France tried decimal time after the French revolution (each day had 10 hours, 100 minutes/hour, 100 seconds/minute)...

They also tried a wacky calendar: 12-month year, 30 days/month each consisting of 10-day "weeks".

Perhaps the whole thing failed because their scheme only suggested 1/10 days of rest instead of 1/7 (because 2/10 was too much historically). Given the current French employment situation, perhaps 3/10 days off is better than 2/7?

If energy became essentially "free", it might be possible for a small number of well financed crazy people to hoarde/concentrate enough energy to cause an event that created enough damage to make the earth uninhabitable. That would be a bad scenaro.

Your "bad" scenaro is merely disappointing...

As I recall, AMD started their march away from global foundaries and started using TSMC for their original Fusion chips at the 40nm bulk CMOS node and Microsoft has historically fabed at TSMC, so there's no reason to doubt that the XboxOne SOC is a 40nm chip fabbed at TSMC.

Although the 28nm node is currently the most advanced process TSMC has in production, it is currently severely wafer start constrained (Qualcomm, AMD, Nvidia, are all big customers chewing up all spare capacity), so TSMC's "N-1" 40nm node is probably easier to ramp a product in if you need large production capacity by christmas and you don't care too much about die size... AFAIK, TSMC's 40nm node is just the "shrink" of their 45nm node, so their isn't any compelling reason to use 45nm for a new design unless you wanted to just have bigger chips with fewer chips on each wafer and more defects***

***you would typically only run a 45nm process if you had millions of dollars of chip reticles sitting around and you wanted to make a few more chips w/o making new reticles for 40nm..

I don't know about phones, but in the distribution world, it is very common for a reseller to not actually buy a product before it is sold.

Many companies these days work on a virtual inventory basis with their primary supply chain. The basic idea is that the seller of the product effectively leases space in the warehouse to the supplier with a contract such that the supplier agrees to maintain a certain amount of virtual inventory. When the seller sells-through a product, they don't actually have to pay for the inventory until the second the unit is "pulled" from this hub and then the supplier bills the seller and is on the hook to replenish this inventory. Of course the seller discontinues that product, then it just never pulls any more units from hub and the supplier is left holding the bag (even though the inventory is in the seller's warehouse). On the sale, the seller often still has "net-90" days to pay for it as well. As you can see, the life of the supplier isn't easy, nowdays they need to pay for both the inventory and the account receivables side...

For the inventory on the shelf there is a similar paradigm, as part of the shelf stocking agreement, a repurchase agreement is made that the seller can require the supplier to purchase back some or all of the inventory (although usually at a discounted rate), if the inventory hasn't been sold in a certain number of days. This type of stock/repurchase agreements happens in industries far and wide, supermarkets to bookseller to electronic's retailers.

The rationale for the seller offering a high repurchase price and percentages is for the seller and supplier to maximise the amount of product on the shelves (to prevent out-of-stock sales loss) given the seller's risk tolerance for the product. Of course the supplier may be irrational, but the seller is covered a bit in this case... Usually the seller says I'll risk $X to stock your product on the shelf and the agreement is structured by the supplier that although $Y of inventory is stocked, $Y - #units X repurchase_price = $X.

I see why you put the word "liberal" in quotes...

I'm curious how "rubber-room" spending is categorized? (fwiw, new york city statistically has one of the highest in-classroom spending ratios in the country)

An interesting read, if you have the time...
http://www.thegatesnotes.com/Books/Education/Where-Do-School-Funds-Go-Book-Review

Apparently, State Senator Yee doesn't readily give up if laws that he supports eventually are declared unconstitutional...

http://arstechnica.com/gaming/2009/07/sensationalist-legal-brief-aims-to-revive-ca-game-law/

FWIW, he's the same state senator that sponsored the bill to ban the sale of violent video games to minors, and has seemed to make harrassment of the video game biz his personal mission...

He's also sponsored legislation to ban any semi-automatics that have conversion kits... I suppose this is the next logical step for him since basically a 3d printer is the ultimate conversion kit.

As an inmos refuge that was initially brainwashed by CSP, it's pretty clear that CSP isn't the only way to skin the concurrent code beast and not every thing that is taught as CSP is actually formal enough to be benefit from any of the CSP verification techniques (e.g., shared data structures are quite common in practical systems), so it has really devolved into more of a "meme" than a programming technique that is taught.

As a trivial example, simple server/client is at its surface a CSP-like programming paradigm, until you realize that underneath the database server, there's probably some ACID implementation stuff that can't really be taught the same way (since they usually rely on some sort of global locks which are difficult and not realistic components in a pure CSP analysis framework).

I liken CSP to OOP. Many claim to lament that the younglings aren't taught the proper way to apply OOP, but those people don't often look in the mirror, as they have devolved OOP into a meme consisting of a few coding guidelines rather than a way to be taught to think about code and potential latent bugs. Not to say that everyone should live and breathe CSP/OOP, but perhaps the most valuable part is just the memes and the coding guidelines (which can be learned quite quickly).

Actually, I doubt it. The longer you live, the more likely you would be to experience an expensive condition which they would have to pay for. Insurance companies would ideally want you to pay whilst you are healthy and then die immediatly after contracting any expensive condition (before they have to pay too much). Chronic conditions where you continue to toil on and pay premiums lower than the cost to the company are probably the worst for insurance companies (although conversely it is the "best" case for the the drug companies).

For those that want a car analogy, this is kinda like how some people want to sell their cars before there are any (inevitable) serious repair costs associated with owning a car that is wearing out. Also, if there a serious incident, at some point, an insurance company would rather "total-out" the car than to actually pay for repair costs. In fact this idea used to be a common feature of health insurance policies (although I think that life-time maximums are no longer legal terms in health insurance policies in the US).

Although this study speculates that GDF-11 is the primary active agent, that information is unknown...

However, as we have discovered, blood is much more complicated that we think it is. Also, actual testing of blood substitutes on humans has been very controversial (since many have been unsuspecting/uninformed participants). For example, I doubt very many people remember PolyHeme... Here's and interesting snipet from the wiki...

I imagine that if we did statistics today on this, not only does much more malware exist in stuffs executed by the browser than from applications that they install, they are also much more likely to get malware from the browser vector than any other vector (because of compromised sites and random surfing).

Sure much of the malware payload isn't pure javascript, but many vulnerablities start there (e.g., xss, csrf, etc) and in the implementations of the "sandboxes" that browsers use to run the malware^H applets/plugins (like java).

I think you trust is misplaced in the current environment.

News at 11...

What this appears to be talking about is how at the airport they now swab your carry-on luggage and put it in a machine. I don't think these boston folks would have been though any "swabing" checkpoint so the existance of a device that did this probably would not have made anyone "safer" in this case (or any similar non-airport/govt-building checkpoint situation).

For those curious, the idea behind an IMS (ion mobility spectrometer), is that you ionize your sample (well sort of, you have water or other liquid vapor with ions dissolved in it, not just pure ions in air or in a vaccum) and waft them into a drift tube and use fact that these ionized vapors have slightly different masses so they have different mobility under an electric field. The "spectra" of the mobility under this electric field helps to identify the original chemicals in the swab.

The specific problem they are trying to solve with black powder is that the ions formed by Sulphur (atomic mass 32) and Oxygen (atomic mass 16) are very difficult to disambiguate for a clean detection signal (since O2n- and S1n- have about the same mass).

The common method of disamgibuating is to add solvents or chemical reagents before ionization. AFAIK, in the case of Oxygen interference, a common way to change the ionic signature is to add dichlorolmethane CH2CL2 and the resulting reaction usually exchanges O2- ions for CL- ions (which is enough different than sulphur ion to make it easy to detect), but unfortuantly, dichloromethane also has a side effect of inhibiting the formation of various Sulphur ion allotropes (i.e., different number of sulphur atoms in the ion). So these folks apparently came up with a technique where you ionize first (avoiding the problem with CH2CL2 and sulphur ion formation) and then pass all the ions through a CH2CL2 "titration region" in the drift tube (effectively replacing many of O2- ions that mask the sulphur ion signature with Cl- ions).

Of course the devil is in the details which I haven't read about yet...

And so does this supercomputer...

That raises up the question of how can a supercomputer simulate what our species would do if it had access to a supercomputer that could simulate what our species would do if it had access to a supercomputer that could simulate...
Out of memory [core dump]
% rm -rf *

Which of course begs the question...

PBKAC