It's not fear of nuclear power that makes it uneconomical. It's cheap fossil fuels. Back in the 70s it was the Saudis opening the oil spigot; today it's fracking natural gas and of course coal.

Which is not to say irrational fear hasn't created nuclear problems -- particularly when it comes to developing long term storage facilities for high level radioactive waste. We also give fossil fuels a break on externalized costs because we're familiar with the and therefore fear them less than we probably ought. But still, it's hard to supplant a mature, entrenched, *cheap* technology.

Well, it depends. You have to look at each situation individually to see what is at stake. If you know that Anne Frank's family is hiding in the office annex, you obviously keep your mouth shut.

In a case like this, it's important to remember that civil disobedience is most effective when it forces the government to mete out a wildly unpopular punishment. What the government has done is bound to be extremely unpopular because it has come perilously close to passing a secret law.

People think they have fourth amendment protections for most of this data, but long established precedent (Smith v. Maryland) is that there is no Constitutional expectation of privacy for metadata on phone calls. When Congress weakened *statutory* protections against collecting call metadata, American citizens *believed* their calling data was still protected by the Constitution. Nobody has bothered to disabuse them of this idea; not Congress (who despite their current posturing passed the law and authorized the program) nor the Obama administration (their posturing on "transparency" and "accountability" notwithstanding). They knew that the majority of Americans had no idea the changes in the law technically allowed the government to run a program like PRISM.

The exposure of this bit of flim-flammery makes Snowden standing up and outing himself incredibly powerful. His doing that means that this issue will *not* die down anytime soon. Look at how long the Bradley Manning case has dragged on, and *this* one, rightly or wrongly, may prove to be far more powerful in the public imagination. I think Snowden might have been morally justified in laying low if he thought he could get away with it, but his outing of himself will keep this issue alive through the next election cycle at least. That could deal a far more serious blow to the PRISM program than quietly leaking it's existence. The cost of that greater impact is that Snowden definitely loses his job, and he faces prosecution and legal punishments.

OK, now you are in a position where the burden of proof is on you.

It's legitimate to look at somebody's evidence and say, "it doesn't convince me." It's sometimes *also* legitimate to say "I've seen enough evidence to convince myself beyond a reasonable doubt, so I won't bother thinking about your evidence; otherwise you'd have to take the time to examine the workings of every proposed perpetual motion machine.

What you can't do is say, "I'll dismiss your evidence because there's a possibility you have a conflict of interest." Everyone *always* has a vested interest in any position they've taken in the past. If you go there, if you call a man a liar because he has stated a professional opinion you disagree with, it's *your* responsibility to show evidence that lying has taken place. If you can't, STFU.

I actually subscribe to the Kindle edition of the Boston Globe -- a newspaper I don't particularly like -- precisely because I think somebody need to pay for things like local investigative reporting.

Which must be why the patrolmen's union wants them as standard equipment. Really, you missed your chance to make this an anti-union rant a well as an anti-cop rant.

I developed a very serious mobile app back way in the mid-90s for public health and disease surveillance. Let me tell you from experience why an app that people rely upon every day for critical work is no way to strike it rich. People *need* a lot of support for that kind of app. Support equals labor, and labor is expensive. Businesses with high expenses don't get rich unless they can command huge prices.

When smartphones came along, my partner used to gnash his teeth at stories of developers scoring windfalls with ringtones or stupid little games, and here we were doing *important* work and only making an OK living. I pointed out that if somebody pays $1.99 for something to amuse himself, he's never going to call tech support. When something represents a total investment of fifty to a hundred thousand dollars in hardware, software and system integration services, he damn well is going to call tech support. But 50K isn't really that much money if you include hardware, third party software licenses, QC'ing the client's existing data and converting it, training the administrators and end uses, and negotiating with IT gatekeepers. That's what you have to face when you do work that everyone agrees is important. Yes, people are willing to spend real money on important problems, but they also subject you to higher standards, intense scrutiny, and exacting ongoing demands, and those things eat into your profits. And the only way to get rich in business is to generate profits -- and salary you pay yourself for your labor IS AN EXPENSE.

That's why the $1.99 app somebody buys on a whim to amuse himself is bound to be more profitable than *important* software that somebody relies on to do something important -- no matter how much you charge for that software. There are exceptions to this rule, of course. Software that is a cheaper, more convenient alternative to something someone already has (e.g. Skype) is practical because what it does may be important, but that software itself is at first dispensable.

Look at the vast amounts of cash going into develop "social media"; it is no accident that most of it goes to support is so trivia. Trivia is profitable. It's easier to try radical new things in the trivial. A lot more people have an early adopter stance towards a service like Facebook than they do to towards things they regard as critical. They take convincing and hand-holding. That's why something like Google Wave couldn't get off the ground, you have to approach something as important as collaboration much more conservatively, usually working around how people already do things (e.g. Sharepoint).

And if chess players were allowed to hit each other with sticks, it'd be a lot more interesting for the average viewer.

I actually think the larger ball idea is interesting, but the real problem is that Americans don't have the attention span they used to. If you've ever tried to explain baseball to someone in a country where it isn't played, you'll know it takes a great deal of information just to follow basic game play. Worse, baseball is a game of evolving situations; it takes a database of dozens, if not hundreds of games watched to really understand what's going on. So tweaks like the larger ball are likely to do little for the casual fan and nothing for the dedicated fan.

The Italian Baseball League has an interesting solution. They reduce the length of the game to seven innings (nine is often too long for non-fanatics) and if there is a tie the winner is decided by a home run derby -- kind of the way soccer tournament advancement is done by a penalty shoot-out in a tie game.

Which *still* doesn't constitute success.

The term "agile development" covers a lot of ground. Much of what people mean by "agile" simply amounts to best practices (e.g. daily commits, unit testing, frequent builds etc.). But "agile" also refers to a kind of iterative and incremental approach to identifying and pursuing business goals in your project (exemplified by Scrum). That turns out to be a sensible and appropriate approach **in many but not all situations**.

Many development projects exist in a business environment where business needs evolve quickly, driven by both endogenous (management decisions) and exogenous (competition and market) factors. Likewise the software project itself, if it is reasonably successful, alters the very business conditions it is designed to address. *Under such conditions* you can't set out to build something in two years with any confidence that it will be what is needed twenty-four months from today. On the other hand, no programmer can be productive if he gets a different set of marching orders every day. You are forced by circumstances to adopt a flexible, iterative approach that allows the programmers to actually complete useful work before its specifications become obsolete, which contains the scope of *change-driven* failure and points your team in the right direction sooner rather than later.

But it is critical to remember that not *all* projects are like that. If you are writing software to control a spacecraft or a nuclear power plant, you don't sit down and bang out a little production code to figure out what it is you need to build. There's a lot more you can and should do to prepare for coding, and the classical engineering principle of discovering requirements as early in the process applies. It applies to the chaotic business situation too, but in that situation many requirements are simply impossible to anticipate.

In any case, the phrase "world's biggest agile project" should give any thinking developer pause. "Huge" and "agile" (in the goal-setting sense) don't go together. It seems to me that the idea of approaching *some things* in a waterfall manner (still using many best practices associated with "agile") and *others* in an interactive, exploratory fashion is the approach they should have been taking from the start.

Except there's nothing to say that a CF process would necessarily yield ridiculous amounts of energy at affordable prices. In fact the first proof of concept, if it ever comes, is likely to be just a barely measurable hair's breadth above break even. And scaling the technology to generate megawatts might well prove to be prohibitively expensive. What if a MW plant required thousands of tons of nickel? There might not be enough nickel in the world to supply a significant fraction of the world's energy supply.

Then there's the flying car problem. There is no doubt that practical flying cars are physically possible. The reason we've never seen one is that it's a fool's investment in the short- to mid-term. Any flying car we can come up with over that time scale is going to be a lot worse than buying a dedicated plane and renting a car at your destination. If there were some immediate niche application for a near-term flying car where it beat a dedicated plane and car combo, we might *all* be driving flying cars in twenty years. But there's no such niche to pay back investors. Even if CF is physically possible, if it doesn't quickly reach a stage where it beats some conventional power source economically (e.g. replacing solar panels in remote applications), it might never become practical.

I agree. Let's have that conversation when that starts happening.

Actually, I don't want to slam drones. I think they can be used responsibly and economically. But *responsible* isn't a given.

As for my playing semantic "games", that's a richly ironic accusation.

The man's life was saved by a policeman using an infrared camera which happened to be mounted on a drone.

It's important to get the gist of the story right here, because the decision to use drones domestically is a matter of trade offs. So it makes a difference whether you draw the spurious lesson "drones save lives", or the correct lesson, "infrared cameras save lives, drones save money in deploying such cameras in comparison to conventional helicopters or fixed-wing aircraft." One might reasonably choose to risk civil liberties because of certain life-or-death situation, but not choose to do so if its a matter of another ten or twenty bucks a year on your state or provincial taxes.

That was my thought when I read the summary, but then I read the article.

The glitch wasn't one that caused the machine to lose; it was one that allowed the player to manipulate the machine into paying out the same jackpot *twice*. Compounding that, it could be made to pay out that jackpot at odds higher than the player actually faced the first time around.

Let's look at this by analogy. Suppose there was a real card game offered by the casino, and this game had a flaw in it; that flaw consists of a no-lose strategy. There'd be nothing immoral about a player noticing that strategy and exploiting it to win. That's the impression of the software glitch the summary gives. In fact the glitch works more like this: suppose the player notices that when the employee hands out winning chips, he leaves a drawer full of chips open where anyone coming along could grab them. The player then plays until he wins, pockets his winnings, then walks around to the other side of the table and stuffs a bunch of chips that don't belong to him into his pocket.

I dunno. Some years ago I had a successful business doing field data collection software on Windows CE, later Windows Mobile devices, and for the most part those devices were sold as semi-useless executive toys.

In an ideal world, form follows function; in the real world vendors create form factors and user try to figure out what the can use those form factors for. Many developers tried to shoehorn desktop style apps onto PDA with limited success, but it turned out that besides looking up phone numbers and appointments, the PDA form factor was ideally suited for the kind of app where your field workers hop out of a truck, note some exotic invasive plant, and record spraying it with Roundup. A laptop, or even a tablet is too bulky; you want something you can carry in your pocket. On the other hand, it was painful to type more than couple of words on a PDA using a stylus (things have got somewhat better with predictive text entry).

When you say "there aren't many places I'd recommend them [tablets] for business," you obviously have a set of applications in mind, and of course if they're typical desktop apps you wouldn't recommend tablets. Tablets are poor choices for content creation. The lack of keyboard means they're not very good for text-centric content creation, and the tradeoffs of performance, I/O capabilities, and storage needed to achieve good hand-holdability and battery life mean that other kinds of content creation aren't going to be their forte, either. What tablets are good for are the very task we saw them used for in Stanley Kubrick's 2001 or in Star Trek TNG: information retrieval, presentation and playback. There's plenty of business applications that fit that bill. Furthermore the middle ground tablets occupy between notebooks and PDA means that while they aren't pocketable like a PDA, they have potential data entry applications where the screen size of a PDA is an important limitation, on one hand, but the bulk of a notebook is inconvenient. For example apps where you retrieve and configure things and then hand around the result (e.g. high end point of sale).

Personally, I like the idea of a tablet with a detachable hardware keyboard. But keep in mind most product developers are unimaginative. They don't redesign their product to take advantage of a form factor, they simply bring their old apps up on the new form factor and expect magic to happen. It doesn't. You have conceive an app around a form factor's potential, and design the app around it's strengths and limitations.

The study was double-blind with a placebo control.

You wasted the time spent on your response; time you will never get back.. Nobody's going to want to associate with you, either, because they'll look silly by reflection.

You might need a couple of Tylenol.