed will help your studies, at least make sure you're buying the right kind or a hybrid which will give you a little of both.

As I see it, it's the American media that has conditioned people to think that any interaction between an adult and a child not related to them is suspect.

This is terribly sad. When I was 14, I would have loved to be able to talk to a real scientist.

Is this exclusively an American problem?

Having done TDD a bunch over the last five or so years, I couldn't ever go back to a world without extensive unit tests.

That said, there are interesting scaling problems with tests that I don't feel like many people seem to write about or know how to deal with; all the TDD books and sites describe techniques that work with simple, small code-bases but which often break down in the face of real-world problems. In particular, writing tests that only fail when the code is broken (and not just because the test is broken) is often incredibly difficult, and if you have (like we do) more than 40,000 unit tests, even a small false-positive rate like .1% will mean that a simple check-in could require hours of test rewriting not because the code is broken but because the tests are poorly written or make too many assumptions about implementation details. Even the best, most disciplined teams create some "bad" tests, and it doesn't take many to start killing you.

Similarly, in a real software environment where a code base lives for years and undergoes numerous changes, you often end up with stale tests that either no longer test anything useful or which enforce requirements that have since been changed.

So while the tests are invaluable and I can't imagine not having them, at the same time there are a lot of issues with test maintenance over the long term that become very difficult to deal with.

It seems like twitter is taking fail to new levels every day.