This sort of thing is why Linux is a failure on the desktop. The "real way" to do something is by editing a text file. Then there are third party hacks to make it "easier", until the third party hack screws up something.
Right. Fogscreen does this commercially. With better image quality, too. "Fogscreen" really is a fog screen. Here's Fogscreen in HD video, so you can see the quality of Fogscreen, which is OK for PR but not that great. They do interactivity, too.
Water screens are available, too. Those things can be huge, hundreds of feet long if desired.
All these technologies suffer from poor resolution. It's hard to keep a layer of fog smooth and flat. Resolution gets worse further from the nozzles, too.
It's not that there's not more to discover, it's that the cost and effort for major discoveries has gone up. This is especially true in high-energy physics, where each generation of accelerators is far more expensive than the previous one. On the other hand, there's been lots of discovery in low-energy physics in recent decades. Exploring physics around absolute zero has been very productive and not hugely expensive. Semiconductor device physics continues to make progress. Lots of low-energy effects once thought useless, like the Hall effect, turn out to have practical applications.
But the return on investment for basic research really has decreased. That's why big corporate research labs have disappeared in the US. AT&T and IBM used to do basic research in physics, and out of that came the transistor and the high density disk drive. Few companies do that today. Even pharma research is very product-focused.
You're going to spend way too much time trying to get Samba to do all the funny stuff that an ancient farm of XPs has set up to share. You're going to have to do this in stages, which means replicating the exact sharing structure of the old machines. The users won't be able to do this themselves. ("After editing/etc/samba/smb.conf, restart Samba for the changes to take effect." - Ubuntu documentation)
Then you have to get everybody converted from Microsoft Office to OpenOffice/LibreOffice. They're not that compatible. Some documents and spreadsheets will be broken. Templates won't transfer very well. Everyone's workflow will be disrupted. The overhead of doing this for a small shop will be higher than the savings.
The small office environment is where the Microsoft environment does best. Upgrade to Windows 7, one machine at a time. Windows 7 is a good OS. (The solid Microsoft OSs were NT 3.51, Windows 2000, and Windows 7.)
This is a technololgy that's almost there. Like laser weapons. Big chemical lasers that could shoot down shells or small missiles were built two decades ago, but they were building-sized installations, used huge amounts of hazardous chemicals, and took a long time to cool between shots. A decade ago, the THEL laser system had that down to three semi-trailers, but it still used big tanks of hazardous chemicals. Recently, big arrays of solid-state lasers have been used to shoot down shells and small missiles, and that system fits on a medium sized truck. The current version is only 10KW, and the consensus is that about 50KW to 100KW is needed to be really effective.
What's "mobile" about it? Runs on batteries? Plays crappy cellphone games? No. It's cordless.
That's good, but it has nothing to do with mobile phones. Even GameFace uses the term "cordless", not "mobile".
The site is kind of vague on what processing takes place in the headgear, and what takes place on the external WiFi connected device where, presumably, the game is playing. This thing is only worth the trouble if the game behind it is rendering very fast and has very high resolution content, and the latency to the game is very low.
Here's the argument for non-obviousness from the patent:
One problem associated with using touch screens on portable devices is the unintentional activation or deactivation of functions due to unintentional contact with the touch screen. Thus, portable devices, touch screens on such devices, and/or applications running on such devices may be locked upon satisfaction of predefined lock conditions, such as upon entering an active call, after a predetermined time of idleness has elapsed, or upon manual locking by a user.
Devices with touch screens and/or applications running on such devices may be unlocked by any of several well-known unlocking procedures, such as pressing a predefined set of buttons (simultaneously or sequentially) or entering a code or password. These unlock procedures, however, have drawbacks. The button combinations may be hard to perform. Creating, memorizing, and recalling passwords, codes, and the like can be quite burdensome. These drawbacks may reduce the ease of use of the unlocking process and, as a consequence, the ease of use of the device in general.
Accordingly, there is a need for more efficient, user-friendly procedures for unlocking such devices, touch screens, and/or applications.
Apple was trying to come up with a way to prevent butt-dialing and other unwanted device actions. The point of using a sliding motion is that it's unlikely to happen via random touches, but is reasonably intutive.
Microsoft's video doesn't really show a slider. It shows touch buttons that look visually like sliders. But you can trip them just by touching in the active area for the desired state. This is shown in the video where the demonstrator runs their finger down a column of switches and they all switch. Apple requires an explicit "click and drag" operation to unlock.
It may seem trivial, but if nobody did it before the patent, and everybody wanted to do it after the patent, it's a valid invention. "Obvious" does not mean "obvious in hindsight".
This is Yet Another Crap Extruder based printer. That whole class of machines sort of works on good days. None of them Just Work.
The fundamental problem is that they're welding a hot thing to a cold thing. That sucks for metal welding, it sucks for soldering, and it sucks for plastic welding. It's how you get bad welds, cold solder joints, and fractures in 3D printing. The heated build plate systems usually start a build OK, but a few cm from the build plate, that heat source isn't close enough to help much. So many taller builds fail around 2-5 cm.
For this process to work, it needs better temperature control. A heated build chamber (that's patented). A hot air jet or small laser aimed at the target just before the weld (larger plastic welders do this). But nobody seems to be doing that. They just keep coming up with variations in the 3-axis motion mechanism (not hard to get right) and the software (not really the problem). Or they add DRM and overcharge for "print cartridges".
It's just 70 gallons of crude oil left in an unused pipeline. No fire. No explosion. Just a mess.
It's not like a few years ago, when a high pressure gas pipeline exploded in Daly City and took out a small subdivision. Now that was a serious problem and an indication of a worse one. The column of smoke was visible 20 miles away. The state of California made PG&E do hydrostatic testing on all their major gas pipelines, over PG&E's claims that it was unnecessary. During hydrostatic testing with water, there were two pipeline bursts. One caused a landslide that blocked parts of I-280 at Woodside CA. No fire, of course; just water and mud, since this happened during testing.
OK, first bypass the click troll and get to the actual paper.
The general idea seems to be to transmit a large amount of noisy data per plaintext bit. Historically, crypto schemes which make the input much bigger are disfavored, but communications bandwidth is cheaper now and that might be OK.
The author of the paper seems to have fallen into the old trap of thinking that that analog signals have infinite amounts of data in them. He writes things like ''The encrypting key space is unbounded." and "The choice of the form of coupling functions comes from a set of functions that is not bounded." ("High-end" audio people also fall for this.) In reality, at some point you hit a noise threshold, and, anyway, down at the bottom, electrons and photons are discrite. Also, to be usable, whatever is used for the key has to be of finite size, and preferably not too big.
"No new cypher is worth looking at unless it comes from someone who has already broken a very hard one. - Friedman.
That looks like library code that the compiler generated. Maybe some kind of strcpy variant.
Read the analysis of the code. It's not. It is, however, decompiled assembly code; the people doing the analysis don't have access to the source.
What you have to keep in mind is that this software was written for Windows 98.
Irrelevant. This code is vulnerable on any OS that lets it get UDP packets.
We don't have enough context to know.
Read the actual vulnerability report. There's enough context there.
The real morons are the ones who tried to network it, presumably against Yokogawa's recommendations.
You have to assume today that if it has an Internet-accessable interface, an attacker will find a way to get to it from the public Internet. Because, in practice, attackers do.
Google already has several TV interface devices on the market. There's Google Chromecast, of course. Google also sells a set top-box used with Google Fiber That also comes with the Google Storage Box, which is a 2TB file server for storing downloaded content.
There's the old Google TV, which is mostly Android software inside.
So Google has this covered already. They have a device for viewing TV over the Internet, and they have a cable box for their cable system. They're probably going to tweak the UI on one of those and promote that as a new product.
Neither Google nor Facebook has ever successfully built a product users will actually pay for. (Google's Nexus phones are rebranded LG, Samsung, and Asus products). For both, all significant revenue is from ads.
Yet both have now acquired hardware companies. Now they have to make a business out of them. They may not succeed.
Google acquired Motorola and had no idea what to do with it. Now they're selling it. Google has an automatic driving R&D project, but they acquired DARPA Grand Challenge technology and seem no closer to deployment than a few years ago.
Google acquired a half dozen advanced robotics R&D firms, but none of those have commercial products or profits yet. Google now has to build an entire industrial business in robotics, which is slow, hard, and will take years to pay off. Google hasn't shown the corporate patience for that. Google products that didn't take off quickly are usually killed. I'm worried that Google will end up trashing the US robotics industry once they realize it's not a Make Money Fast business.
Facebook hasn't really tried yet in hardware. But they have no expertise at it. The Oculus Rift is still a prototype/low volume device. Facebook has never run a factory. They'll have to outsource manufacturing, which means everybody else will be making goggles if it turns out to be profitable to do so.
What's a good no-nonsense registrar for major TLDs? It doesn't have to be super cheap. I want to dump Network Solutions because they gave me an unsolicited domain (I had.com and.net; they gave me a useless.info) which they then expect me to pay to renew.
I have about five domains. I want to avoid the "bulk" domain companies like GoDaddy.
I'm afraid, however, that the Real Men Don't Need Bound Checks mentality that is prevalent among C programmers will be a big obstacle.
I've run into that. Usually from second-rate programmers. Programmers who think that way should be put them on maintenance programming for a while. Have them debug program crashes in code written by others.
for ( i = 0; v3 != '\n'; ++v2)// Dangerous loop, copying data to a stack buffer, until an end of line is found
{
if ( v3 == '\r' )
break;
*(_BYTE *)(i + a1) = v3;// Byte copy to the stack, without having destination size into account.
v3 = *(_BYTE *)(v2 + 1);
++i;
}
The company that let that code out the door should be sued for gross negligence, and managers fired. That's not the only example; they failed to do basic checks at least three times. This isn't a subtle bug. This is failing C Programming 101.
(Several times, I've tried to convince the C standards committee to put a "strict mode" in the language and move towards a form of C that's resistant to buffer overflow problems. Maybe I should try again.)
C - now with over thirty years of buffer overflows.
Google has re-invented LittleBits, a family of electronics modules which are attached with magnets. With their new "Cloud" module and their Arduino module, you might even be able to build a wireless VOIP phone.
This is a fun hobbyist concept, but you don't actually use things built that way.
Either this will be bulky or the components will be fragile. You pay a penalty for all that casing and standard form factor.
Somebody (Wyse, I think) built a PC like this in the early 80s. Each module looked like a book, and plugged into the module next to it. You lined up all the modules, pushed them together, and put a big pin with a knob through the stack to lock them all together. Total failure as a product.
What Google should be doing, after buying all those robotics companies, is designing a phone for 100% robotic assembly, so they don't need Foxconn. (Except that Motorola did that a decade ago.)
I think a fire alarm is an instance where I'd like something to have as simple and foolproof a mechanism as possible.
Yes. That's why fire sprinklers are so successful. There's nothing between the water and the fire except a low-melting-point component in the sprinkler head.
This is an example of webcrap-level programmers doing things they're not qualified to do. I'm beginning to think that "Internet of Things" programmers should be required to have Registered Professional Engineer credentials, like structural engineers.
Bitcoin is about as important as the Segway. Bitcoin works, and it does what it's supposed to do - allow one-way irrevocable value transfers between anonymous parties. That's a poor basis for practical commerce, although a boon to a broad range of crooks. It's especially bad that over half of Bitcoin exchanges have gone bust, despite the fact that they shouldn't be at financial risk.
It looks like Bitcoin will be around for a while, but it's not growing. The block chain transaction rate now is about what it was a year ago. Like the Segway, Bitcoin is kind of cool, and has niche applications, but it's not changing the world.
Most low-level employees today are depressed. And underpaid. In a poll, 81% of fast-food employees report having wages stolen from them in the form of unpaid time.
Slashdot Mirror
I don't know why it says to edit smb.conf directly when the easy button way to set up Samba is with "system-config-samba"
Me either. But that's what the manual says.
This sort of thing is why Linux is a failure on the desktop. The "real way" to do something is by editing a text file. Then there are third party hacks to make it "easier", until the third party hack screws up something.
Right. Fogscreen does this commercially. With better image quality, too. "Fogscreen" really is a fog screen. Here's Fogscreen in HD video, so you can see the quality of Fogscreen, which is OK for PR but not that great. They do interactivity, too.
Water screens are available, too. Those things can be huge, hundreds of feet long if desired.
All these technologies suffer from poor resolution. It's hard to keep a layer of fog smooth and flat. Resolution gets worse further from the nozzles, too.
It's not that there's not more to discover, it's that the cost and effort for major discoveries has gone up. This is especially true in high-energy physics, where each generation of accelerators is far more expensive than the previous one. On the other hand, there's been lots of discovery in low-energy physics in recent decades. Exploring physics around absolute zero has been very productive and not hugely expensive. Semiconductor device physics continues to make progress. Lots of low-energy effects once thought useless, like the Hall effect, turn out to have practical applications.
But the return on investment for basic research really has decreased. That's why big corporate research labs have disappeared in the US. AT&T and IBM used to do basic research in physics, and out of that came the transistor and the high density disk drive. Few companies do that today. Even pharma research is very product-focused.
An "accidental bug" which enables not only the microphone (even when it's supposed to be turned off) but text to speech conversion? No way.
If anyone can find an honest prosecutor, criminal prosecution is in order.
You're going to spend way too much time trying to get Samba to do all the funny stuff that an ancient farm of XPs has set up to share. You're going to have to do this in stages, which means replicating the exact sharing structure of the old machines. The users won't be able to do this themselves. ("After editing /etc/samba/smb.conf, restart Samba for the changes to take effect." - Ubuntu documentation)
Then you have to get everybody converted from Microsoft Office to OpenOffice/LibreOffice. They're not that compatible. Some documents and spreadsheets will be broken. Templates won't transfer very well. Everyone's workflow will be disrupted. The overhead of doing this for a small shop will be higher than the savings.
The small office environment is where the Microsoft environment does best. Upgrade to Windows 7, one machine at a time. Windows 7 is a good OS. (The solid Microsoft OSs were NT 3.51, Windows 2000, and Windows 7.)
This is a technololgy that's almost there. Like laser weapons. Big chemical lasers that could shoot down shells or small missiles were built two decades ago, but they were building-sized installations, used huge amounts of hazardous chemicals, and took a long time to cool between shots. A decade ago, the THEL laser system had that down to three semi-trailers, but it still used big tanks of hazardous chemicals. Recently, big arrays of solid-state lasers have been used to shoot down shells and small missiles, and that system fits on a medium sized truck. The current version is only 10KW, and the consensus is that about 50KW to 100KW is needed to be really effective.
What's "mobile" about it? Runs on batteries? Plays crappy cellphone games? No. It's cordless.
That's good, but it has nothing to do with mobile phones. Even GameFace uses the term "cordless", not "mobile".
The site is kind of vague on what processing takes place in the headgear, and what takes place on the external WiFi connected device where, presumably, the game is playing. This thing is only worth the trouble if the game behind it is rendering very fast and has very high resolution content, and the latency to the game is very low.
Here's the argument for non-obviousness from the patent:
One problem associated with using touch screens on portable devices is the unintentional activation or deactivation of functions due to unintentional contact with the touch screen. Thus, portable devices, touch screens on such devices, and/or applications running on such devices may be locked upon satisfaction of predefined lock conditions, such as upon entering an active call, after a predetermined time of idleness has elapsed, or upon manual locking by a user.
Devices with touch screens and/or applications running on such devices may be unlocked by any of several well-known unlocking procedures, such as pressing a predefined set of buttons (simultaneously or sequentially) or entering a code or password. These unlock procedures, however, have drawbacks. The button combinations may be hard to perform. Creating, memorizing, and recalling passwords, codes, and the like can be quite burdensome. These drawbacks may reduce the ease of use of the unlocking process and, as a consequence, the ease of use of the device in general. Accordingly, there is a need for more efficient, user-friendly procedures for unlocking such devices, touch screens, and/or applications.
Apple was trying to come up with a way to prevent butt-dialing and other unwanted device actions. The point of using a sliding motion is that it's unlikely to happen via random touches, but is reasonably intutive.
Microsoft's video doesn't really show a slider. It shows touch buttons that look visually like sliders. But you can trip them just by touching in the active area for the desired state. This is shown in the video where the demonstrator runs their finger down a column of switches and they all switch. Apple requires an explicit "click and drag" operation to unlock.
It may seem trivial, but if nobody did it before the patent, and everybody wanted to do it after the patent, it's a valid invention. "Obvious" does not mean "obvious in hindsight".
Batteries for grid storage have different properties than batteries for cars.
So grid storage tends to use different battery technologies than vehicles.
This is Yet Another Crap Extruder based printer. That whole class of machines sort of works on good days. None of them Just Work.
The fundamental problem is that they're welding a hot thing to a cold thing. That sucks for metal welding, it sucks for soldering, and it sucks for plastic welding. It's how you get bad welds, cold solder joints, and fractures in 3D printing. The heated build plate systems usually start a build OK, but a few cm from the build plate, that heat source isn't close enough to help much. So many taller builds fail around 2-5 cm.
For this process to work, it needs better temperature control. A heated build chamber (that's patented). A hot air jet or small laser aimed at the target just before the weld (larger plastic welders do this). But nobody seems to be doing that. They just keep coming up with variations in the 3-axis motion mechanism (not hard to get right) and the software (not really the problem). Or they add DRM and overcharge for "print cartridges".
Here's the programming manual for FIRST robotics competitions. This is not dumbed down.
It turns out that large pad + IV + crypto-secure hash...
Did this guy just reinvent a book cypher?
It's just 70 gallons of crude oil left in an unused pipeline. No fire. No explosion. Just a mess.
It's not like a few years ago, when a high pressure gas pipeline exploded in Daly City and took out a small subdivision. Now that was a serious problem and an indication of a worse one. The column of smoke was visible 20 miles away. The state of California made PG&E do hydrostatic testing on all their major gas pipelines, over PG&E's claims that it was unnecessary. During hydrostatic testing with water, there were two pipeline bursts. One caused a landslide that blocked parts of I-280 at Woodside CA. No fire, of course; just water and mud, since this happened during testing.
OK, first bypass the click troll and get to the actual paper.
The general idea seems to be to transmit a large amount of noisy data per plaintext bit. Historically, crypto schemes which make the input much bigger are disfavored, but communications bandwidth is cheaper now and that might be OK.
The author of the paper seems to have fallen into the old trap of thinking that that analog signals have infinite amounts of data in them. He writes things like ''The encrypting key space is unbounded." and "The choice of the form of coupling functions comes from a set of functions that is not bounded." ("High-end" audio people also fall for this.) In reality, at some point you hit a noise threshold, and, anyway, down at the bottom, electrons and photons are discrite. Also, to be usable, whatever is used for the key has to be of finite size, and preferably not too big.
"No new cypher is worth looking at unless it comes from someone who has already broken a very hard one. - Friedman.
That looks like library code that the compiler generated. Maybe some kind of strcpy variant.
Read the analysis of the code. It's not. It is, however, decompiled assembly code; the people doing the analysis don't have access to the source.
What you have to keep in mind is that this software was written for Windows 98.
Irrelevant. This code is vulnerable on any OS that lets it get UDP packets.
We don't have enough context to know.
Read the actual vulnerability report. There's enough context there.
The real morons are the ones who tried to network it, presumably against Yokogawa's recommendations.
You have to assume today that if it has an Internet-accessable interface, an attacker will find a way to get to it from the public Internet. Because, in practice, attackers do.
Google already has several TV interface devices on the market. There's Google Chromecast, of course. Google also sells a set top-box used with Google Fiber That also comes with the Google Storage Box, which is a 2TB file server for storing downloaded content. There's the old Google TV, which is mostly Android software inside.
So Google has this covered already. They have a device for viewing TV over the Internet, and they have a cable box for their cable system. They're probably going to tweak the UI on one of those and promote that as a new product.
Neither Google nor Facebook has ever successfully built a product users will actually pay for. (Google's Nexus phones are rebranded LG, Samsung, and Asus products). For both, all significant revenue is from ads. Yet both have now acquired hardware companies. Now they have to make a business out of them. They may not succeed.
Google acquired Motorola and had no idea what to do with it. Now they're selling it. Google has an automatic driving R&D project, but they acquired DARPA Grand Challenge technology and seem no closer to deployment than a few years ago. Google acquired a half dozen advanced robotics R&D firms, but none of those have commercial products or profits yet. Google now has to build an entire industrial business in robotics, which is slow, hard, and will take years to pay off. Google hasn't shown the corporate patience for that. Google products that didn't take off quickly are usually killed. I'm worried that Google will end up trashing the US robotics industry once they realize it's not a Make Money Fast business.
Facebook hasn't really tried yet in hardware. But they have no expertise at it. The Oculus Rift is still a prototype/low volume device. Facebook has never run a factory. They'll have to outsource manufacturing, which means everybody else will be making goggles if it turns out to be profitable to do so.
What's a good no-nonsense registrar for major TLDs? It doesn't have to be super cheap. I want to dump Network Solutions because they gave me an unsolicited domain (I had .com and .net; they gave me a useless .info) which they then expect me to pay to renew.
I have about five domains. I want to avoid the "bulk" domain companies like GoDaddy.
I'm afraid, however, that the Real Men Don't Need Bound Checks mentality that is prevalent among C programmers will be a big obstacle.
I've run into that. Usually from second-rate programmers. Programmers who think that way should be put them on maintenance programming for a while. Have them debug program crashes in code written by others.
The code:
The company that let that code out the door should be sued for gross negligence, and managers fired. That's not the only example; they failed to do basic checks at least three times. This isn't a subtle bug. This is failing C Programming 101.
(Several times, I've tried to convince the C standards committee to put a "strict mode" in the language and move towards a form of C that's resistant to buffer overflow problems. Maybe I should try again.)
C - now with over thirty years of buffer overflows.
Google has re-invented LittleBits, a family of electronics modules which are attached with magnets. With their new "Cloud" module and their Arduino module, you might even be able to build a wireless VOIP phone.
This is a fun hobbyist concept, but you don't actually use things built that way. Either this will be bulky or the components will be fragile. You pay a penalty for all that casing and standard form factor.
Somebody (Wyse, I think) built a PC like this in the early 80s. Each module looked like a book, and plugged into the module next to it. You lined up all the modules, pushed them together, and put a big pin with a knob through the stack to lock them all together. Total failure as a product.
What Google should be doing, after buying all those robotics companies, is designing a phone for 100% robotic assembly, so they don't need Foxconn. (Except that Motorola did that a decade ago.)
I think a fire alarm is an instance where I'd like something to have as simple and foolproof a mechanism as possible.
Yes. That's why fire sprinklers are so successful. There's nothing between the water and the fire except a low-melting-point component in the sprinkler head.
This is an example of webcrap-level programmers doing things they're not qualified to do. I'm beginning to think that "Internet of Things" programmers should be required to have Registered Professional Engineer credentials, like structural engineers.
Bitcoin is about as important as the Segway. Bitcoin works, and it does what it's supposed to do - allow one-way irrevocable value transfers between anonymous parties. That's a poor basis for practical commerce, although a boon to a broad range of crooks. It's especially bad that over half of Bitcoin exchanges have gone bust, despite the fact that they shouldn't be at financial risk.
It looks like Bitcoin will be around for a while, but it's not growing. The block chain transaction rate now is about what it was a year ago. Like the Segway, Bitcoin is kind of cool, and has niche applications, but it's not changing the world.
Most low-level employees today are depressed. And underpaid. In a poll, 81% of fast-food employees report having wages stolen from them in the form of unpaid time.
So what does he want CS students to learn? Embedded system programming? Control theory? Labview? System safety engineering? Mechatronics? Robotics?