I see no evidence that the police are immune from epic fails in judging character, or are indeed better at it than anyone else. But they do have a lot of experience with *investigations*, and that counts for something.

If you folks on the right had asked one of us *liberals* back in '08, we'd have told you Obama wasn't one of us. He's essentially what would have been a centrist Republican thirty years ago. These were people, like Bob Dole, that we liberals didn't agree with, but could respect and work with. In fact, "Obamacare" pretty much follows the private sector oriented reform plans of Bob Dole. If Obama were a liberal he'd have gone with single payer, and negotiated tough price concessions with pharmaceutical manufacturers (which is the source of America's runaway heath care spending). You'd have seen banks regulated or broken apart, and criminal investigations in response to the financial crisis of '08, not an attempt to put the system back together again the way it was before the crash.

In fact Obama is very much the kind of president Dole would have been: an economic pragmatist, a diplomatic multilateralist, and an aggressive user of military force where he perceives an imminent threat to national security.

If you want to stop state intrusion into private affairs, you've got to stop being afraid, and convince others around you to stop being afraid. The more fear there is in the political climate, the more impunity the government has in its actions.

Liberals got behind Obama in '08 for the same reason we got behind Obamacare: we backed the best alternative achievable in a climate of fear -- a climate, by the way, that makes the state internal security apparatus feel empowered to do anything it wants in the search for terrorists.

I don't think much of your marketing strategy -- after all, do we have any *evidence* that people want [note 1] to run their desktop windows apps on a tablet? That said, I think it's better than Microsoft's strategy. If you enter a crowded market, you've got to offer something other vendors don't have. Dumping money on advertising in an attempt to generate excitement seems hopeless when people are already divided into two camps; iPad or Android. Plus, you don't want to further confuse customers by giving them too many choices to make in your own line.

For years Jobs showed the industry how to do this: streamlined product introductions that focus on stand out features.

note 1: By "want" I mean "are willing to pay for", not "think is a pretty neat idea". Years in business have taught me the neat ideas are common as muck, but ideas that people pull out their checkbook for are very rare indeed.

Available here.

A quick scan indicates it does not say exactly what news reports are claiming it does. The title gives a hint: "TSA Could Strengthen Monitoring of Allegations of Employee Misconduct".

The media (including /.) has seized on one fact out of the report, that the number of misconduct investigations has increased about 27% (not 26% as reported), and erroneously concluded that the rate of misconduct at the agency has increased by 26% (e.g. the title of this /. piece). This conclusion is not necessarily *wrong*, mind you, but the data in the report simply doesn't give us any basis for drawing it. For one thing, one of the main criticisms of the report is that the TSA is not tracking the *outcome* of investigations. For all we know the increase is the result of a higher rate of investigation, or even the increase in the agency's head count.

The whole point of the report is that the TSA has been so slapdash at tracking investigations of employee misconduct it doesn't know the degree which employees are violating policies or even the law. Consequently nobody really knows whether the rate of misconduct has gone up or down. That's damning enough to be going on with.

School does not exist as a vocational training facility for industry. It should train people to be productive citizens. Over the long term that means fundamental skills. By "fundamental" I don't mean "introductory", I mean skills upon which *other* skills can be built: to analyze, to imagine, to communicate and *to learn*.

In terms of computer skills, students should be used to adjusting to doing things different ways, because changes in the software on the market will force them to do that. They should be able to create a problem-solving strategy and execute it with the tools at hand, rather than let the tools at hand dictate their capabilities.

After all, which Windows should they train to use? Windows XP? Windows 8? By the time they hit the market Windows 10 might be the current MS standard, and people may well be using operating systems targeted to non-desktop form factors as much or more than Windows.

But this is aviation, so the instruments are probably all calibrated in some domain-specific units, like "nautical furlongs per kilojiffy.

We're narrow-minded wimps.

Like anything else, the gustatory qualities of an insect depend on how the insect is prepared. You wouldn't care for a raw shrimp, and you wouldn't care for a raw silkworm either. For that matter you probably wouldn't like raw chicken.

Crunchy ants straight from the mound is a taste many people might never acquire, but it doesn't mean you can't use your culinary skills to transform them into something else. For example there are forest people in India who grind stinging ants into a paste and make it into a spicy chutney. You wouldn't know that you were eating insects if you weren't told. For that matter the crunch of a big ant might be just the thing in a confection where you'd otherwise use puffed rice.

Then there is just getting used to the texture and the fact that you're eating bugs. I know people who are researchers who eat handsful of live crickets as a snack because they've got hundreds of pounds of them in their lab, and they like the crunchiness. A lot of people have exactly the same kind of difficulties you are reporting the first time they try raw shellfish, but once you get used to it there are few things tastier than a raw oyster on the half shell with a squeeze of lemon.

Trust me, a raw oyster doesn't have the texture Americans associate with meat.

She sums it up pretty well... http://judgybitch.com/2013/07/29/policing-twitter-is-dumb/

My son was entering middle school, and we asked him what kind of a girlfriend he thought he'd have. And I swear to god this is what he said: "Of course nobody wants an ugly girlfriend, but it's more important for a girl to be smart than beautiful. It'd be better to go out with an ordinary looking girl that you like than a beautiful idiot."

Of course, that was back when he was 12, before the hormones really kicked in.

Having bean a lead developer in a company that was an Oracle reseller (pretty much a necessity in some markets), your characterization of Oracle is partly wrong; the part that isn't wrong is a gross oversimplification.

I've visited some of the places where Oracle's developers work, and as you might expect I am (or rather *was*) pretty familiar with their product. Trust me, they pour an almost unthinkable amount of money into developing unique and useful technology. As you might suspect they don't do it out of the goodness of their heart; they don't even do it out of pride in the product. They do it in order to encourage large, institutional customers to make their systems dependent on features they can only get from Oracle.

There's good and bad aspects to this lock-in strategy. Some of the things Oracle simply does better than anyone else, such as transaction isolation (in an ACID environment). When you develop and test on Oracle, you can pretty much proceed like the user has exclusive access to the database -- no worrying about things like dirty reads or the like (although the DBA had better make sure he's allocated enough rollback segments). It's nice, but not critical; but it also makes switching to a different RDBMS inconvenient. Oracle has gone farther down this path than you probably ever imagined, right up to creating something they call "virtual private databases" -- super-long duration wrapping transactions that persist across database connections and function something like a fork in a source control system. I've known *very* large data acquisition and management operations (e.g. a commercial vendor of worldwide street data for GIS) that depend on capabilities they can *only* get from Oracle.

There are some things about Oracle I really like, like their transaction log management tools, which make it easy to find a past set of changes to your data and undo them with a wave of your magic wand, as if they never happened. For me that's a killer feature. On the other hand they've also done sleazy, bottom-feeder things to lock clients in, like making the way their JDBC drivers handle BLOBs incompatible with everyone else. They may have fixed that, but I don't think it was accidental this annoying incompatibility persisted so long.

I've also visited Oracle sales offices, and know about how they handle "channel" sales. It's all very numbers driven. Oracle's corporate culture is that they don't care about the customer, once he's good and locked in. Oracle's licensing is very complex, it take days of study to figure out what you're allowed to do with your Oracle installation. If a customer makes a mistake he doesn't get any slack; he's got to pay up fast. On the flip side, if a customer accidentally spends five or ten times what he needs (very easy to do), or if he licenses his installation in a way that won't allow for the growth he needs to plan for (also very easy to do), nobody is going to tell him. He's a sucker, and they've got quarterly targets to meet. It flies in the face of most people's instincts to treat customers this way.

Frankly, I find Oracle's corporate values detestable; but it's possible to work with them. They make sure it's *always* possible to work with them, because they want your money. But *don't* expect your Oracle salesman or reseller to take care of you, to look out for you, to warn you if you are about to make a mistake that's in their favor, or to have pity on you if such a mistake leaves you strapped over a barrel. Oracle's business strategy is *built* upon exploiting locked-in customers. You must approach a relationship with Oracle in a defensive posture -- as indeed you should with any agreement other than free software licenses.

Well, I don't know about engineering simulations, but I've worked with systems that did public health simulations. What laymen *think* a computer model can do is predict the future. And maybe in some cases a model can come close to doing that. But the real value of models is to generate questions and hypotheses for investigation.

The problem with models is that they're only as good as the input data you feed them, and in many cases the data is unknowable or based on assumptions you aren't sure of. And that leads to a practical application of a model. You don't say, "I know that X is true, therefore Y will or will not happen" because you almost certainly don't know everything you'd need to know to make such a positive prediction. Rather, you say, "If you are worried about Y, you'd better check on X."

Tthat Halliburton destroyed the documentation when it knew that documentation was needed for the DWH investigation makes me wonder whether simulation results suggested Transocean (the operators of DWH) ought to be paying attention to certain preventable factors that contributed to the disaster. Even if that didn't let Transocean off the hook, it might change the distribution of damages and fines paid by the responsible parties.

It's not just lawyers. I've tried for years to get my writer friends to use some form of version control, but in vain. One of the great benefits of version control is makes it safe to try radical things with your codebase (e.g. novel or story) because even without branching you can simply recall the state of your work at any point in the past.

I myself used a literate programming tool to generate manuscripts, synopses, excerpts, and even alternate versions, but eventually I gave up because the dominant workflow task in writing, other than sitting down and banging out text, is exchanging manuscripts in MS Word format.

Newsflash: People write major systems in Java that work pretty well. People do mission critical, bet-the-company stuff in Java, and it works. *Your* mileage may vary, but it always does.

This doesn't mean it' the best choice for everything, because *nothing's* the best choice for everything

And it doesn't mean Java doesn't have serious flaws. There's something deeply ingrained in Java that encourages over-engineering. But every language has its pitfalls.

Psst... I live in Massachusetts, where we have had Obamacare since back when it was Romneycare (but after it was Bob Dolecare). The sky has not fallen. Initially there has been some supply pressure as people who were priced out of the market for certain services (adolescent mental health care was a biggy) lined up to get services they could now afford. That's a problem, but not an entirely a bad thing.

People always piss and moan about change, but change was coming in health care, even without Obamacare. You can stick your head in the sand and pretend change wasn't coming, but health care spending as a percent of GDP rose from about 5% of GDP in 1960 to 17.9% of GDP in 2009. That's twice what socialist paradise Sweden pays. Do you think things would remain the same when spending reached 25% of GDP? 30%? Or even remained at 17.9%?