Public facts?

If I subscribe to magazine X, is that a public fact? If I purchase book Y, is that a public fact? If my car spends time each week in the parking lot of organization Z, is that a public fact? Library borrowings? Video rentals? Websites clicked on? The list of information about us that might be "shared" goes on and on.

What companies are selling is not just one's info in the phone book, but the association of that info with one's spending habits and one's behavior. I know at present time we have no right to control this associated info, but, yes, I believe we ought to have a right to control it.

Sounds like we need to talk about what privacy really is. A good definition of privacy is "control of your personal information" (probably from This paper.) Of course, keeping personal information entirely secret is the best means of control, but in the modern world, complete secrecy is getting more and more impractical. So what else could we do?

One option I've heard is a property right, such as ownership (similar to copyright) of personal information. Joe "owns" his name &* address, and he'll loan a copy to Time Magazine for the purpose of delivering the periodical he has paid for. Any other use of Joe's information by Time Magazine is a violation, unless Joe & Time have come to some other agreement. This is very similar to copyright, so let's just call it personal copyright.

Copyright might be too blunt an instrument though, because remedies mostly involve (expensive) civil suits. A number of European governments passed legislation called Fair Information Practices. These laws basically say that personal information can only be used for the purpose for which it was given, and cannot be repurposed without consent of the person involved. Probably the governments involved have given themselves a loophole for national security, but I haven't investigated the details. This option reduces the cost to the individual, and makes it the job of the government to enforce the law. I see this as a benefit, though some may not.

Writing Fair Information Practices into law would probably explode the business models of the currently most successful tech companies in the USA, so maybe there's a way to ease into the laws and allow the tech companies time to adjust their business methods...

The "Flying Spaghetti Monster" AKA Pastafarianism was originally created to illustrate how easily one can invent a non-testable "theory" of creation and existence.

A poster above also posits the "10 minutes ago theory," which is likewise non-testable (what is there to prevent an all powerful being from planting memories in every brain; old photos in every album, and ancient dinosaur bones in the rocks?)

Science is about testable theories; in fact I would argue the word "theory" implies testability, so we'll call the non-testable ones "explanations". I'm not sure where to classify the non-testable explanations, philosophy is a reasonable guess. Perhaps the main point to be made with non-testable explanations is that they are so easy to invent.

In any case, the science classroom is the place for discussing methods for testing testable theories, with perhaps a quick glance at several non-testable explanations to see how non-testability operates.

Intel x87 scalar FP instructions use an 80 bit internal format for higher precision. Intel SSE2 vector FP instructions use 64 bits. You will see last bit variations depending on which instructions the compiler chooses. In fact, I've heard of cases where a JIT compiler vectorized a calculation sometimes (directing code to SSE2 hardware), and left it scalar other times (directing it to 80 bit x87 hardware). Might only make a difference in the last bit, but last bit variations can add up over a few billion calculations.

I can recall a physics simulation I was involved in years ago that got differences of 10% depending on what hardware we ran it on. Turned out the Sun &SGI workstations used 64 bit FP, while the IBM box used some 128 bit or something like that. Took a while to track that one down...

I just came across this animation of world-wide weather-station measured temperatures from 1880 to the present. I know it's only 130 years, but unless one can prove a bias in early hand-made thermometers, it seems to show a hockey stick.

I, too, would be amused by folks who used the 1970s as a baseline for global warming data. So, just for the heck of it, I googled images for global warming hockey stick and it seems most of the graphs start at the year 1000 or before. However, among the top four there is one graph that starts at 1970; amusingly, it was created by a global warming sceptic. I suppose you can cite example an example somewhere of someone who bases their global warming theories only on the last 45 years, but it certainly isn't the mainstream.

It would be very wrong of the White House to give one US corporation carte-blanche to ignore a patent. Although the ITC ban may be too strong a response, there's still the fact that Apple has been ignoring a patent for years. They shouldn't be free to continue indefinitely.

At what point are charges of targeting no longer valid?

When the facts appeared to be that the IRS was mostly investigating groups with conservative sounding names, it was valid to accuse the IRS of targeting a particular part of the political spectrum, and that would be very bad indeed.

But now we know the IRS was also investigating groups with names including words like "occupy" and "progressive." Apparently they were investigating both Jewish and Christian groups. Apparently the IRS was investigating broadly across the political spectrum to make sure groups applying for 501(c)(4) status were really non-political.

At this point, don't we need to start easing off our claims of targeting? At this point, don't we begin to notice that the claims of political use of the IRS might have been premature and off base? At this point, don't we begin to concede the IRS might, just possibly might, have been doing the job they were created to do?

If so, then the IRS wasn't "targeting" open source groups at all. Maybe its time to cool down a little.

Nope.

AltiVec was Motorola's 1999 SIMD instructions & hardware, a response to the SIMD instructions & hardware released by AMD in 1998 (AMD called theirs 3DNow!). Intel also released SIMD instructions & hardware in 1999, called SSE. 3DNow!, AltiVec & SSE were all 128 bit wide pipes that could handle 4 single precision floating point operations simultaneously in parallel. Some of them may have also been able to do two double precision floats also (not AltiVec though), and they all did various integer ops in parallel too.

Xeon Phi is a chip that contains around 60 independent specialized Intel X86 cores, plus caches & ring busses for the cores to communicate with each other. The core count is inexact probably because Intel is figuring out the expected number of dead cores on a chip they can ship and still call it a complete chip. Each of the 60 or so specialized cores has a 512 bit wide pipe that will do 16 parallel single precision floating point operations or 8 parallel double precision floats. To call it a "pipe" means a new instruction & data can be issued every clock cycle, and there are a number of instructions "in flight" streaming down the pipeline, with results issuing out of the bottom of the pipe every clock. The pipe is a "fused multiply add" architecture (useful for vector dot products) so theoretically, every clock cycle, the CPU could issue 16 single precision mults and 16 single precision adds, a total of 32 flops per clock per core. Most high performance computing uses double precision, so cut that 32 in half, and multiply 16 flops per clock times 60 cores times about 1.2GHz to get about 1.2 DP teraflops (theoretical) per Xeon Phi chip. Actual flops will be considerably lower if the problem doesn't fit well in cache.

The bottom half of this article has a nice overview of Xeon Phi specs.

How many "Intel Inside" stickers will they be posting on Tianhe-2? I can see an a argument for a mere 16000 - one per node; 32000 - one per Ivy Bridge chip; and 80000 - one per Intel core carrying chip. But I think Intel's marketing dept should hold out for 3.12 million stickers - one per core!

It's too bad Thinking Machines Incorporated never had a sticker policy, because the "Fat Tree" routing topology is straight out of TMI (the prior TMI topology, hypercube, didn't allow the customer as much choice to balance cores vs interconnect).

True, no CPU chips draw 220 watts. But that doesn't mean it's impossible to cool a 220 watt chip

The Nvidia Titan GPU card, with a 7 billion transistor chip at its heart, draws an additional 236 watts when it goes from idle to full load. It's not hard to imagine 200 watts feeding into the GPU chip. Other GPU cards on that page draw even more power than the Titan. The Radeon HD 6950 CFX card drew 329 watts. It's not hard to imagine the chip at its heart drew over 220 watts.

If you want to cool a 220 watt CPU you might need water cooling, but it's by no means impossible.

As of 2013, the user can attach arbitrary tags to a file and search by tag as well as filename (e.g. "important"). Turns out MS introduced file "comments" with Win XP, although searching by comment was awkward. Vista improved upon it, and of course Win 7 inherited it from Vista.

Check out Children's Technology Workshop, AKA iCamp, http://www.ctworkshop.com/

The teach basic animation, stop motion animation, video game design, basic video editing, and probably some other stuff. They use all freeware tools on Windows laptops. Looking at their website, they seem to be actively exporting their stuff to other countries. Definitely worth a look.

Good points. Another thing that makes the S1200 similar to Atom & different from Core is the S1200 doesn't do out-of-order execution. Core chips have something like a 50 instruction re-order buffer, and that helps Core execute an average of 1.5 instructions per clock per thread (at the cost of greatly increased complexity). Atom on the other hand, so far, does no re-ordering, which makes it much simpler and a bit slower.

Yeah, I've heard of offline GPS. But the excellence of Google Maps led me to some false conclusions about smartphone maps in general.

Almost eight years ago, we did some extensive driving in Italy with a Garmin Quest of that era. Every night I would have to connect it to the laptop and swap in a new set of maps for the next day's drive. And sometimes it led us astray -- roads that didn't go through; dead ends that shouldn't have been, etc (not to mention construction detours). But it was still far better than trying to navigate city centers with paper maps.

Since then I've used a newer fancier Garmin and a couple cars with built-in GPSs. They frequently don't find the optimal route, and occasionally they have a few roads wrong or the GPS "destination" is a 100 yards from the true destination. When I got an Android phone, I was blown away by how much better the route quality was than my prior GPSs. Even if the Google maps had only been equal to my older Garmins, the traffic info is so good you don't want to leave home without it.

But my point is this: all along, we have been learning what kinds of errors to expect from our GPS devices. We've learned that minor errors are common. We've learned to be alert when we encounter detours. We've learned that newer devices work better. But errors of seventy kilometers? for multiple cities? I mean, that's so far outside my eight years experience with GPS devices that I wouldn't have believed it. And to think that Apple replaced Google maps with this crap? The quality of Apple maps is so far below the quality of any other GPS device I have ever used that Apple doesn't belong in the same category as the other devices.

At long last, Apple, have you no sense of decency? .