I prefer duplicati:
I prefer duplicati:
No and no. That's the whole point.
"with big data comes big responsibility"
Simple: any company that leaks private information of US citizens (or allows it to move off-shore) should be subject of $10,000 fine per person (throw in some jail time for executives and board of directors). Plus, the law should explicitly allow class action suits and forbid contracts that take away that right. And proof of damages should not be required.
This leak supposedly disclosed data for 500,000 people, so that would be $5,000,000,000 (Experian goes bankrupt) and Experian executives will rot in prison. Wanna bet there would be class action attorneys at the ready to represent the plaintiffs?
Suddenly privacy protection would become #1 on everybody's mind...
Here is a simple way to make telecoms move on the spectrum they are sitting on: make the lease non-permanent.
If each lease lasted, say, 15 years, and had to be rebid, say, 5 years before the lease expires, the incentive to sit on spectrum would diminish greatly. The prices that companies are willing to pay for spectrum might diminish somewhat, but not utilizing spectrum would start costing real money, and new competition would have a chance to enter the market every now and then.
The problem with the current system is that obtaining a lease to spectrum gives companies a permanent monopoly on the spectrum forever, which decreases the incentive for competition. The spectrum is a sunk cost and delaying utilization of it is merely a loss of revenue, but not a direct cost.
I have wondered why spectrum licenses have been perpetual. It makes a lot more sense to have the lease lapse after 10 or 20 years, and re-auction it. This would provide for the more effective allocation, while allowing big carriers to have return on their infrastructure investment.
Finer grained licensing (what this proposal seems to suggest) is also good, but you can only invest so much in infrastructure without knowing how/when exactly the spectrum will be available. So this will be useful for something like WiFi, but not so well for large installations of cell base stations.
I think your post is inconsistent. If you buy the premise that Iran is trying to build an atomic bomb (you don't question this premise), then you have to accept that Iran may use its atomic bomb. It is possible it will use said atomic bomb for conquest of the land which was conquered by your ancestors and which you are holding right now. This applies to both US and Europe. And yes, Israel.
So the sanctions against Iran are a defensive act against potential conquest. Best defense is offense, right?
You have clearly never lived under Communism. Diapers were made of fabric (because babies do not wait for disposables to be delivered), but everyone was on a constant search for toilet paper. And not the soft stuff that feels like soft fabric that is sold in the West, we're talking about nasty, sand-paper resembling stuff that you knew was made out of recycled newspapers because you could read a letter here and there.
All these posts about the cost to Comcast (Netflix is outside Comcast's network while Comcast's content is within the network) are missing the point.
If Comcast allowed Netflix to host their servers within Comcast's infrastructure for a reasonable fee, then Netflix would have an option of how to host the content. Pay a little more, and Comcast customers do not have to deal with the cap. But Comcast does not allow that. And that is the real issue.
Netflix would love this, because not only would they have a better product for Comcast customers, they would also save on fees for sending their content to Comcast's network. Comcast would of course hate it because they would have to compete with Netflix on a more even playing field.
This seems like something that FCC should take a look at - either apply caps evenly to all content (even within network/infrastructure) or allow competitors to host their content within your network/infrastructure. This sort of a rule should be one of the basic principles of net neutrality.
It is very nice that AMD Opterons are mentioned and petaflops are celebrated, but aren't those petaflops mostly delivered by NVIDIA's Kepler Tesa cards?
From the TFA:
Cray XK6 blades with NVIDIA(R) Tesla(TM) GPUs, based on NVIDIA
(NASDAQ: NVDA) next-generation 'Kepler' architecture, which is
expected to more than double the performance of the Fermi GPU on
Seriously? This is about a cross platform framework that so far has produced a single application that runs on only one platform?
A little premature, don't you think?
emacs or vi? Flamewar at 6!
Meanwhile, they have to continuously remove about 2MW of heat or things get worse.
I wonder if they could use that extra heat to generate some electricity...
The real problem with all patents is that it is difficult to judge which applications are sufficiently non-obvious, and the current system chooses to err on the side of granting more patents rather than fewer. In my opinion, one test for patents should be to check if the patent depends on a technology that has not been widely available for, say, at least 5 years. If it does depend on such a technology, the invention should not be patentable. This way, everybody would have 5 years to develop the same idea. If nobody does, then it is non-obvious. If everybody does, then it is obvious, and shouldn't be patentable.
So, for example, if someone figured out how to efficiently solve the traveling salesman algorithm on a regular computer, that would be patentable: computers have been around for more than 5 years. But if their solution requires a quantum computer, then the idea is not patentable, because quantum computers are not widely available.
By this logic, inventions of internal combustion engine, steam engine, rocket engines, etc, would be patentable. But being the first to create some gene just because you invented the sequencing technology would not allow you to patent the gene. The sequencing technology may be patentable. Most current software patents would not be allowed under this system.
I don't know about Medicare, but there is not enough money to pay Social Security.
Social Security does not hold cash, it holds bonds. And today, due to the reduced tax receipts, it is collecting less in taxes than they are obligated to pay. So, in order to pay all the beneficiaries, some bonds have to be sold. The problem is that these are special bonds, which can only be sold back to the Treasury. But the Treasury is out of cash to pay for them.
The reason they cannot be much cheaper is *because* they get the OS for free.
None of the other vendors can match Apple on purchasing power alone, so they will have a hard time competing on cost for comparable hardware. And since Apple owns the software and the store that comes with it, they can sell the hardware below cost, and make it all back in the App Store, iTunes, etc.
To understand a program you must become both the machine and the program.