Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!


Forgot your password?

Comment Re:Add weights? (Score 1) 173

Come now, you're interrupting the Two Minutes Hate.

The proper fix is probably just to reduce the ejection force somewhat, and the seats probably allow it, but from what I know of milspec equipment, the problem is likely that 27 tons of paperwork haven't been completed, so that's not an approved adjustment procedure yet.

Comment Re:Might not need? (Score 1) 146

When a car is tested for emissions, its drive wheels are usually placed on a treadmill. The other wheels are left on the ground standing still. Most cars today have wheel speed sensors for the stability control systems, so brakes can be applied to tires that lose traction. The algorithm to cheat is simple: If the drive wheels are turning at highway speeds, but the car clearly isn't moving at highway speeds, cut the power (and emissions) because the power isn't going anywhere useful.

Comment Re: inadequate (Score 2) 159

We do have a choice. We can either trust others with our information, or we can live without the modern services they provide.

You can live without telephone or Internet service. You can live without credit. You can live without running water, electricity, cable TV, or any other privatized "public" utility. There's your alternative choice.

For most of the last century, America has been opposed to widespread government control. Out of a fear of "socialism", we campaign against raising the government-supplied standard of living. We say we don't want the government to take away our choice, without realizing that the only other option in the choice we have is to return to a standard of living set shortly after the Civil War.

Comment Re:Who gives a shit? (Score 2) 576

We might not be in this mess now, and it might not be with these belligerents, but we would still be in much the same mess, as it dates back to the dawn of recorded history. Any "stability" is simply a nice political way of saying "we're resting between rounds".

The short version is that the nice relatively-fertile land we now call the "Middle East" was a convenient destination for refugees and conquerors from the kingdoms to the north, east, and west. As each group settled, they called the area their home, and ignored the claims of anyone who already happened to live there, who in turn were often exiled or subjugated.

Three thousand years later, the settlers are now distinct cultures, and their tales of settling and exile have turned into legends of a traditional homeland, often mixed with religious justification. As a result, any time an opportunity arises, a traditionalist can easily stir up support for reclaiming the land for his people. Other folks like to call his cause "extremist" to trivialize his intent, and we also like to point to an arbitrary moment in history when his group wasn't in control, downplaying his claim to the area. Do note that I'm not talking about any particular group or ethnicity... their histories are all pretty much the same.

The only time there hasn't been some kind of uprising is when there was a ruler oppressive enough to keep all such rebellions under control. That tactic really started in the area with the Assyrians, until they grew too complacent, and a rebellion erupted with foreign assistance. More recently, the Soviet Union and Saddam Hussein played the role for various areas, with similar ends.

There are only two ways that I see to have lasting peace in the area. One option is to nuke the entire area completely, with fallout, denying it to everyone. The tales of exiled peoples will get worse, but nobody will really want to go home for a few thousand decades. The more reasonable option is to keep working toward "political ends", supporting whatever group says they won't try to kill their neighbors. Perhaps in a few thousand more years, the distinct cultures will merge together enough that they no longer care about conquests predating the alphabet.

Comment Re:it's the weather, stupid (Score 1) 43

There's not much in the way of ballast in modern lighter-than-air craft. Aerostats control their altitude by adjusting their density. They control their density by inflating lift bags inside their outer shell. For weather-induced vertical movements, the craft doesn't need to do much of anything. It'll float or sink to its intended altitude naturally. As for the tether, it's plenty strong enough, and serves as an inertial mass to dampen the movement of the balloon.

Regarding air resistance, consider that the air at 10,000 feet is roughly one third as dense as at sea level. On the ground, aerostats are big unwieldy hulks. At their designed altitude, they can be maneuvered with relative ease.

Comment Re:I don't know about your org.. (Score 1) 131

In the meantime nobody's installing anything else if I get to hear about it.

Therein lies the problem.

If I'm picking a different product to install, it's because I've done my own research, and found that something else will make my system better, even if it's for silly reasons like "easier integration" or "doesn't require a call to a sales rep every time it breaks". Having an architect tell me I can't use it because something else has already been declared the standard just tells me that our architects are wasting our time and money.

I understand the benefits of standardizing, and of course the standard will be one option I consider. I'm just not going to let some other team, who's never seen my requirements, dictate the solutions I employ.

As proof by analogy, consider that I'm currently fighting my way through management trying to explain why a cloud-based monitoring solution is not going to work on systems that are designed to have no network connectivity.

Comment Re:it's the weather, stupid (Score 3, Informative) 43

At the altitudes where such balloons normally reside, the weather is pretty stable. With no pesky land features to complicate things, the weather is mostly just influenced by whatever's already happening upwind. As for the aerostats themselves, they are surprisingly maneuverable at their normal altitudes, usually having just a few horsepower of motors turning a few small propellers. With so little air resistance, the craft can avoid inclement weather easily.

Even if a lighter-than-air vehicle is caught in poor weather, the majority of the effects are mitigated. It's a balloon. It moves with the wind, doesn't make a good conductor, and is usually unmanned. Unlike an airplane, it's not trying to fight against the wind, so the forces on the structure are greatly reduced. In turbulence, the outer bag flexes and accommodates any stress. It might get a little shaky for the instruments, but not unreasonably so.

Source: I used to work with some folks that now design aerostats.

Comment Re: But but muh trooth defector! (Score 1) 262

You're far enough into hyperbole territory that I'm not sure if you intend satire or not.

By rules, you mean laws? You think there's a problem with discovering lawbreakers?

Yes, some rules are explicitly laws. Others are only indirectly so, such as when the law gives decisionmaking authority to a security officer, and he decides to require a particular policy. I also never said I had a problem with it.

Groups "adversarial to the government and its security" prominently include military enemies and their agents, people out to destroy those the government protects.

Those are certainly prominent in the media, yes, but they're not the biggest threat, directly. It's rather unlikely that a military enemy will lay siege to the FBI to get confidential investigation notes. Security is far more likely to be breached by someone who just doesn't follow the rules, or, as I said before, is simply adversarial.

  • Some folks think that they're just so good at their job that the normal folks' rules don't really apply to them. Sure, they think can take home that classified document, because they're so dedicated that they're working overtime on it!
  • Some folks think that the rules are unimportant. Of course the log book should be filled out every time the door is opened, but they're in a hurry!
  • Some folks grow to hate their organization. Yesterday, they filed a questionable expense report. Today, they're beating a polygraph because they can. Tomorrow, they'll leak a secret to the public, just to see the uproar.

Comment Re: But but muh trooth defector! (Score 3, Insightful) 262

Well, yes. The government-security world works that way.

It's a matter of trust. The government will only trust you with its secrets if you play the game and follow the rules, including the silly ones that everybody knows are silly. The polygraph isn't really meant to magically find spies. It's meant to find the people who think they're above the rules or are adversarial to the government and its security.

If it actually catches any liars, that's a bonus, but it's not really the goal.

Comment Re:False understanding (Score 1) 93

And how is that any worse than today, where any bank can change internal records unilaterally, and we rely on an infrequent audit to catch it?

That simple-majority rule is configurable, being just a convenient way to decide which competing blockchain to accept if it diverges for any reason. In this implementation, it's fairly straightforward to resolve, since we already have a centralized auditor: the US government (in various offices for various jurisdictions). If any participants' blocks cause the chain to diverge, both sides of the disparity get an audit.

This violates Bitcoin's rule against central authority, but we don't really care much, because our goals are different. With Bitcoin, the goal is to avoid government. With this blockchain transaction system, the goal is to avoid on a third party for routine operations. Transactions would be validated in blocks, and each bank would ideally unanimously verify each others' findings. The tedious third-party validation need only happen if banks disagree... and that would mean something failed in a big way, so it's probably a government matter.

Comment As Usual (Score 3, Informative) 128

Yet again, Bennett Haselton inspires us with a short-sighted solution, having never considered whether his system will actually work.

1. If a user recognizes a joke as a re-worded version of someone else's tweet, they can flag it as a "duplicate", with a link to the earlier tweet that they think is similar.

Right there in step 1 is the problem. By requiring a link to a sentence someone read months ago, the burden on the user is raised unacceptably. Users won't bother policing when it's difficult, unless the case is severe enough to stir up an outrage - which would already result in more damage than just flagging a user's tweets.

Of course, the potential for abuse is also high. Changing a single word can parody an original post, yet changing a different single word may not avoid plagiarizing. An automated algorithm won't likely be able to tell the difference, so it will fall to manual effort to identify which flagged duplicates are actually malicious. In context, even an identical phrase may be making a very different statement, so taking the tweet out of context for manual review makes false positives very likely.

Shakespeare plagiarized. Plato plagiarized. Tom Lehrer penned many verses praising plagiarism. The bottom line is that plagiarism goes hand-in-hand with creation, and it should always be evaluated only in the entire context of both works - the plagiarizing and the plagiarized. What is being said is often not what's being written.

Comment Re:The Linux community is destroying itself. (Score 2) 232

So how is that different from when the init script goes and fubars itself, or shits all over a log and you can't fix it? Or what about the kernel itself? What if your CPU is really a small explosive device, planted to sabotage your mission-critical systems? I guess you should go fire everyone who ever puts anything into mission-critical systems, because they might fail sometime.

I am not suggesting that critical failure is acceptable, but I do argue that the expertise, testing, and trust of a new system should all scale proportionally to how critical that system is.

If you absolutely cannot have any failure, your critical systems should be running on a nice mature and well-understood flavor of an old Linux or Unix, stripped down to the bare minimum of utilities and subsystems. It should be attended by top-tier admins, with round-the-clock monitoring and redundancy, and every piece (hardware and software) should have been tested for years before deployment. You should probably also replace the manager who put your company in this single-point-of-failure position as well.

At the other extreme, if a system doesn't matter much, it can be staffed by entry-level admins who can barely spell "Unix", running on the newest and shiniest hardware, and those admins should have the freedom to experiment with settings and configurations until the whole thing catches fire. Then they can fix it and do it again. By the time that new technology is ready to go into production, the admins are now experienced enough to understand the machine they're responsible for, and they should know how to configure it to avoid those critical failures.

Ultimately, you are trusting your production system to work consistently within required parameters. That trust is gained by having reliable testing conducted by people who understand the system. You should be hiring people who can build that understanding, not just those who agree with your prejudices.

The disks are getting full; purge a file today.