Publishers cannot relent. The law doesn't allow them to require payment for snippets (like the German law did), it requires them to require payment. Which is why Google is shutting Google News down entirely in Spain... since all Spanish publisher are required to get paid, and Google isn't going to pay them, there will be no Spanish content for the Spanish Google News, making it useless.

Particularly for a service which doesn't generate any revenue.

Nearly all of what you say are valid points. But one carries a misconception:

By it's very nature of being a focused, collimated beam a laser does not affect anything in "the general direction" of the target - if it was not focused and accurate, it wouldn't be an effective weapon and might not even be dangerous.

That's SO not true. There are two issues here:
  - Forward (and back) scatter: A laser beam "leaks" light, primarily in the "general direction" of the main beam and, to a lesser extent, in the general direction of back toward the source. It's not a big percentage. But when you start out with kilowatts of colimated light it can be more than adequate to burn out a human eye.
  - Scattering (also specular reflection) from the target, or the cloud of gas that remains of the target. This can be a substantial fraction of the incident beam.

"Do not look at the beam or the target with the remains of your face."

Deuterium/Hydrogen (D/H) isotope ratio is significantly higher (more than three times, in fact) than that of water found on Earth.

Q: How do you separate heavy water from light water?
A: Distillation. Light water boils off / evaporates more easily, because the molecules are lighter, and leaves the heavier water behind.

Why shouldn't this be true of vacuum sublimation as well?

Leave a chunk of dirty ice orbiting the sun in a hard vaccuum for a few million years, with the water quietly sublimating away. Seems to me the result would be that last remaining chunk of dirty ice would have a substantially larger fraction of heavy water molecules than the water on the planet where the deep gravity well hangs on to the lighter molecules.

Is it enough to explain a 3:1 enrichment? No clue. But I'd like to see that the analysis was done and what the scientists' estimates were.

(Not to say they ignored it. The last time I raised a similar question about a scientific paper reported here it turned out that the scientists HAD examined the issue.)

This will cost us billions of dollars in the private and public sector,

who is this "us" he is talking about?

The taxpayers. It's a clear violation of the "takings" clause of the US Fifth Amendment (long since incorporated against the states and their subdivisions, including the City and County of Los Angeles.) This means, after a bunch of legal wrangling, the courts are very likely to rule that applying such a law against a pre-existing building is a "partial taking" and the government must make the owner whole, i.e. reimburse him for his costs of compliance.

The takings clause:

No person shall ... be deprived of ... property, without due process of law; nor shall private property be taken for public use, without just compensation.

if the public good is really being served here by improving safety of citizens, why isn't the discussion framed more along these lines?

When it gets to the courts, it will be. Count on it.

One of the religious prohibitions in Islam is making war with fire.

If this is used it will be interesting to see the effects on recruiting by the Islamic State and other anti-US organizations among those Muslims who are currently either opposed to them or unaligned.

Also: How do you keep a 30 kW laser, at any frequency, from blinding everybody in the general direction of the target? The last I heard, weapons that blind are banned by the current "laws of war" as recognized by the western powers - and that's been the major impeidment so far to deploying laser (and other directed energy) weapons. Has something changed? Or did the current administration just decide to play with the new toy despite past promises to the other kids?

Private and unimplemented is the key, on old compilers.

On new compilers use:

class MyClass {
MyClass(const MyClass&) = delete;
// ...

Doesn't matter what access specifier you use, because any attempt, from any code, to copy a MyClass will be diagnosed as an error by the compiler.

Nonsense. The only way you don't know what that means in C++ is if the code was written by a complete, drooling idiot.

Seriously, this is the first thing C programmers pull out to criticize C++... but in 23 years of professional C++ programming, I have never seen bizarre arithmetic operator overloading used in practice, except for iostreams, and you get used to that pretty quickly, given that it's been part of the standard library since very early on.

I have a few times seen libraries of mathematical operations, say on vectors or tensors, that made heavy use of arithmetic operator overloading, but it was so they could say "i = j * 5" where i and j are n-dimensional matrices. In those cases, operator overloading not only makes sense, it's a dramatic improvement over C.

If you want to criticize C++, there are lots of valid criticisms, mostly around the huge variety of features and the complexity of their interactions, which can get really subtle. And you can criticize many of the insane template metaprogramming constructs (though those can be really useful sometimes, particularly in building up infrastructure that allow the compiler to diagnose all sorts of errors you might make). If you don't like "invisible" stuff, you can criticize the abuses that can be made of constructors and destructors. But operator overloading? Faugh.

Or, in the case of the last two, intentionally declaring them private and NOT implementing them (or if you have a C++11 compiler, explicitly deleting them), so as to make the class non-copyable. Relatively few classes need to be copyable. Declaring the copy ctor and assignment operator private ensures that client code can't accidentally copy your non-copyable objects. Not implementing them ensures that if class code accidentally copies an instance, you get a link error.

Anyone know what the efficiencies are on these sorts of "tabletop" laser particle accelerators versus say a linac? I'm curious as to whether it'd make an effective "tabletop" spallation neutron source

I don't know about efficiency, but the problem with the tabletop synchrotrons (which accelerate electrons, but X-rays are the primary product) is that their X-ray flux is much lower than the football-field-sized rings, which means they're not as useful for molecular imaging applications. My guess would be that the same problem would apply to a tabletop neutron source.

How is this "directly connected to the internet" when it is using a router to access the net.

By that definition, NOTHING connects directly to the internet.

Anyone with a better understanding care to explain ?

The proper definition of a host running an internet-facing application being "directly connecting to the internet" is using IP for the first hop, with the packets having a route from there to and from the rest of the Connected (capital-I) Internet.

Bluetooth 4.2 added support for IPv6 to/from bluetooth devices. This means IP packets formed on, or directed to, the Bluetooth 4.2 hosts, for delivery to/from other Internet-connected devices, do not require a protocol-translation gateway to select and translate some subset of the packet types, services, and features, modifying the transport semantics to support some tiny subset of functionality that the gateway explicitly understands. An IP packet formed on the bluetooth device goes all the way to its destination semantically unmodified, and ditto packets going from some other device to the bluetooth device. The full feature set of IP (or as much of it as the stack implementer choses to support) is available, while the routers can be "as dumb as rocks" and totally ignorant of what the application on the Bluetooth device is up to, in classic Internet style.

A Bluetooth 4.2 device, using IPv6 and with a route, IS on the Internet, and is a peer to all other internet-connected hosts.

This is relevant to nerds and technology how?

Some of us are eco-nerds.

Seriously. Planets and space habitats will need ecological engineering - the real stuff, not the eco-wacko knee-jerks.

Examinations of how this horrendously complex system works when tweaked are definitely "news for nerds" and "stuff that matters".

There are lots of different sorts of nerds, and lots of nerds geek out on many different technologies each. If you sometimes see nerd-fodder that isn't on one of YOUR subjects on Slashdot, suck it up and shut up, while the nerds of THAT topic finally get to have THEIR conversation.

We get enough of that disruptive raining-on-our-parade from the jocks.

Sure, and if they wanted to they could sell gasoline-powered cars, too.

Tesla is taking a shot at modernizing the car industry, and not just with their choice of powerplant. The dealer system made sense when you needed local expertise, but information is much easier to distribute today. Dealers are an anachronism -- and they know it, which is why they're fighting so hard to retain the regulatory restrictions on direct sales.

There are lots of practical reasons why Tesla doesn't want to go the franchise route, but besides all of those, it's just not where Tesla wants to go. They may fail. Saturn tried to buck the old model, too, not by eliminating dealerships but by enforcing tight rules on franchises and requiring a set-price sales model, and they ultimately failed, first falling back to the old model of jerking customers around, and then ultimately getting shut down entirely. Maybe Tesla will fail, too, but I don't think so, and I'm glad they're sticking to their guns.

That's a rather strong statement. Do you mean that they don't compete right now, or that they never can/will compete? Because your statement sounds like the latter, and I don't buy it. Central generation clearly can benefit from economies of scale, but distribution is enormously expensive, both in terms of infrastructure cost and power losses.

That the list contains people without convictions means that you can be added, and your sentence affected, by things you haven't been proven guilty of: Due Process Fail.

That stuck out like a big sore thumb to me. It's police and prosecutorial misconduct, pure and simple. (I'm appalled that this wasn't brought up until this far down in the discussion.)

Other items, just from the little bit quoted here:
  - 'people whom the D.A. considers "uncooperative witnesses,"'

One of the big differences between the US and English systems is that in the US you are NOT REQUIRED to risk your own life to do the police department's work by testifying about what you've seen. (You aren't allowed to lie, but you are allowed to be silent.) The police often can't, or won't, provide you with protection against criminal retaliation for your testimony, at the same time that they block you from obtaining or using the means to protect yourself. Don't want to be a martyr? Just say nothing.

But these guys are turning that principle on its head: If they decide you're an "uncooperative witness", into the database you go, to be harassed and minutely scrutinized from then on, threatened with arrest at any slip-up, treated differently, and far worse, than other citizens. That's selective enforcement at its worst, and denial of civil rights under cover of law.

Then there's "gang members". If some policeman don't happen to like you and the friends you hang out with, they he can define your group as a "gang", regardless of whether you've committed any crime, and treat you and your group as they would big-time repeat offenders. Any bets on whether this gets used against political opponents of the prosecutors' party?