The "bounce" is just the natural step response of an underdamped second order system. You cannot patent, or at least you should not be able to patent fundamental mathematical principles. The only people who think this is in any way novel or patent-worthy are those who've never taken a higher level math or physics course. But ignorance of the natural laws of math and physics is not an excuse.
That's easy. Just hack into their web site, put a child porn image on there, and MP's reasoning Google should swiftly block their site. What? It wasn't their porn? And the site has significant non-porn uses? But you just told us it's easy to block sites with child porn. Which is it?
No, OP is correct. Any picture of a naked child can be treated by the authorities as child porn. Numerous people ran afoul of this back in the film camera days. They'd take pics of their kids taking a bath and drop the fiilm off to be developed. Some state laws required photo developing labs to report any "child porn" without defining what child porn was, so in an abundance of caution the labs would report anything with naked kids in it, leaving the judgement of what constitutes child porn up to the police. The police and especially child services usually try to err on the side of caution and assume child abuse unless proven otherwise (backwards from the innocent until proven guilty standard used elsewhere).
The number of these incidents has just dropped because photos are now digital and photos of private family moments like bath time are never seen by outsiders. You can still run afoul of the nebulous definition of child porn if you post such photos on the web though.
']' was the Apple DOS equivalent to the '>' cursor in most Unix shells. A person seeing it would assume they were at the command prompt, type in a valid command, and get confused when the computer told them the command was invalid.
Of course we were only out for a little fun so we were content with confusing a classmate for a minute or two (about how long it took the average person to figure out they weren't at the command prompt. But if you imagine someone with more malicious intent, they could've made the program appear exactly like a login screen and used it to harvest people's usernames and passwords.
ctrl-alt-del is a workaround for that. It's a non-maskable interrupt so hooks directly to a specific function in the OS which can't be overridden. It'll bypass any program that's running (like our mischievous bit of code above) and always do what ctrl-alt-del is supposed to do. Normally it reboots the computer (so malware can't fake a reboot by blacking out the screen and displaying what looks like the normal bootup routine). But in Windows Server it's remapped to the login prompt so you can't be duped by malware presenting a fake login prompt to get the admin password and thus root access. In client versions of Windows it brings up the task manager, presumably so you can always kill wayward processes which have taken over the mouse input queue.
Incidentally, the IBM PC wasn't the first with a 3-finger salute. Most older computers had it too. On the Apple IIe it was ctrl + open apple + reset.
I'd modify that idea slightly. If you try to force an idea onto people who don't like it, they're going to figure out a way to make it fail. Instead of setting the idea up to fail out of the starting gate, decouple those who would oppose it from the setting where it would be implemented.
Set up/mandate an alternative market which implements such fees. If it works as you theorize and high frequency traders are driving up prices, the people who don't do high-frequency trades will transition over to the alternative market because it offers them more value (they get more money from sales, pay less money for purchases, even after the per-trade fee). The high frequency traders would then be left by themselves in the old market to (try to) make money off of each other. If they want to participate in the alternative market where the regular people went, then it's their choice to accept the per-trade fee, not something being forced onto them.
If OTOH if you're wrong (I happen to think you're right, I'm just covering all possible outcomes), then the alternative market would fail because people would be paid more/pay less on the original market in spite of the high frequency traders. And we could carry on knowing that high frequency trading isn't the bogeyman you're making it out to be.
You see small businesses make this mistake all the time: "If we only double down, and do what is NOT working HARDER..."
Then, they go under.
I'll disagree and say it's the other way around. The company that goes under is the one that sticks with its tried and true model and refuses to even attempt to change with changing market conditions.
Microsoft is selling a tablet form PC because they have to. Their revenue comes primarily from Windows and Office. Those are for the most part tied to the PC platform. The growth market for PC hardware has shifted from desktops, to laptops, briefly to netbooks, and now to tablets. Currently tablets don't use Windows, and most use alternatives to Office. So they're doing the logical thing (some would say the only thing they can do) to protect their Windows and Office business - trying to push the PC platform into the tablet form factor.
It may or may not work, but at least they're trying to change. Succeeding in business isn't about always making the right business choice. Nobody can predict things that reliably. You have to take more of a shotgun approach - throw a lot of different ideas out there and see which ones stick. If you don't take risks because you're afraid of losing money, you won't survive.
I dunno Microsoft's stance on this, by my stance is that LTE service should only be on your phone. Your laptop, tablet, etc should be able to wirelessly tether to your phone for their Internet connection.
Requiring every portable device to have its own LTE hardware and subscription to a plan is inefficient, wasteful, and only makes sense for the inordinately wealthy who can splurge on such inefficiencies, and for the cellular companies who eagerly looking forward to reselling you something you've already bought from them.
Also beats pattern or password unlocks, which can be 'beaten' by just a bit of careful spying.
A pattern or password will secure your data on the device if an unscrupulous person happens to find it on the bar stool where you left it. Your fingerprints unfortunately are likely to be all over the device, and thus represent less security in this scenario.
This may have been true a few years ago with Android handsets generally being underpowered, but the hardware caught up a while ago already.
I'm pretty sure the lag when task switching is due to insufficient RAM, not being underpowered. I'm still on a 3 year old Galaxy S (single core, 512 MB). If I use the carrier-provided Android 2.3 it's pretty snappy and only lags if I force it to task switch while it's busy processing something else.
But I've loaded Jelly Bean on it to get some of the newer Android features. With a clean install it's still snappy. But it starts to lag after I've loaded just a few apps. In order to keep it responsive, I've actually had to ditch most of the 200+ apps I originally had in 2.3. When I check system memory, about 450 MB is in use after it boots. A good chunk of that is "default" Google stuff that I need, like the Play store, gmail, account manager. I'm saving a good chunk by refusing to install Hangouts (I uninstalled Google+, but for some reason Hangouts keeps wanting to install).
(And before you ask why I don't upgrade, I'm still trying to find a decent water resistant phone which works with polarized glasses. The S4 Active seems to work, but alas it's an AT&T exclusive.)
Hitler's minions thought they were okay because they were just doing their job, also.
That didn't help them much when it came time to hand out the war-crimes awards.
Just something the NSA folks might want to think about. They also might want to take a gander at the Constitution and, in particular, the Bill of Rights. Read them all, including Amendment X.
Bear in mind that there are two different things NSA does/did, with very different implications.
1) They weakened cryptographic standards. This deserves all the criticism you're dishing out.
2) They researched how to break crypto. This is completely within their (and anyone else's) right to do. The alternative viewpoint - that merely trying to break crypto should be illegal - is exactly what the MPAA and RIAA have been trying to foist upon us with the draconian provisions in the DMCA prohibiting breaking DRM.
Up to a month ago such a comment would've been modded to -1 because historically, NSA had helped improve the security of encryption standards. As Schneier has said, the revelations about recent NSA activity has completely evaporated the goodwill NSA earned in the cryptographic community from back then.
I can understand why they do it. You have stories like these, where people are getting electrocuted by iphone chargers. That kind of thing could keep people from wanting to buy iPhones. Easier to just stop people from using 'unauthorized' cables altogether, and stop that kind of story.
I dunno. I haven't seen any stories of people being electrocuted by microUSB cables. If you search google for "electrocuted by USB cable" aside from a few people getting shocked (not electrocuted) by malfunctioning USB ports/chargers, every other hit is for Apple's Lightning/USB cable.
It really sounds like Apple/Intel's Lightning interface has a design flaw which makes it easy to either fail or improperly manufacture in a way which leads to potential electrocution. If you choose to reject the industry standard and roll your own, you damn well better make sure it's as safe or safer. KISS. (I thought it might be due to Lightning's higher max power, but some quick research says it maxes out at 12 Watts vs. USB 2.0's 2.5 Watts. While that's higher, it shouldn't be enough to electrocute someone.)
You can get em as cheap as $0.65. I've run into a few microUSB cables which don't work, but that's the beauty of using standardized cables: You can test them on other hardware to figure out if the cable is faulty or the hardware is faulty.
Ostensibly, Apple refused to comply with the EU requirement for microUSB charging because their port allows other functions like audio and video out. But other phones which had similar multi-function outputs just made a microUSB port for charging/data, and a microHDMI port for audio and video.
There's more going on than that. I have an informal email list set up with a few friends. It's basically a simple forwarder - anything sent to list@example.com gets resent to everyone in the mailing list. Somehow list@example.com ended up with a linkedin account, and we were getting emails sent "from" list@example.com asking us to join linkedin. I never saw a mail inviting list@example.com to join linkedin, the first linkedin-related mail I got was "from" list@example.com asking me to join linkedin. I ended up having to take over the account (using a password reset sent to list@example.com) so I could close it.
So clearly (1) it's possible to create an account in your name at linkedin without you actually ever seeing nor agreeing to the TOS, and (2) they're using email addresses obtained via other means to spam people with their invites. I can assure you list@example.com doesn't have a mail account nor an address book. They have to have figured out that my email and list@example.com were related via a third party's address book.
If you are using a thermonuclear bomb (and the only reason that a bomber with such a bomb would be over enemy territory is the intent to drop it on some target) then it means that you are prepared to destroy a city or some other large area.
This is a common misconception. The nukes developed during the Cold War weren't for destroying cities. Yeah a few dozen were targeted at the larger cities as an afterthought. But the vast majority were targeted at hardened missile silos. The idea being (as silly as it sounds) to destroy the enemy nukes in their silos before they can be launched at you. Since accuracy wasn't that great back then (remember, no GPS), you had to make up for it with warheads with larger yields (sufficient to destroy several hundred feet thick reinforced concrete), and by trying to make sure you had several times more nukes than the enemy had missile silos.
That last bit is what caused the arms race. If the nukes were only for wiping out cities, a few hundred of them would've been enough for both the U.S. and USSR and there would've been no arms race. The numbers got into the thousands because the U.S. tried to make sure it had 3x more nukes than the Soviets, the Soviets tried to make sure it had 3x more nukes than the U.S., repeat ad nauseum. Mobile launchers, nukes aboard B-52s being flow 24/7, and ballistic missile submarines were all invented as ways to avoid this vulnerability of nukes in missile silos.
Different species of bacteria in your gut form competitive colonies. Antibiotics can wipe out the dominant species, allowing a different species to gain dominance and inhibit the previous dominant species from regaining its original population. Several years ago, my doctor diagnosed my constant stomachaches and vomiting as being caused by a certain type of bacteria which had colonized my stomach. He put me on a treatment of strong antibiotics to wipe them out and allow a more benign gut bacteria to take over. My symptoms went away after the treatment.
Similar things have happened on a macro scale. It's suspected the cod fishery off New England has suffered such a fate after severe overfishing led to its collapse in the 1990s. There have been draconian limits on commercial cod catches for two decades, but no rebound in the cod population. It's suspected that capelin have now taken over as the dominant species in that ecosystem. Capelin used to be eaten by the larger cod. But when overfishing decimated the cod stocks, the capelin were able to grow both in size and population. The theory is the tables have turned now and the capelin are eating the juvenile cod, preventing the cod from reaching the size and numbers which would threaten the dominance of the capelin.
Can't speak for the Linux version of ZFS, but I run ZFS for FreeBSD in a VM (specifically, FreeNAS 8.0->8.2). My understanding is that after Solaris, the FreeBSD version of ZFS is the most mature. I watched the memory profile for the VM. It would start at about 512 MB, then gradually fill up all the RAM I gave it, then "purge" itself and drop down to 512 MB and repeat.
Upon researching, it looks like the only feature that's really memory-intensive is deduplication. That keeps checksums for every file on disk in memory, so it can quickly detect if a new file you're writing is a dupe. But when I turned it on, it absolutely crushed performance. Writes which had been going at 60-80 MB/sec dropped to about 5 MB/s. I turned off dedupe.
My dad has abnormally dry hands and a thin (as in not very high) fingerprint ridge layer. He hates typing passwords so uses the same short password on everything. A few years ago I got him a Thinkpad with a fingerprint scanner in hopes of beefing up his security without much additional effort.
The scanner only works about 10% of the time on him. He doesn't use it because of the high failure rate. This tells me that although the tech may read some of its data from the interior structure of the finger, the majority of its functionality depends on the fingerprint ridges themselves. (And yes it's the same technology. Apple bought Authentec in 2012, Authentec bought UPEK in 2010, and UPEK made the fingerprint scanners for the Thinkpads.)
Specifically, cable-tapping they used to do with NR-1 (the smallest nuclear-powered sub ever made). It was retired in 2008. In retrospect, it should've been obvious they retired it because they had a better, probably autonomous, method of accomplishing the same task.
we instead have a massive catastrophic climate change when one of those chambers springs a leak.
A lot of people forget that material properties change with pressure and depth. The first time the Alvin submersible found black smokers (active volcanic vents) on the mid-oceanic ridge, they moved in for a closer look. They found out afterwards that they'd recorded temperatures close to 400 C. The melting point of Alvin's portholes was far less than 400 C, and they would've died if they'd stayed there too long. People see liquid water, and just assume the temperature is below 100 C and therefore the glass portholes are safe. But at the depth they were at, the pressure is much higher and thus the boiling point of water was around 400 C.
Looking at the phase diagram for CO2, that's in the supercritical fluid phase. So the CO2 wouldn't need to be pressurized at that depth like it has to be at sea level. The ground pressure alone would be enough to prevent it from reverting to a gas, and thus it would be impossible for the chamber to catastrophically spring a leak. The only way that could happen is if another drilling operation tapped the chamber and suffered a blowout. Normally that doesn't happen - they keep the bore filled with heavy mud to maintain the pressure at depth. But occasionally (e.g. Deepwater Horizon) there is a blowout, the pressurized mud is lost, and the liquid/gas underneath is then squeezed out by the surrounding rock through the "straw" (bore). I don't see this as being any more risky than regular oil drilling. If anything it's safer since CO2 is pretty inert and won't catch fire. The biggest risk would be the CO2 gas pooling in a depression and suffocating anyone/anything inside.
Not only does the U.S. still have CDMA, most of the rest of the world does too. CDMA won the standards war. The only part of GSM which uses its original TDMA is the voice comms. Most GSM carriers have adopted CDMA or WCDMA for 3G and 3.5G data service (including HSDPA/+).
TDMA sucks because it allocates a timeslice to each phone regardless of whether or not that phone actually transmits during the timeslice. The way CDMA works, every phone can transmit simultaneously and the bandwidth per phone decreases proportionally to the increasing noise floor. i.e. it scales automatically with number of phones transmitting, instead of scaling with the number of phones connected to the tower like TDMA. If it weren't for CDMA, 3G data speeds on GSM would've been limited to about 150 kbps.
That's also the reason GSM phones can do voice and data simultaneously. They have a TDMA radio for voice, and a separate CDMA radio for data. CDMA phones typically have only one CDMA radio, so they can only do either voice or data, not both simultaneously.
CDMA is finally being supplanted by OFDMA (what most implementations of LTE use) because processors have finally become powerful enough to decode the OFDMA signals without draining your battery in 30 minutes. Conceptually, OFDMA is very similar to CDMA, except it operates in the frequency domain instead of code domain. In CDMA each phone is assigned an orthogonal set of codes (e.g. Three phones could be assigned codes AB, BC, and CD. If the phones 1 and 2 transmit simultaneously, the signal the tower sees is ABC, and it knows phones 1 and 2 transmitted while 3 did not. In this simple example, instead of losing 1/3rd of your bandwidth because phone 3 didn't transmit like would happen in TDMA, you only lose 1/4 the bandwidth. The more complex the codes, the less bandwidth you lose). In OFDMA each phone is assigned an orthogonal set of frequencies.
That doesn't work. If the people profiting from an invention wouldn't have had the idea for the invention on their own, then you are not extorting them by making them pay to license the patent.
For stopping (or for that matter, distinguishing) patent trolls, I think we can take a page out of copyright law. With copyrights, you're automatically granted a copyright on anything you make. But if someone violates your copyright, you're limited to compensatory damages. i.e. the offender can only be forced to pay you for actual damages you suffered. To collect statutory damages (the fines that go up to $150k per work regardless of damages suffered), you have to have first registered your copyright. I think if we switch this around a bit, it could solve the problem of patent trolls.
Make it so the original patent filer can collect both compensatory and statutory damages (to protect the little guy who comes up with a great idea, but has trouble bringing it to market while a big company shamelessly steals the idea and takes over the market). But if the patent is transferred, the new patent holder can only collect compensatory damages. That would make it worthless for a person or company to buy a patent solely to sue others for infringement. If they aren't actually building something which uses that patent, then they suffered no damages from the patent infringement and thus aren't able to collect anything from others using that patent. In order to be able to collect damages, you need to be able to show your income was negatively impacted by the infringement, which means you need to be making something which uses the patent.
That would eliminate all the speculating going on with patents. You wouldn't buy a patent in the hopes that you'd be able to collect millions from others for infringement. You'd only buy a patent because you plan to start building something which uses it or it'll improve a product you're already building, and you decide it'll be cheaper to own the patent rather than license it from whoever owns the patent. The main problem I can think of with this idea is you'd end up with a bunch of shell corporations set up to file for patent(s), and people would buy/sell the shell corporation (which is the original patent filer) instead of the patents themselves.
The only reason skeumorphism is maligned is because it's become unfashionable, not because of any inherent flaw in the aesthetic. Mind you, I've a big fan of Microsoft's flat look, but I also think Apple's former approach was distinctive and quite good.
I was a big critic of the bubbly look that was introduced with Windows XP. The new flat look in Office 2013 has finally made me understand one of the advantages of that bubbly look.
Office 2013 (on Win 7, dunno about Win 8) has the flat look but only puts a drop shadow on the foreground window. If I have multiple Excel spreadsheets open in the background, the lack of a drop shadow on background windows makes it very difficult to tell where one ends and another begins. They're all flat and blend together. The bubbly look that Windows XP introduced with its pronounced window borders solves that problem - it makes it (painfully) obvious where the window borders are. XP went overboard with it and Win 7 toned it down. But the way Microsoft has completely abandoned it for the flat look hinders usability in this case.
Back in the 1980s, Apple, IBM, and Microsoft did these huge graphical user interface studies where they tested how people used and reacted to different elements of the interface. They did this to scientifically narrow down what ideas worked and what didn't. With the newer OSes, I get the feeling this methodological approach to user interface design has been abandoned in favor of graphics artists making up whatever they think looks cool.
Slashdot Mirror
The "bounce" is just the natural step response of an underdamped second order system. You cannot patent, or at least you should not be able to patent fundamental mathematical principles. The only people who think this is in any way novel or patent-worthy are those who've never taken a higher level math or physics course. But ignorance of the natural laws of math and physics is not an excuse.
That's easy. Just hack into their web site, put a child porn image on there, and MP's reasoning Google should swiftly block their site. What? It wasn't their porn? And the site has significant non-porn uses? But you just told us it's easy to block sites with child porn. Which is it?
No, OP is correct. Any picture of a naked child can be treated by the authorities as child porn. Numerous people ran afoul of this back in the film camera days. They'd take pics of their kids taking a bath and drop the fiilm off to be developed. Some state laws required photo developing labs to report any "child porn" without defining what child porn was, so in an abundance of caution the labs would report anything with naked kids in it, leaving the judgement of what constitutes child porn up to the police. The police and especially child services usually try to err on the side of caution and assume child abuse unless proven otherwise (backwards from the innocent until proven guilty standard used elsewhere).
The number of these incidents has just dropped because photos are now digital and photos of private family moments like bath time are never seen by outsiders. You can still run afoul of the nebulous definition of child porn if you post such photos on the web though.
Way back in the Apple II days, sometimes we'd goof off in the lab by writing a quick program:
10 PRINT "]'
20 INPUT A$
30 PRINT "ERROR - INVALID COMMAND"
40 GOTO 10
']' was the Apple DOS equivalent to the '>' cursor in most Unix shells. A person seeing it would assume they were at the command prompt, type in a valid command, and get confused when the computer told them the command was invalid.
Of course we were only out for a little fun so we were content with confusing a classmate for a minute or two (about how long it took the average person to figure out they weren't at the command prompt. But if you imagine someone with more malicious intent, they could've made the program appear exactly like a login screen and used it to harvest people's usernames and passwords.
ctrl-alt-del is a workaround for that. It's a non-maskable interrupt so hooks directly to a specific function in the OS which can't be overridden. It'll bypass any program that's running (like our mischievous bit of code above) and always do what ctrl-alt-del is supposed to do. Normally it reboots the computer (so malware can't fake a reboot by blacking out the screen and displaying what looks like the normal bootup routine). But in Windows Server it's remapped to the login prompt so you can't be duped by malware presenting a fake login prompt to get the admin password and thus root access. In client versions of Windows it brings up the task manager, presumably so you can always kill wayward processes which have taken over the mouse input queue.
Incidentally, the IBM PC wasn't the first with a 3-finger salute. Most older computers had it too. On the Apple IIe it was ctrl + open apple + reset.
I'd modify that idea slightly. If you try to force an idea onto people who don't like it, they're going to figure out a way to make it fail. Instead of setting the idea up to fail out of the starting gate, decouple those who would oppose it from the setting where it would be implemented.
Set up/mandate an alternative market which implements such fees. If it works as you theorize and high frequency traders are driving up prices, the people who don't do high-frequency trades will transition over to the alternative market because it offers them more value (they get more money from sales, pay less money for purchases, even after the per-trade fee). The high frequency traders would then be left by themselves in the old market to (try to) make money off of each other. If they want to participate in the alternative market where the regular people went, then it's their choice to accept the per-trade fee, not something being forced onto them.
If OTOH if you're wrong (I happen to think you're right, I'm just covering all possible outcomes), then the alternative market would fail because people would be paid more/pay less on the original market in spite of the high frequency traders. And we could carry on knowing that high frequency trading isn't the bogeyman you're making it out to be.
I'll disagree and say it's the other way around. The company that goes under is the one that sticks with its tried and true model and refuses to even attempt to change with changing market conditions.
Microsoft is selling a tablet form PC because they have to. Their revenue comes primarily from Windows and Office. Those are for the most part tied to the PC platform. The growth market for PC hardware has shifted from desktops, to laptops, briefly to netbooks, and now to tablets. Currently tablets don't use Windows, and most use alternatives to Office. So they're doing the logical thing (some would say the only thing they can do) to protect their Windows and Office business - trying to push the PC platform into the tablet form factor.
It may or may not work, but at least they're trying to change. Succeeding in business isn't about always making the right business choice. Nobody can predict things that reliably. You have to take more of a shotgun approach - throw a lot of different ideas out there and see which ones stick. If you don't take risks because you're afraid of losing money, you won't survive.
They had already outsourced all the IT professionals who could've told them that.
I dunno Microsoft's stance on this, by my stance is that LTE service should only be on your phone. Your laptop, tablet, etc should be able to wirelessly tether to your phone for their Internet connection.
Requiring every portable device to have its own LTE hardware and subscription to a plan is inefficient, wasteful, and only makes sense for the inordinately wealthy who can splurge on such inefficiencies, and for the cellular companies who eagerly looking forward to reselling you something you've already bought from them.
A pattern or password will secure your data on the device if an unscrupulous person happens to find it on the bar stool where you left it. Your fingerprints unfortunately are likely to be all over the device, and thus represent less security in this scenario.
I'm pretty sure the lag when task switching is due to insufficient RAM, not being underpowered. I'm still on a 3 year old Galaxy S (single core, 512 MB). If I use the carrier-provided Android 2.3 it's pretty snappy and only lags if I force it to task switch while it's busy processing something else.
But I've loaded Jelly Bean on it to get some of the newer Android features. With a clean install it's still snappy. But it starts to lag after I've loaded just a few apps. In order to keep it responsive, I've actually had to ditch most of the 200+ apps I originally had in 2.3. When I check system memory, about 450 MB is in use after it boots. A good chunk of that is "default" Google stuff that I need, like the Play store, gmail, account manager. I'm saving a good chunk by refusing to install Hangouts (I uninstalled Google+, but for some reason Hangouts keeps wanting to install).
(And before you ask why I don't upgrade, I'm still trying to find a decent water resistant phone which works with polarized glasses. The S4 Active seems to work, but alas it's an AT&T exclusive.)
Bear in mind that there are two different things NSA does/did, with very different implications.
1) They weakened cryptographic standards. This deserves all the criticism you're dishing out.
2) They researched how to break crypto. This is completely within their (and anyone else's) right to do. The alternative viewpoint - that merely trying to break crypto should be illegal - is exactly what the MPAA and RIAA have been trying to foist upon us with the draconian provisions in the DMCA prohibiting breaking DRM.
Up to a month ago such a comment would've been modded to -1 because historically, NSA had helped improve the security of encryption standards. As Schneier has said, the revelations about recent NSA activity has completely evaporated the goodwill NSA earned in the cryptographic community from back then.
I dunno. I haven't seen any stories of people being electrocuted by microUSB cables. If you search google for "electrocuted by USB cable" aside from a few people getting shocked (not electrocuted) by malfunctioning USB ports/chargers, every other hit is for Apple's Lightning/USB cable.
It really sounds like Apple/Intel's Lightning interface has a design flaw which makes it easy to either fail or improperly manufacture in a way which leads to potential electrocution. If you choose to reject the industry standard and roll your own, you damn well better make sure it's as safe or safer. KISS. (I thought it might be due to Lightning's higher max power, but some quick research says it maxes out at 12 Watts vs. USB 2.0's 2.5 Watts. While that's higher, it shouldn't be enough to electrocute someone.)
You can get em as cheap as $0.65. I've run into a few microUSB cables which don't work, but that's the beauty of using standardized cables: You can test them on other hardware to figure out if the cable is faulty or the hardware is faulty.
Ostensibly, Apple refused to comply with the EU requirement for microUSB charging because their port allows other functions like audio and video out. But other phones which had similar multi-function outputs just made a microUSB port for charging/data, and a microHDMI port for audio and video.
There's more going on than that. I have an informal email list set up with a few friends. It's basically a simple forwarder - anything sent to list@example.com gets resent to everyone in the mailing list. Somehow list@example.com ended up with a linkedin account, and we were getting emails sent "from" list@example.com asking us to join linkedin. I never saw a mail inviting list@example.com to join linkedin, the first linkedin-related mail I got was "from" list@example.com asking me to join linkedin. I ended up having to take over the account (using a password reset sent to list@example.com) so I could close it.
So clearly (1) it's possible to create an account in your name at linkedin without you actually ever seeing nor agreeing to the TOS, and (2) they're using email addresses obtained via other means to spam people with their invites. I can assure you list@example.com doesn't have a mail account nor an address book. They have to have figured out that my email and list@example.com were related via a third party's address book.
This is a common misconception. The nukes developed during the Cold War weren't for destroying cities. Yeah a few dozen were targeted at the larger cities as an afterthought. But the vast majority were targeted at hardened missile silos. The idea being (as silly as it sounds) to destroy the enemy nukes in their silos before they can be launched at you. Since accuracy wasn't that great back then (remember, no GPS), you had to make up for it with warheads with larger yields (sufficient to destroy several hundred feet thick reinforced concrete), and by trying to make sure you had several times more nukes than the enemy had missile silos.
That last bit is what caused the arms race. If the nukes were only for wiping out cities, a few hundred of them would've been enough for both the U.S. and USSR and there would've been no arms race. The numbers got into the thousands because the U.S. tried to make sure it had 3x more nukes than the Soviets, the Soviets tried to make sure it had 3x more nukes than the U.S., repeat ad nauseum. Mobile launchers, nukes aboard B-52s being flow 24/7, and ballistic missile submarines were all invented as ways to avoid this vulnerability of nukes in missile silos.
Different species of bacteria in your gut form competitive colonies. Antibiotics can wipe out the dominant species, allowing a different species to gain dominance and inhibit the previous dominant species from regaining its original population. Several years ago, my doctor diagnosed my constant stomachaches and vomiting as being caused by a certain type of bacteria which had colonized my stomach. He put me on a treatment of strong antibiotics to wipe them out and allow a more benign gut bacteria to take over. My symptoms went away after the treatment.
Similar things have happened on a macro scale. It's suspected the cod fishery off New England has suffered such a fate after severe overfishing led to its collapse in the 1990s. There have been draconian limits on commercial cod catches for two decades, but no rebound in the cod population. It's suspected that capelin have now taken over as the dominant species in that ecosystem. Capelin used to be eaten by the larger cod. But when overfishing decimated the cod stocks, the capelin were able to grow both in size and population. The theory is the tables have turned now and the capelin are eating the juvenile cod, preventing the cod from reaching the size and numbers which would threaten the dominance of the capelin.
Didn't you get the memo? The courts think sniffing open wifi networks is a violation of wiretap laws.
Can't speak for the Linux version of ZFS, but I run ZFS for FreeBSD in a VM (specifically, FreeNAS 8.0->8.2). My understanding is that after Solaris, the FreeBSD version of ZFS is the most mature. I watched the memory profile for the VM. It would start at about 512 MB, then gradually fill up all the RAM I gave it, then "purge" itself and drop down to 512 MB and repeat.
Upon researching, it looks like the only feature that's really memory-intensive is deduplication. That keeps checksums for every file on disk in memory, so it can quickly detect if a new file you're writing is a dupe. But when I turned it on, it absolutely crushed performance. Writes which had been going at 60-80 MB/sec dropped to about 5 MB/s. I turned off dedupe.
My dad has abnormally dry hands and a thin (as in not very high) fingerprint ridge layer. He hates typing passwords so uses the same short password on everything. A few years ago I got him a Thinkpad with a fingerprint scanner in hopes of beefing up his security without much additional effort.
The scanner only works about 10% of the time on him. He doesn't use it because of the high failure rate. This tells me that although the tech may read some of its data from the interior structure of the finger, the majority of its functionality depends on the fingerprint ridges themselves. (And yes it's the same technology. Apple bought Authentec in 2012, Authentec bought UPEK in 2010, and UPEK made the fingerprint scanners for the Thinkpads.)
Specifically, cable-tapping they used to do with NR-1 (the smallest nuclear-powered sub ever made). It was retired in 2008. In retrospect, it should've been obvious they retired it because they had a better, probably autonomous, method of accomplishing the same task.
A lot of people forget that material properties change with pressure and depth. The first time the Alvin submersible found black smokers (active volcanic vents) on the mid-oceanic ridge, they moved in for a closer look. They found out afterwards that they'd recorded temperatures close to 400 C. The melting point of Alvin's portholes was far less than 400 C, and they would've died if they'd stayed there too long. People see liquid water, and just assume the temperature is below 100 C and therefore the glass portholes are safe. But at the depth they were at, the pressure is much higher and thus the boiling point of water was around 400 C.
I did some quick research. Fracking is typically done 2-3 km underground. The ground temperature at that depth is about 75 C. The pressure at that depth is about 200-300 bar (atmospheres).
Looking at the phase diagram for CO2, that's in the supercritical fluid phase. So the CO2 wouldn't need to be pressurized at that depth like it has to be at sea level. The ground pressure alone would be enough to prevent it from reverting to a gas, and thus it would be impossible for the chamber to catastrophically spring a leak. The only way that could happen is if another drilling operation tapped the chamber and suffered a blowout. Normally that doesn't happen - they keep the bore filled with heavy mud to maintain the pressure at depth. But occasionally (e.g. Deepwater Horizon) there is a blowout, the pressurized mud is lost, and the liquid/gas underneath is then squeezed out by the surrounding rock through the "straw" (bore). I don't see this as being any more risky than regular oil drilling. If anything it's safer since CO2 is pretty inert and won't catch fire. The biggest risk would be the CO2 gas pooling in a depression and suffocating anyone/anything inside.
Not only does the U.S. still have CDMA, most of the rest of the world does too. CDMA won the standards war. The only part of GSM which uses its original TDMA is the voice comms. Most GSM carriers have adopted CDMA or WCDMA for 3G and 3.5G data service (including HSDPA/+).
TDMA sucks because it allocates a timeslice to each phone regardless of whether or not that phone actually transmits during the timeslice. The way CDMA works, every phone can transmit simultaneously and the bandwidth per phone decreases proportionally to the increasing noise floor. i.e. it scales automatically with number of phones transmitting, instead of scaling with the number of phones connected to the tower like TDMA. If it weren't for CDMA, 3G data speeds on GSM would've been limited to about 150 kbps.
That's also the reason GSM phones can do voice and data simultaneously. They have a TDMA radio for voice, and a separate CDMA radio for data. CDMA phones typically have only one CDMA radio, so they can only do either voice or data, not both simultaneously.
CDMA is finally being supplanted by OFDMA (what most implementations of LTE use) because processors have finally become powerful enough to decode the OFDMA signals without draining your battery in 30 minutes. Conceptually, OFDMA is very similar to CDMA, except it operates in the frequency domain instead of code domain. In CDMA each phone is assigned an orthogonal set of codes (e.g. Three phones could be assigned codes AB, BC, and CD. If the phones 1 and 2 transmit simultaneously, the signal the tower sees is ABC, and it knows phones 1 and 2 transmitted while 3 did not. In this simple example, instead of losing 1/3rd of your bandwidth because phone 3 didn't transmit like would happen in TDMA, you only lose 1/4 the bandwidth. The more complex the codes, the less bandwidth you lose). In OFDMA each phone is assigned an orthogonal set of frequencies.
That doesn't work. If the people profiting from an invention wouldn't have had the idea for the invention on their own, then you are not extorting them by making them pay to license the patent.
For stopping (or for that matter, distinguishing) patent trolls, I think we can take a page out of copyright law. With copyrights, you're automatically granted a copyright on anything you make. But if someone violates your copyright, you're limited to compensatory damages. i.e. the offender can only be forced to pay you for actual damages you suffered. To collect statutory damages (the fines that go up to $150k per work regardless of damages suffered), you have to have first registered your copyright. I think if we switch this around a bit, it could solve the problem of patent trolls.
Make it so the original patent filer can collect both compensatory and statutory damages (to protect the little guy who comes up with a great idea, but has trouble bringing it to market while a big company shamelessly steals the idea and takes over the market). But if the patent is transferred, the new patent holder can only collect compensatory damages. That would make it worthless for a person or company to buy a patent solely to sue others for infringement. If they aren't actually building something which uses that patent, then they suffered no damages from the patent infringement and thus aren't able to collect anything from others using that patent. In order to be able to collect damages, you need to be able to show your income was negatively impacted by the infringement, which means you need to be making something which uses the patent.
That would eliminate all the speculating going on with patents. You wouldn't buy a patent in the hopes that you'd be able to collect millions from others for infringement. You'd only buy a patent because you plan to start building something which uses it or it'll improve a product you're already building, and you decide it'll be cheaper to own the patent rather than license it from whoever owns the patent. The main problem I can think of with this idea is you'd end up with a bunch of shell corporations set up to file for patent(s), and people would buy/sell the shell corporation (which is the original patent filer) instead of the patents themselves.
I was a big critic of the bubbly look that was introduced with Windows XP. The new flat look in Office 2013 has finally made me understand one of the advantages of that bubbly look.
Office 2013 (on Win 7, dunno about Win 8) has the flat look but only puts a drop shadow on the foreground window. If I have multiple Excel spreadsheets open in the background, the lack of a drop shadow on background windows makes it very difficult to tell where one ends and another begins. They're all flat and blend together. The bubbly look that Windows XP introduced with its pronounced window borders solves that problem - it makes it (painfully) obvious where the window borders are. XP went overboard with it and Win 7 toned it down. But the way Microsoft has completely abandoned it for the flat look hinders usability in this case.
Back in the 1980s, Apple, IBM, and Microsoft did these huge graphical user interface studies where they tested how people used and reacted to different elements of the interface. They did this to scientifically narrow down what ideas worked and what didn't. With the newer OSes, I get the feeling this methodological approach to user interface design has been abandoned in favor of graphics artists making up whatever they think looks cool.