I bet if a terrorist suspect had used IronKey what would happen is the police would ask GCHQ who would ask the NSA who would ask IronKey. IronKey would then give the NSA information how to get in, because the US government is a very important customer and the nice lady from the NSA had explained it was a matter of public safety for those charming folks in the UK, literally ticking time bomb situation. Then the NSA would tell GCHQ who'd provide the information back to the police. I.e. US and UK companies will by and large roll over when the government tells them and the US and UK government cooperate very efficiently.
Or alternatively suppose there was a company stupid enough to make a device which self destructed and foiled law enforcement and didn't roll over. They'd get sued for obstruction of justice. Probably the people who run the company would find they got audited by the IRS too. Also in the UK the Regulation of Investigatory Powers Act 2015 would pass creating an offence of "conspiring to obstruct justice by storing data relevant to an ongoing investigation on a storage medium which self destructs when law enforcement personnel attempt to access it". Or something like that. Basically usage of the device would be criminalized and the company that made it harassed.
A lot of the time you need to look at the non internet version of things to see why things like the RIPA 2000 Section 49 which people on slashdot complain about are the way they are. Also you can see why most of the technical workarounds to them proposed here are probably either illegal now or would be made illegal if anyone used them successfully.
E.g. suppose a suspect is ordered by a court to provide some documentation and refuses. Does that mean the suspect gets to walk? No, they are prosecuted for contempt of court. This is analogous to refusing to provide a password or key to decrypt data. Knowing how to use TrueCrypt doesn't give you immunity from prosecution. Nor should it.
Similarly suppose a suspect rigs a system to destroy documents when police attempt to access it. Does that mean they get to walk? No, they are prosecuted for obstruction of justice, destroying evidence, contempt of court etc. This is analogous to using a storage medium that self destructs when police attempt to access it. Knowing about IronKey doesn't give you immunity from prosecution. Nor should it.
Clones used a 'clean room engineering' technique. So there were too groups of engineers. One read the source code and the manual and wrote a specification. The other took the specification and wrote a Bios.
With the success of the IBM PC in 1983, Phoenix decided to provide an IBM PC compatible ROM BIOS to the PC market. A licensable ROM BIOS would allow clone PC manufacturers to run the same applications, and the MS-DOS that was being used by IBM. However, to do this Phoenix needed a strategy for defense against IBM copyright infringement lawsuits. Phoenix used a clean room technique that isolated the engineers who had been contaminated by reading the IBM source listings in the IBM Technical Reference Manuals. The contaminated engineers wrote specifications for the BIOS APIs and provided the specifications to a single "clean" engineerâ"one with experience programming the Texas Instruments TMS9900, and without experience with the Intel 8088 or 8086[13]â"who had not been exposed to IBM BIOS source code. The "clean" engineer developed code to mimic the BIOS APIs. This technique provided Phoenix with a defensibly non-infringing IBM PC-compatible ROM BIOS. Because the programmers who wrote the Phoenix code had never read IBM's reference manuals, nothing they wrote could have been copied from IBM's code, no matter how closely the two matched.[14] This reverse engineering technique is commonly referred to as a "Chinese wall." The first Phoenix PC ROM BIOS was introduced in May, 1984, and helped fuel the growth in the PC industry.
Isn't it better if the Evil Big Corporations who use the results of research pay for the research via patent licenses than the tax payer pays?
Suppose I'm a government bureaucrat. I need to decide who gets grants and who does not. Now with most academic fields it's actually non trivial to decide if a particular piece of research is actually worth anything. So I end up doing things like handing out cash to my alma mater, my favourite area, or who pays me off. This system doesn't really work very well because it depends on the government bureaucrat being some sort of philosopher king. Or you can hand out equal amounts of money to all the universities. In which case the good ones have no incentive to stay good - they get the same money as the bad ones regardless.
Now under the patent system academics invent things - and this particular invention is pretty damn non obvious. They patent them. The ones that are doing something useful get cash in terms of patent licenses. The ones that don't, don't. There's no discretion on the part of bureaucrats. In fact inside the university there's no discretion needed to decide which research is useful and which is not - the revenues from patents tell all.
It reminds me of a wonderful story I read about a very early laptop design. It run off NiCd rechargeable AA cells and trickle charged them when plugged in. In the UK where it was invented it sold with a warning label saying "Rechargeable batteries only, laptop will overheat if non rechargeable batteries are used". Trying to trickle charge a non rechargeable battery will make it overheat and potentially burn the user. When they decided to market in the US they were advised that if you burn your users they will sue you, regardless of warning label. So they added a thermistor and turned off the trickle charge if the batteries overheated. Now one way of looking at this is that the scumbag personal injury lawyers are actually carrying information. Even the threat of scumbag personal injury lawyers carries information. And so it is with the patent system. Most people will settle out of court if they're infringing a patent. Thus people can use whatever technology they want. The legal system takes care of making sure the inventor gets paid - either he keeps the patent himself and gets royalties, or he signs it over to his employer and they get them. Or, potentially he or his employer could sell the patents to a patent troll company - i.e. you can choose cash up front or a stream of royalties.
People complain about lawyers, patents and patent trolls but all of them are playing a useful role in a system which automagically channels money to research which is useful without needing any sort of central authority making the decisions.
If you look at Betamax it was regarded as being a bit better than VHS but was less widely licensed. Betamax started off with almost all of the market but gradually lost it because Betamax machines tended to be expensive.
I'd say the analogy is pretty good. High end but proprietary system gradually loses market share to more open, cheaper competitor.
You can buy a very good, cheap Android handset from one of the zillions of Android OEMS. That enables Android to gain market share amongst people who can't afford a more expensive iPhone.
Top selling phones seem to be between £158 (Samsung S3 mini) to £369 (Samsung S4). Now the lifetime is 24 months. So people spend £10 per month to keep their smartphone up to date. Most people don't do this explicitly, rather their telco sells them a plan for much more than £10 a month and gives them a 'free' upgrade every so often as a sweetener.
Prices seem to be £300-400. On the other hand I bet the replacement time is longer. Many people mention 5 years. That's £5 or so a month. So they'd need to spend significantly more on laptops to get to the level of cash they spend on phones.
So it's plausible that people spend more money on keeping their smartphone up to date than their PC.
In fact that's quite plausible. Most people seem to have horrible, sluggish laptops but the very latest smartphone.
Of course if they bought one of these every five years it would work out differently
#6 on the best seller list and £855. So that would be £14 per month assuming you replace it every five years. Incidentally this is one of the reasons why Mac OS taking over from Windows is not a good thing. Most people could get save money by buying one of the vast number of Windows machines compared to buying one of Apple's limited selection of admittedly very high quality machines. A small selection of high end machines means you probably need to spend extra cash to get all the features you need because of the cheapest machine lacks a few.
Of course Microsoft are doing their best to fuck up Windows, so it's not that surprising that people are jumping ship for Mac OS. Windows OEMs must be pretty pissed off at this.
Fish don't split water into hydrogen and oxygen. Rather they extract oxygen dissolved in water. However it seems like there are significant theoretical barriers to such a device because humans need a lot of O2 and seawater only has 7ppm. So you'd need to pass 192 litres of water per minute over the gill surface to get 1 litre or oxygen.
As sea water contains 7 ppm oxygen, 1,000,000 kg (1,000 tonnes) of sea water holds 7 kg (1,000 short tons holds 14 lb) of O2, the equivalent of 5,350 litres (1,410 US gal) of oxygen gas at atmospheric pressure.
An average diver with a fully closed-circuit rebreather needs 1 liter (roughly 1 quart) of oxygen per minute.[8] As a result, at least 192 litres (51 US gal) of sea water per minute would have to be passed through the system, and this system would not work in anoxic water.
On the other hand
Another potential source of oxygen generation is plastron respiration.[10] A foam with hydrophobic surfaces immersed in water becomes superhydrophobic, which provides a water-air interface across which oxygen can diffuse into the foam. In nature, this method is used by some aquatic insects (such as water boatman, Notonecta) and spiders (such as Dolomedes triton) to breathe underwater without a gill. This method was experimentally proven by professor Ed Cussler on his dog
They don't say how big the apparatus was or what the flow rate was. There's an interview with Cussler here.
Suppose I find a vulnerability in some software. I've got two choices
1) Make it public and at best get a mention on slashdot when it is fixed.
2) Sell the details to either the NSA/GCHQ etc or to criminal types. In which case no mention on slashdot, but cash up front.
See the problem with security - any security - is that revealing vulnerabilities to the project so they can be fixed is likely to be much less lucrative than selling them other people who want to exploit them.
If I were cynical here's what I'd do
1) I'd sell details of the exploit to whoever paid the most (Russian Mafia/NSA etc) using an untraceable identity. At this point the vulnerability starts to be exploited by them.
2) I then wait until other security researchers notice this or look like they're about to figure it out. However before they can figure it out completely I report it to the vendor with my normal identity. E.g. Microsoft and Google for example pay cash, so I'd get that.
3) Then even later I'd then announce it publicly at Black Hat and say the vendor hadn't fixed it quickly enough so I've decided to go public. For an open source project (e.g. TrueCrypt) I'd submit a patch and say "Look, I fixed this before anyone knew about it") and make the Black Hat talk about that. So I skip the vendor report stage completely because they won't pay me. However I'd keep stage 1 i.e. "flog it on the open market to the mafia", because that's where the money is.
This - call it Irresponsible Disclosure - optimizes my income - I get it from the criminal types and the vendor if they pay it. It also optimizes my publicity.
Of course the downside is that if the NSA/FBI etc think you're doing this they'll seize your laptop when you go through customs
Then again, that's no bad thing for publicity too - tech sites will cover it as "Fascist government harassing well meaning security researchers". And of course if you get detained for a few hours just use it as an opportunity to negotiate a deal with them to sell the exploits to them exclusively. The government has loads of cash and may well use it to buy up your worthless one man company in return for you agreeing to sell to them exclusively in future.
Actually if you fig a bit deeper there is something interesting about this project. Look at the USB stack - it's all done in software using GPIO pins! Very clever. There's a company that wrote the USB stack. You can get it GPL licensed for free or you can pay for a BSD license, but they start real cheap
This was never about people getting health insurance. This was about making sure those who have chosen not to take personal responsibility for themselves can leech off everyone else without changing their ways. If the President truly wanted to make us all healthier, he would have pushed for higher taxes on tobacco products, forced people to undertake exercise to keep their weight down and would have gone full bore against drug dealers.
Had no idea that it was possible to build a device that interfaces to USB in so few components (it does USB in software on a tiny microcontroller, and the firmware is around 1kb in size...)
That's the genius of USB really. Most early USB devices probably had a Serial Interface engine in hardware and a few hundred bytes of firmware written in assembler in flash or masked prom.
Looking at the firmware main.cpp it implements HID class device. Writing to reportId 1 means you set the 3 PWM oscillators for red, green and blue leds. The other reportIds seem to read and write the EEPROM.
Cheating on benchmarks is only a problem is some people do it. If everyone does it, the benchmarks work again.
Thus you could argue that the companies who refuse assign an intern to reverse engineer their competitors cheats and port those cheats to their products are the problem.
That was just a little app that Office put in the startup folder that called made a dummy OLE call and exited. Calling OLE pulled the OLE dlls into memory where they stayed. Since Office depended very heavily on OLE it made Office apps launch faster. There's nothing uncompetitive about that - Office Fast Start was something that came with Office, not with Windows and OLE was documented on MSDN.
Now, CIFS is pretty much just a serialization of NT I/O semantics over the wire (for some reason this surprises and confuses people from the UNIX/TCP camps. What do they expect?).
In the same way that CIFS aka SMB is a "serialization of NT I/O semantics over the wire", the.doc data formats are a serialization of Word's internal representation of a document. There's a lot of subtlety in that representation as he mentions. The Mac and Windows versions had different epochs, there's a whole bunch of things like "keep this paragraph with next" in the UI that need to be encoded in the file format. Also Office depends heavily on OLE (which was developed for Office) and thus OLE compound documents are part of the spec. They can include Windows Metafiles which are a serialization of GDI commands.
So you ended up with something that had a load of features and run fast on 80's and 90's machines with by current standards glacially slow CPUs and disks. It was cross platform in the sense that it worked on Windows and Mac. It was never really designed to be something that people outside Microsoft could reimplement easily. In fact I bet the original Mac version of Word wasn't intended to run on Windows - the point was to ship it on Mac.
Mind you I use OpenOffice on Windows these days and it seems like.doc and.xls files are now supported pretty well there. On Android Polaris Office seems to have no problems opening.xls files. Now the Open Office and Polaris Office teams must have spent ages getting this stuff to work, but it does. MS is still musing whether it should release Office on Android, but essentially the world has moved on without it.
Here's a novel idea. How about have tablets default to Start Screen and Metro mode and desktops and notebooks defaulting to looking just like Windows 7 - i.e. Start Menu and desktop mode? And having a user option to override that default.
Then the 0.001% of users who exclusively use Metro apps on their tablet would be happy and the rest of us could just ignore it completely. The only reason they're pushing Metro down everyone's throat is so that people write and use Metro apps and the Microsoft store has something to do.
As it is they've got the boat anchor that is Metro dragging down Windows 8 because people who like Windows 7 hate it. It's dragging down Windows RT too because no compelling Metro apps means that Windows RT is screwed. It's dragging down the Windows Store because no one actually wants Metro.
They've got one very unpopular product - Metro and a number of very popular ones - most notably Windows itself. They've tried to force the people that like and use Windows to use Metro. And probably the reason for that is because if Metro apps take off then so will Windows Phone. Which right now is tanking too.
However instead of this strategy making Metro and Windows Phone more popular they've actually managed to make desktop Windows less popular. PC sales are down and they've made Windows run much less well on non touchscreen machines but the tablets people are buying instead are running Android and iOS, not Windows.
Exactly. The British weren't as obviously vicious as Imperial Japan or Nazi Germany or the earlier European empires but still there were regular famines during British rule that stopped after they left. So British rule killed millions of people.
And Islamic rule of India was no joke either. The Hindu Kush is so named because Hindu slaves died in vast numbers on their way to slave markets in Islamic lands.
The world is not a very nice place and countries need to be able to defend themselves. Otherwise their inhabitants will be enslaved or slaughtered.
Therefore, India isn't as was considered earlier a nation with a failed thermonuclear test but one with a compact dial-able thermonuclear warhead capable country with places them in a block of just three countries or four countries(if you include the likes of piggybacker UK) with such advanced warheads.
I wish people in the UK who claim that Trident is "too expensive" would realise this. if we keep Trident we've got a system that would have cost us far far more to develop from scratch than to co-develop with the US.
If the UK scrapped Trident the sharing would stop and we'd probably sack all the scientists. Then if we needed nukes in the future we'd have to develop from scratch with a new set of scientists and no help - indeed active attempts to stop "proliferation" - from the US.
The few billion the UK will spend per year to stay part of the club is a lot less than it would have to pay on a crash program to rebuild a scrapped program with no help from anyone else.
The whole thing is insane really. At the start MS had 90% of the desktop market. Windows Mobile had about 10-20% of the mobile market. Most importantly they had a load of ISVs producing software, the old stuff run on Win32 and the new stuff on.Net.
MS introduce the Kin and Zune. These were spectacular failures - based on.Net and C#
Then MS decide to replace Windows Mobile with Windows Phone 7. It is based on C#/.Net and is locked to prevent Win32 code. It share a lot with Kin and Zune. It is a failure. They replace it with WP8. WP8 is locked to prevent Win32 code except for Microsoft's code - IE and Office are still Win32. Everyone else is supposed to use the WinRT API in C++. Then they move the WP8 API to Windows 8 and release an ARM version which is locked to prevent Win32 code. Windows Phone is now down to a few percent market share. Most of the ISVs defected to Android and iOS and show no sign of coming back.
So you've got a UI which they used on their phone project which is not selling on their desktop OS which is. At that point it seems like people stopped buying machines with Windows 8 - if you look at what happened Windows 7 is still outselling it.
Now if you look at Vista it sold poorly and they rushed out Windows 7. So you'd expect them to rush out a Windows 9 which had the start menu restored. But if you look at Windows Blue the biggest change is apparently "an improved charms menu".
If you could compile apps you'd see a bit of third party support. Perhaps some web browsers and some other stuff.
Now admittedly that was true of Alpha, MIPS and PowerPC but they didn't sell well enough to see much software getting cross compiled. Mind you at that point the world was x86. Now the world is x86/x64 - i.e. a lot more stuff is already running on two architectures. In fact quite a lot of mobile stuff already ran on Arm because of Windows CE/Windows Mobile - Opera Mobile for example.
So if Windows RT had allowed it you'd see some applications running on Arm. Banning all third party Win32 applications in the hope of moving everyone to Metro clearly isn't working.
Slashdot Mirror
They deport you to Saudi Arabia where you'd be water boarded, drugged, beaten and raped for years.
No, just kidding. Nothing at all.
Mind you though, I'd definitely write your passwords down on a Post it note and stick it under your desk if I were you just to be safe.
I bet if a terrorist suspect had used IronKey what would happen is the police would ask GCHQ who would ask the NSA who would ask IronKey. IronKey would then give the NSA information how to get in, because the US government is a very important customer and the nice lady from the NSA had explained it was a matter of public safety for those charming folks in the UK, literally ticking time bomb situation. Then the NSA would tell GCHQ who'd provide the information back to the police. I.e. US and UK companies will by and large roll over when the government tells them and the US and UK government cooperate very efficiently.
Or alternatively suppose there was a company stupid enough to make a device which self destructed and foiled law enforcement and didn't roll over. They'd get sued for obstruction of justice. Probably the people who run the company would find they got audited by the IRS too. Also in the UK the Regulation of Investigatory Powers Act 2015 would pass creating an offence of "conspiring to obstruct justice by storing data relevant to an ongoing investigation on a storage medium which self destructs when law enforcement personnel attempt to access it". Or something like that. Basically usage of the device would be criminalized and the company that made it harassed.
A lot of the time you need to look at the non internet version of things to see why things like the RIPA 2000 Section 49 which people on slashdot complain about are the way they are. Also you can see why most of the technical workarounds to them proposed here are probably either illegal now or would be made illegal if anyone used them successfully.
E.g. suppose a suspect is ordered by a court to provide some documentation and refuses. Does that mean the suspect gets to walk? No, they are prosecuted for contempt of court. This is analogous to refusing to provide a password or key to decrypt data. Knowing how to use TrueCrypt doesn't give you immunity from prosecution. Nor should it.
Similarly suppose a suspect rigs a system to destroy documents when police attempt to access it. Does that mean they get to walk? No, they are prosecuted for obstruction of justice, destroying evidence, contempt of court etc. This is analogous to using a storage medium that self destructs when police attempt to access it. Knowing about IronKey doesn't give you immunity from prosecution. Nor should it.
Clones used a 'clean room engineering' technique. So there were too groups of engineers. One read the source code and the manual and wrote a specification. The other took the specification and wrote a Bios.
http://en.wikipedia.org/wiki/Phoenix_Technologies#Cloning_the_IBM_PC_BIOS
With the success of the IBM PC in 1983, Phoenix decided to provide an IBM PC compatible ROM BIOS to the PC market. A licensable ROM BIOS would allow clone PC manufacturers to run the same applications, and the MS-DOS that was being used by IBM. However, to do this Phoenix needed a strategy for defense against IBM copyright infringement lawsuits. Phoenix used a clean room technique that isolated the engineers who had been contaminated by reading the IBM source listings in the IBM Technical Reference Manuals. The contaminated engineers wrote specifications for the BIOS APIs and provided the specifications to a single "clean" engineerâ"one with experience programming the Texas Instruments TMS9900, and without experience with the Intel 8088 or 8086[13]â"who had not been exposed to IBM BIOS source code. The "clean" engineer developed code to mimic the BIOS APIs. This technique provided Phoenix with a defensibly non-infringing IBM PC-compatible ROM BIOS. Because the programmers who wrote the Phoenix code had never read IBM's reference manuals, nothing they wrote could have been copied from IBM's code, no matter how closely the two matched.[14] This reverse engineering technique is commonly referred to as a "Chinese wall." The first Phoenix PC ROM BIOS was introduced in May, 1984, and helped fuel the growth in the PC industry.
Isn't it better if the Evil Big Corporations who use the results of research pay for the research via patent licenses than the tax payer pays?
Suppose I'm a government bureaucrat. I need to decide who gets grants and who does not. Now with most academic fields it's actually non trivial to decide if a particular piece of research is actually worth anything. So I end up doing things like handing out cash to my alma mater, my favourite area, or who pays me off. This system doesn't really work very well because it depends on the government bureaucrat being some sort of philosopher king. Or you can hand out equal amounts of money to all the universities. In which case the good ones have no incentive to stay good - they get the same money as the bad ones regardless.
Now under the patent system academics invent things - and this particular invention is pretty damn non obvious. They patent them. The ones that are doing something useful get cash in terms of patent licenses. The ones that don't, don't. There's no discretion on the part of bureaucrats. In fact inside the university there's no discretion needed to decide which research is useful and which is not - the revenues from patents tell all.
It reminds me of a wonderful story I read about a very early laptop design. It run off NiCd rechargeable AA cells and trickle charged them when plugged in. In the UK where it was invented it sold with a warning label saying "Rechargeable batteries only, laptop will overheat if non rechargeable batteries are used". Trying to trickle charge a non rechargeable battery will make it overheat and potentially burn the user. When they decided to market in the US they were advised that if you burn your users they will sue you, regardless of warning label. So they added a thermistor and turned off the trickle charge if the batteries overheated. Now one way of looking at this is that the scumbag personal injury lawyers are actually carrying information. Even the threat of scumbag personal injury lawyers carries information. And so it is with the patent system. Most people will settle out of court if they're infringing a patent. Thus people can use whatever technology they want. The legal system takes care of making sure the inventor gets paid - either he keeps the patent himself and gets royalties, or he signs it over to his employer and they get them. Or, potentially he or his employer could sell the patents to a patent troll company - i.e. you can choose cash up front or a stream of royalties.
People complain about lawyers, patents and patent trolls but all of them are playing a useful role in a system which automagically channels money to research which is useful without needing any sort of central authority making the decisions.
If you look at Betamax it was regarded as being a bit better than VHS but was less widely licensed. Betamax started off with almost all of the market but gradually lost it because Betamax machines tended to be expensive.
I'd say the analogy is pretty good. High end but proprietary system gradually loses market share to more open, cheaper competitor.
You can buy a very good, cheap Android handset from one of the zillions of Android OEMS. That enables Android to gain market share amongst people who can't afford a more expensive iPhone.
The interesting thing about Android is the amount of money involved. E.g. in the UK
http://www.amazon.co.uk/Best-Sellers-Electronics-Mobile-Phones-Smartphones/zgbs/electronics/356496011
Top selling phones seem to be between £158 (Samsung S3 mini) to £369 (Samsung S4). Now the lifetime is 24 months. So people spend £10 per month to keep their smartphone up to date. Most people don't do this explicitly, rather their telco sells them a plan for much more than £10 a month and gives them a 'free' upgrade every so often as a sweetener.
Now for PCs
http://www.amazon.co.uk/Best-Sellers-Computers-Accessories-Laptops/zgbs/computers/429886031/ref=zg_bs_nav_computers_1_computers
Prices seem to be £300-400. On the other hand I bet the replacement time is longer. Many people mention 5 years. That's £5 or so a month. So they'd need to spend significantly more on laptops to get to the level of cash they spend on phones.
So it's plausible that people spend more money on keeping their smartphone up to date than their PC.
In fact that's quite plausible. Most people seem to have horrible, sluggish laptops but the very latest smartphone.
Of course if they bought one of these every five years it would work out differently
http://www.amazon.co.uk/Apple-13-inch-MacBook-2-5GHz-Graphics/dp/B008BEYEL8/ref=zg_bs_429886031_6
#6 on the best seller list and £855. So that would be £14 per month assuming you replace it every five years. Incidentally this is one of the reasons why Mac OS taking over from Windows is not a good thing. Most people could get save money by buying one of the vast number of Windows machines compared to buying one of Apple's limited selection of admittedly very high quality machines. A small selection of high end machines means you probably need to spend extra cash to get all the features you need because of the cheapest machine lacks a few.
Of course Microsoft are doing their best to fuck up Windows, so it's not that surprising that people are jumping ship for Mac OS. Windows OEMs must be pretty pissed off at this.
Fish don't split water into hydrogen and oxygen. Rather they extract oxygen dissolved in water. However it seems like there are significant theoretical barriers to such a device because humans need a lot of O2 and seawater only has 7ppm. So you'd need to pass 192 litres of water per minute over the gill surface to get 1 litre or oxygen.
http://en.wikipedia.org/wiki/Artificial_gills_(human)
As sea water contains 7 ppm oxygen, 1,000,000 kg (1,000 tonnes) of sea water holds 7 kg (1,000 short tons holds 14 lb) of O2, the equivalent of 5,350 litres (1,410 US gal) of oxygen gas at atmospheric pressure.
An average diver with a fully closed-circuit rebreather needs 1 liter (roughly 1 quart) of oxygen per minute.[8] As a result, at least 192 litres (51 US gal) of sea water per minute would have to be passed through the system, and this system would not work in anoxic water.
On the other hand
Another potential source of oxygen generation is plastron respiration.[10] A foam with hydrophobic surfaces immersed in water becomes superhydrophobic, which provides a water-air interface across which oxygen can diffuse into the foam. In nature, this method is used by some aquatic insects (such as water boatman, Notonecta) and spiders (such as Dolomedes triton) to breathe underwater without a gill. This method was experimentally proven by professor Ed Cussler on his dog
They don't say how big the apparatus was or what the flow rate was. There's an interview with Cussler here.
http://www.naturesraincoats.com/Experiments_Plastron%20Respiration.html
If you look here it seems like artificial gills do need a high flow rate.
There's an interesting New Scientist article about artificial gills here
http://s3.amazonaws.com/lcp/artedi/myfiles/Breathing%20in%20oceans.pdf
Suppose I find a vulnerability in some software. I've got two choices
1) Make it public and at best get a mention on slashdot when it is fixed.
2) Sell the details to either the NSA/GCHQ etc or to criminal types. In which case no mention on slashdot, but cash up front.
See the problem with security - any security - is that revealing vulnerabilities to the project so they can be fixed is likely to be much less lucrative than selling them other people who want to exploit them.
If I were cynical here's what I'd do
1) I'd sell details of the exploit to whoever paid the most (Russian Mafia/NSA etc) using an untraceable identity. At this point the vulnerability starts to be exploited by them.
2) I then wait until other security researchers notice this or look like they're about to figure it out. However before they can figure it out completely I report it to the vendor with my normal identity. E.g. Microsoft and Google for example pay cash, so I'd get that.
3) Then even later I'd then announce it publicly at Black Hat and say the vendor hadn't fixed it quickly enough so I've decided to go public. For an open source project (e.g. TrueCrypt) I'd submit a patch and say "Look, I fixed this before anyone knew about it") and make the Black Hat talk about that. So I skip the vendor report stage completely because they won't pay me. However I'd keep stage 1 i.e. "flog it on the open market to the mafia", because that's where the money is.
This - call it Irresponsible Disclosure - optimizes my income - I get it from the criminal types and the vendor if they pay it. It also optimizes my publicity.
Of course the downside is that if the NSA/FBI etc think you're doing this they'll seize your laptop when you go through customs
http://yro.slashdot.org/story/10/11/20/0332243/whitehat-hacker-moxie-marlinspikes-laptop-cellphones-seized
Then again, that's no bad thing for publicity too - tech sites will cover it as "Fascist government harassing well meaning security researchers". And of course if you get detained for a few hours just use it as an opportunity to negotiate a deal with them to sell the exploits to them exclusively. The government has loads of cash and may well use it to buy up your worthless one man company in return for you agreeing to sell to them exclusively in future.
It's that Southern Cypherpunk series of books about a hacker/waitress, Mary Sue Stackdump.
I have just calculated that in my head. Am I safe?
Actually if you fig a bit deeper there is something interesting about this project. Look at the USB stack - it's all done in software using GPIO pins! Very clever. There's a company that wrote the USB stack. You can get it GPL licensed for free or you can pay for a BSD license, but they start real cheap
http://www.obdev.at/products/vusb/license.html
They've got a load of projects too
http://www.obdev.at/products/vusb/prjobdev.html
If you go above 10,000 units you probably pay more but by that point you can afford it.
Very interesting mix of clever code and a well thought out business model I think.
This was never about people getting health insurance. This was about making sure those who have chosen not to take personal responsibility for themselves can leech off everyone else without changing their ways. If the President truly wanted to make us all healthier, he would have pushed for higher taxes on tobacco products, forced people to undertake exercise to keep their weight down and would have gone full bore against drug dealers.
https://www.youtube.com/watch?v=PWkWQ-39KLo
Had no idea that it was possible to build a device that interfaces to USB in so few components (it does USB in software on a tiny microcontroller, and the firmware is around 1kb in size...)
That's the genius of USB really. Most early USB devices probably had a Serial Interface engine in hardware and a few hundred bytes of firmware written in assembler in flash or masked prom.
This device is actually quite high end
http://www.atmel.com/devices/attiny85.aspx
http://www.atmel.com/Images/Atmel-2586-AVR-8-bit-Microcontroller-ATtiny25-ATtiny45-ATtiny85_Datasheet.pdf
You've got 8KB of flash. You can program it in C and you get a USB driver.
http://www.blinkstick.com/help/firmware
Looking at the firmware main.cpp it implements HID class device. Writing to reportId 1 means you set the 3 PWM oscillators for red, green and blue leds. The other reportIds seem to read and write the EEPROM.
Cheating on benchmarks is only a problem is some people do it. If everyone does it, the benchmarks work again.
Thus you could argue that the companies who refuse assign an intern to reverse engineer their competitors cheats and port those cheats to their products are the problem.
Nonsense. People would dress well in Hugo Boss uniforms and ze trains would run on time.
You mean Office Fast Start?
http://smallbusiness.support.microsoft.com/en-us/kb/132755
That was just a little app that Office put in the startup folder that called made a dummy OLE call and exited. Calling OLE pulled the OLE dlls into memory where they stayed. Since Office depended very heavily on OLE it made Office apps launch faster. There's nothing uncompetitive about that - Office Fast Start was something that came with Office, not with Windows and OLE was documented on MSDN.
Exactly. It reminds me a bit of this
http://blogs.msdn.com/b/oldnewthing/archive/2007/02/01/1573160.aspx#1591874
Now, CIFS is pretty much just a serialization of NT I/O semantics over the wire (for some reason this surprises and confuses people from the UNIX/TCP camps. What do they expect?).
In the same way that CIFS aka SMB is a "serialization of NT I/O semantics over the wire", the .doc data formats are a serialization of Word's internal representation of a document. There's a lot of subtlety in that representation as he mentions. The Mac and Windows versions had different epochs, there's a whole bunch of things like "keep this paragraph with next" in the UI that need to be encoded in the file format. Also Office depends heavily on OLE (which was developed for Office) and thus OLE compound documents are part of the spec. They can include Windows Metafiles which are a serialization of GDI commands.
So you ended up with something that had a load of features and run fast on 80's and 90's machines with by current standards glacially slow CPUs and disks. It was cross platform in the sense that it worked on Windows and Mac. It was never really designed to be something that people outside Microsoft could reimplement easily. In fact I bet the original Mac version of Word wasn't intended to run on Windows - the point was to ship it on Mac.
Mind you I use OpenOffice on Windows these days and it seems like .doc and .xls files are now supported pretty well there. On Android Polaris Office seems to have no problems opening .xls files. Now the Open Office and Polaris Office teams must have spent ages getting this stuff to work, but it does. MS is still musing whether it should release Office on Android, but essentially the world has moved on without it.
Here's a novel idea. How about have tablets default to Start Screen and Metro mode and desktops and notebooks defaulting to looking just like Windows 7 - i.e. Start Menu and desktop mode? And having a user option to override that default.
Then the 0.001% of users who exclusively use Metro apps on their tablet would be happy and the rest of us could just ignore it completely. The only reason they're pushing Metro down everyone's throat is so that people write and use Metro apps and the Microsoft store has something to do.
As it is they've got the boat anchor that is Metro dragging down Windows 8 because people who like Windows 7 hate it. It's dragging down Windows RT too because no compelling Metro apps means that Windows RT is screwed. It's dragging down the Windows Store because no one actually wants Metro.
They've got one very unpopular product - Metro and a number of very popular ones - most notably Windows itself. They've tried to force the people that like and use Windows to use Metro. And probably the reason for that is because if Metro apps take off then so will Windows Phone. Which right now is tanking too.
However instead of this strategy making Metro and Windows Phone more popular they've actually managed to make desktop Windows less popular. PC sales are down and they've made Windows run much less well on non touchscreen machines but the tablets people are buying instead are running Android and iOS, not Windows.
He's jiggling it back and forth, muttering 'work you fucking stupid bastard' under his breath.
Exactly. The British weren't as obviously vicious as Imperial Japan or Nazi Germany or the earlier European empires but still there were regular famines during British rule that stopped after they left. So British rule killed millions of people.
And Islamic rule of India was no joke either. The Hindu Kush is so named because Hindu slaves died in vast numbers on their way to slave markets in Islamic lands.
The world is not a very nice place and countries need to be able to defend themselves. Otherwise their inhabitants will be enslaved or slaughtered.
Soylent green tastes a bit gamey?
Therefore, India isn't as was considered earlier a nation with a failed thermonuclear test but one with a compact dial-able thermonuclear warhead capable country with places them in a block of just three countries or four countries(if you include the likes of piggybacker UK) with such advanced warheads.
I wish people in the UK who claim that Trident is "too expensive" would realise this. if we keep Trident we've got a system that would have cost us far far more to develop from scratch than to co-develop with the US.
If the UK scrapped Trident the sharing would stop and we'd probably sack all the scientists. Then if we needed nukes in the future we'd have to develop from scratch with a new set of scientists and no help - indeed active attempts to stop "proliferation" - from the US.
The few billion the UK will spend per year to stay part of the club is a lot less than it would have to pay on a crash program to rebuild a scrapped program with no help from anyone else.
Should call it the Instagramcorder.
Calling it 'Bollocks' is designed to subliminally attract the young, female demographic that use Instagram.
The whole thing is insane really. At the start MS had 90% of the desktop market. Windows Mobile had about 10-20% of the mobile market. Most importantly they had a load of ISVs producing software, the old stuff run on Win32 and the new stuff on .Net.
MS introduce the Kin and Zune. These were spectacular failures - based on .Net and C#
Then MS decide to replace Windows Mobile with Windows Phone 7. It is based on C#/.Net and is locked to prevent Win32 code. It share a lot with Kin and Zune. It is a failure. They replace it with WP8. WP8 is locked to prevent Win32 code except for Microsoft's code - IE and Office are still Win32. Everyone else is supposed to use the WinRT API in C++. Then they move the WP8 API to Windows 8 and release an ARM version which is locked to prevent Win32 code. Windows Phone is now down to a few percent market share. Most of the ISVs defected to Android and iOS and show no sign of coming back.
So you've got a UI which they used on their phone project which is not selling on their desktop OS which is. At that point it seems like people stopped buying machines with Windows 8 - if you look at what happened Windows 7 is still outselling it.
Now if you look at Vista it sold poorly and they rushed out Windows 7. So you'd expect them to rush out a Windows 9 which had the start menu restored. But if you look at Windows Blue the biggest change is apparently "an improved charms menu".
If you could compile apps you'd see a bit of third party support. Perhaps some web browsers and some other stuff.
Now admittedly that was true of Alpha, MIPS and PowerPC but they didn't sell well enough to see much software getting cross compiled. Mind you at that point the world was x86. Now the world is x86/x64 - i.e. a lot more stuff is already running on two architectures. In fact quite a lot of mobile stuff already ran on Arm because of Windows CE/Windows Mobile - Opera Mobile for example.
So if Windows RT had allowed it you'd see some applications running on Arm. Banning all third party Win32 applications in the hope of moving everyone to Metro clearly isn't working.