They can track it by laptop since when you are in an office, the laptop will request an address from some local WiFi system. These likely get tracked to determine what devices have been connected in order to flag any suspicious attempts by unrecognised devices. Since each laptops would be associated with a specific employee and the laptop belongs to the employer this would allow them to track who is in the office on a given day. I don't know if German laws prevent this approach as being similar to use of badges

The GPL license states 3 a) Accompany it with the complete corresponding machine-readable source code, which must be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or, If you can download the binaries, then making the source available at the same time is sufficient to satisfy the above term. There is no requirement that it is in perpetuity or for anything beyond an implied reasonable amount of time to allow for download to have completed

I'm not overly familiar with the laws mentioned, but I suspect you can't be arrested for shouting the word "fire" in a theatre, after all if there was a fire it's rather important that someone can notify everyone, however you will be arrested for doing so when there is no fire and causing a stampede to the exit that could result in injuries. I'd bet the law in question would be "reckless endangerment to the public", similar to "incitement to violence" and various other laws around the impact of what was said or the intended impact rather than the fact that you said those words.

I'm sure it's been said before, your speech might be protected so you can't be arrested for the words, however you are not immune from the consequences of what you've said.

So I'd say they absolute can prosecute someone for intent to harm or cause harm by releasing information or speaking about vulnerabilities, without violating the 1st amendment. And it's sufficiently damaging to one's health and general well being to have to get to the point that you could attempt to try and argue it under the 1st amendment that I can't see anyone trying that defence.

virtuozzo/openvz is more analogous to lxc than docker, and when you consider that most users had to craft their own environments from scratch rather than having some working base examples. I used openvz to provide a build farm to test building of software against multiple iterations of RHEL/SLES/CentOS/Debian, where we needed to ensure the client software would compile on and end customers system no matter if they were on RHEL 3.4, 3.5, 3.6 or 3.7 (typically it was the kernel back then that was different and required different patch sets). Creating these was no fun what so ever.

Add to this that both required running a custom kernel where only a subset pre-packaged were available, and it puts a real dampener on adoption because it requires two hops to be jumped through

Docker arrived at the right time where nearly all of the functionality required for containers was now part of the kernel that many people running linux were already using.

On top of this Docker solved the packaging and distribution problem so that once someone provided a container image with the basics it is easy to adapt to your needs.

So it's timing and solving a real problem that enabled adoption to suddenly soar.

I think the study is trying to side-step the arguments for UBI by using very certain tax approaches for costing that distort the overall cost/impact, e.g.

• Up to X amount of income is tax free
• Next Y amount is taxed at 'a'%
• Next Z amount is taxed at 'a+b'%
• Remaining amount is taxed at 'a+b+c'%

With UBI the goal is to eliminate the cliff, but that doesn't mean that you shouldn't tax initial income earned after UBI at a higher amount until you've recouped the cost of the UBI being paid to the person. e.g.

• UBI income is tax free
• Initial income after UBI is taxed at 'a+b+c+d'% until tax received covers cost of UBI
• Next Y amount is taxed at 'a'% (this level may need to be eliminated)
• Next Z amount is taxed at 'a+b'%
• Remaining amount is taxed at 'a+b+c'%

In order words you don't want the current margin tax rate system, you want something closer to a an inverted bell distribution curve, so that initial income above UBI quickly pays back to provide for the social net while avoiding the current welfare cliff.

All the costs in the submission (I won't even call it a report) referenced by the article look for a blanket tax band across all income which would be substantially higher than what most people are currently paying and then use that as a way of hinting that it would be too expensive. Many will baulk at being taxed at a flat rate of 50% and it being so high will cause most to assume that altering it to more progressive tax bands/rates would cause the subsequent rates to have to be much much higher.

Somehow I don't think we have much to worry about a cooler planet in the next 50-100 years, but if the warming keeps happening then most of north western Europe will turn into an ice box during the winter (it's been getting progressively worse over the last number but still not a patch on what could happen) in the near future due to the ocean currents switching, probably the same in a few other places. So saying it's better to be warming rather than cooling is missing the reality of what impacts a general planet warming has. It's not just the whole planet getting warmer, it's the average, with some places going to get much cooler and both of these will cause widespread displacement of people.

Right now, cooling the planet would be better than warming it further as it would reduce the impact of the current changes, reduce the likelihood of widespread conflict, and provide enough time for us to work out how to raise the planet's temperature in 100-200 years time if we accidentally cooled it too much.

More likely the person behaved as through the car was driven by a person, and in spotting them, despite the car having right of way would slow down and brake.

I've seen this many times with pedestrians where I live, and often heard the comment "well if they hit me I'll just sue for loads of money", which ignores the risk of never walking again or being killed outright.

Automonous car manufacturers are starting to learn the lesson that airline manufactures have been learning for some time, if you automate something so that 95% of the time you don't need a human, your failure rate when you need that person for the 5% of cases is extremely high.

For anyone used to driving manual and then very rarely switching to auto should have experienced the habit of looking for a non-existant clutch and in the reverse scenario likely to have stalled the car when coming to a stop a few times before getting back into the swing of things. People vastly underestimate the difference between reacting to danger when doing something you've done for hours on end over the previous days/weeks versus trying to take manage the situation while also having to do something you do very infrequently.

As long as driverless cars aim to make a transition to almost driverless before driverless, we're going to see something like an uncanny valley of deaths occur before things get better. Really needs to jump from almost zero automation (auto braking is fine as long as the person had to remain steering) to full driver-less where someone can drive if they want through a manual override to have a steering control appear, but the car will not hand over control in case of danger.

There's really only a few places where there is real extra API overhead, because mostly, on Windows the API you are calling will eventually result in a series of system calls being made to various parts of the Windows kernel to get stuff done, and on Linux the WINE API will result in just a different series of system calls being made to various parts of the Linux kernel. In most cases the amount of work to be done is the same, it's just that one takes place on Windows and the other Linux.

The performance problems usually come down to places were there is no underlying Linux equivalent of something that is done in the Windows kernel via a single (or few) system calls, resulting in many more operations needing to be performed on Linux to produce the desired result.

DirectX usage would be a good example of where there would be some API overhead in translating the original DirectX calls into corresponding OpenGL calls, however unless a data transformation has to be done, the overhead is relatively small. That's not saying it can't happen, but usually it's around specific sets of calls that need some data transformation before being passed on that performance is going to be terrible rather than any of them.

For GPU based computing, WINE can use nVidia's cuda libraries for Linux to handle all the require processing, it even works with Nvidia Physx engine. Note however I'm using WINE with the staging patches applied and locally compiled to pick up the necessary cuda libraries to turn on support (even runs the FluidMark benchmark with decent numbers on my old GTX 570). For those using a distro provided WINE you will be dependent on whether your distro has build WINE (might require the staging patches applied as well) in a way that allows it to use the cuda libraries when installed.

I recall a radio program a year or two back commenting on a study that looked at the global GDP investment cost of remaining with fossil fuels versus switching to renewables, and it found that they were roughly the same, just that the money ended up going to different places.

Basically, we do have the money, it's just a question of who gets paid.

Really wish I could find that study now....

This is not 'Muslim', it's fanaticism. It's reared it's head in most religions, and in reality has nothing to do with a religion and all to do with the desire of some people to control and oppress others. What religion they follow is incidental.

the 1st of April, so yay or nay on this being an epic April fools that duped the CNNIC as well?

If the cops knocked before hand and announced themselves then it's highly unlikely the householder would shot them thinking someone was trying to break in.

The whole reason swatting is likely to go wrong is because of no-knock entry.

It is however true of many of the good practitioners

Whoosh!

That video is hilarious, right up to the point where you realise there's a kind of creepy truth to the benefit social networking sites provide to the likes of the NSA.