I've heard this story, but it was after my time there. It's definitely in the classic style of MIT lame nerd humor. There's an often element of ironic self-deprecation in MIT humor.

Up until the 80s at least MIT had an archaic phone system in all the dorms. It was almost certainly maintained in part by student labor, since due to tuition costs most students had work study jobs -- often quite technical.

Well, there actually is a legitimate issue here.

Not every takedown notice in the Chilling Effects database is bogus. By putting the text of legitimate notices in a searchable database, Chilling Effects can be used to find infringing content. For example I didn't see "Interstellar" when it was in the theaters near me. Using Chilling Effects I very easily found a number of sites offering bootleg downloads.

If Google removes an infringing link from search result, having the takedown notice copy stored at Chilling Effects appear in Google search result effectively nullifies the takedown. The offending URL is right there in the takedown text.

So what is being balanced here is Chilling Effects' mission -- serving as a database for researching takedowns -- vs. the legitimate copyright interests of the people issuing the takedowns. It won't stop legitimate or illegitimate users of the Chilling Effects database, but it won't guide casual search engine users to infringing content either.

Of course this won't satisfy intellectual property interest groups, whose only mode of operation appears to be "scorched earth".

Diving to 30+ meters? Probably air. And I'm guessing it's probably not a good idea to work divers to exhaustion.

They still use dot matrix printers in some rental car agencies -- if you're getting nostalgic.

As for the old rotary phones they were quite ingenious. The technology didn't exist to have out-of-band signaling between the terminal (phone) and the central office switch. Instead as he dial unwound it would interrupt the circuit between the phone and the switching station, essentially hanging up very briefly. Each of these brief pulses in the circuit current would rotate a series of servos at the switching office by certain amounts. What that meant was that you could dial a phone buy tapping the receiver cradle at a certain speed. When I say "you could" I mean in the same sense as "you can pick a lock with a piece of bent wire and a thin lever." In other words your mileage may vary.

When I was an MIT student a club I was in had a lock on their phone's dial to prevent people making unauthorized calls (long distance call used to cost lots of money. The lock was next to useless because so many people knew how to dial phones by tapping the number out on the receiver cradle.

They looked at the radiation coming out of the sample to find evidence of Am-241, an impurity that would be formed if the sample were created in a cyclotron but not if it were created in a reactor. This test doesn't require the sample even to be touched.

OK, I read it, and I wasn't impressed.

The reason is that your definition is circular:

That's fine as an operational definition of what a society *treats* as privacy, but it does no good in telling us what those boundaries should be.

I worked in an engineering lab at MIT when Mount Saint Helens erupted in 1980, and we'd developed one of the first digital field seismometers, and we used a similar technique. Seismometers that were left in the field for weeks were designed to start recording on to mag tape when an event started, but the problem was you'd lose the crucial minutes *before* where interesting things might be happening. Memory was fabulously expensive, so we fed the data off the A/D converter into an array of discrete flip-flops that functioned as a shift register. When recording was triggered, the mag tape would start recording the seismic reading from thirty seconds ago.

The thing is, memory is *not* fabulously expensive anymore. You can find 128 GB USB flash drives for under $20 retail, so the memory chips must be tiny fraction of that. It should be feasible to record an officer's entire shift -- even a double shift -- from an affordable device. I think it's much more practical just to load up on memory than to try to wire up an patrolman with cables and switches. And as with a volcano exploding, the seconds, even minutes leading up to an event are crucial to understanding it.

I took a course on computer privacy law a few years ago, and one of the big questions is "what is privacy"? After looking at all the various philosophical and legal definitions, I came away with this definition: privacy is autotomy -- the right to conduct your affairs without unreasonable and uninvited interference.

So I would define online harassment as deliberate and uninvited interference. Unpleasantness is simply one *means* by which the interference is accomplished, but it is not in and of itself harassment.

Example 1: you make the mistake of delving into Youtube comments. That's like crossing a Norwegian footbridge with a sack of goats. You have chosen to dive into a pool of nastiness, and unpleasant feelings are an unfortunate but non-actionable consequence of that decision.

Example 2: you decide to block some of the more obnoxious trolls. One of them figures this out and creates a new account so he can continue harassing you. Now that's harassment, because you have explicitly un-invited that interaction. He is interfering with your right to ignore him.

Example 3: one of the trolls doxes you and follows you to another website. That's harassment too because his *intent* is to interfere with your enjoyment of that website.

Example 4: You are on a website and someone violates the site's "harassment" policy. This is a matter for the site admins, not the police or courts, unless the person is cyber-stalking you. A reasonable person doesn't expect site policies to be strictly and swiftly enforced -- it almost never happens. By choosing to use any website you choose to expose yourself to obnoxious people.

citation needed.

Back when people rode horses relatively few people did it. In the 1870s US, the cost to keep a horse in stable would cost almost a third of a typical laborer's daily wage ($0.50 out of $1.75), most of which was already spoken for.

You mean like this one (text available here)? Or this one? Or this one? Or this one?

The problem is that people who demand Muslims condemn violence actually don't care what Muslims have to say. It's just posturing.

Except there's no way of telling whether those addresses weren't being used proxies too.

This is an exercise in Bayesian logic. If you had a high degree of prior suspicion that NK was behind this, it'll look like a smoking gun. If you have a low degree of prior suspicion, it won't look nearly so significant. Personally, I'm in the middle. I think this makes it more likely that NK was behind the attack, but I don't regard it as a "smoking gun". It seems perfectly credible that someone who can orchestrate the Sony hack could hack an NK host. We know that the attackers *sometimes* used proxies. So which is more likely, that the NK addresses are just another red herring, or that they "got sloppy"?

The reason for my agnosticism is the sheer diversity and chaos of the Internet. Arguments that "it makes sense" for so-and-so to have done something hold no water with me, because there are people out there who will do things for reasons that make no sense to me, or won't do things when I think they should. It makes perfect sense for NK (as we understand them) to be behind this, but that doesn't signify.

Motivations are weak evidence for anything. It's like me and my brother-in-law, who is a big-shot cultural studies professor at a prestigious university. I once mentioned to him I always wanted to have a Unimat -- a miniature desktop machine shop. This totally mystified him. He couldn't imagine why someone would want to have such a thing. On the other hand, if I'd said I'd wanted to meet third wave feminist philosopher Judith Butler he'd have found this perfectly understandable and logical. Many people who understand the attraction of mini-machine tools might not understand the appeal of meeting with a major post-structuralist thinker, and vice versa. Unless you see the attraction of both, your understanding of one or the other group's motivations is bound to be unreliable.

Our reading of other people's motivations is apt to say more about ourselves than about them. Hard evidence is what is needed before motivations can contribute to our beliefs one way or the other. Tracing the attack (in part) is a step in the right direction, but far from conclusive.

That's not a crisis, it's the way things should be.

I've been in the business 30 years now, and I've never seen anything like Google's attempt to support older versions of the platform with backports new APIs. In Android you can target an API level like 20 (Lollipop) but support installation on an older API like 16 (early Jelly Bean). On API 16 all the stuff that goes around your app (notifications) and under your app (the VM) will still be "archaic", but the app itself will look modern because of the backports in the support library. Of course you should still *test* on older devices because not everything may work, but getting something to work across the vast majority of devices (> 86% run API 16 or later) is practical for a competent programmer.

So a user who doesn't immediately jump on the upgrade bandwagon can still run the latest version of the vast majority of apps. You miss out on some of Lollipop's bells and whistles, and most importantly on the new VM which reported extends battery life by 1/3 on some devices. You also miss out on all the bleeding edge bugs, of which reportedly there are still quite a few. If your device is perfectly good, it remains practically useful even if it's four years or so behind the bleeding edge.

That's something new. Forced march upgrades have been the norm over my career, but there's no reason to take that stance when you're talking about mobile devices. Most people replace them every two years or so, and in any case the Li-ion batteries soldered into the things begin to lose capacity after three years regardless of use. And with hundreds of independent Android device makers version fragmentation is inevitable, so it's sensible to make it benign.

It makes no sense to wring your hands because everyone isn't jumping on the bleeding edge release when there are so few practical consequences for anyone. It's almost as if the lack of a crisis is perceived as a crisis. People eventually *will* move to Lollipop because they'll eventually replace their old devices. Manufacturers *will* put Lollipop on their devices to get the battery life and performances, but in their own time. In the meantime their customers are OK on KitKat, and so should they be.

Well then, http://en.wikipedia.org/wiki/2011_Norway_attacks.

5% of the general population are sociopaths; 1% are psychopaths. So in any sufficiently large group you you will find plenty of individuals acting in deplorable ways -- even horrifically deplorable. Christians, Muslims, rural Southerners, lesbian golfers, people who like avocados -- any group is bound to have it's share of monsters.

Oh for Pete's sake.

Writing a book is a lot of work, and when you're done you promote it for the simple reason that you want people to read it.

Very few people get rich from writing a book, and it's fairly safe to say that nobody sits down to write a book on space policy because he thinks he'll "strike it rich".