You don't want your long-running processes to have root privileges. It's a massive security hole. Many of the Linux daemons for server use run on less than root privileges (Apache, MySQL, etc).
A better approach would have been to have a group that had the ability to make processes run after logout. That would be a security improvement, since you could then determine which users had the rights to have persistent processes.
This is change overturns about 40 years of Linux/Unix computer history. The concept of nohup is used everywhere in Linux server land, and breaking that programming idiom will have significant ramifications.
The issue with Edison and Galileo is that they tend to be crippled platforms. It's been a while since I last looked at them, but I always ask the same questions. Does it have Ethernet? WiFi? 5V/3.3V serial port? Bunch of pins for miscellaneous I/O? SATA port or SD card? Enough RAM to run a standard version of Linux? HDMI and keyboard ports for testing?
The latest Raspberry Pi has most of that. However, there always seems to be a catch with Intel products. Intel doesn't want the Edison or Galileo to compete in the PC market, and as a result, they are crippled platforms.
Birds fly towards your car because it boosts lift and windspeed. The car generates a huge aerodynamic wake, both in front and behind the vehicle. The birds are taking advantage of this fast moving air. It is a clever piece of aerodynamics. However, the margin of failure on the manoeuvre is frighteningly small.
We're expecting these CS graduates to go fourth and do something,...
Historically, universities were about perpetuating knowledge and the advancement of knowledge. Apprenticeships and professional programs are where people learn do to something practical. Universities were the hallowed halls of pure learning.
In the beginning, no one foresaw that a pure math specialization would have huge practical use. Some of the greats in computer science never thought their work would ever see use outside of the math department. If memory serves, Bool was extremely pleased that Boolean Algebra existed as a theoretical exercise that would never have a practical application.
However, it was a horribly defined standard. They defined voltage levels and pins, but nothing about the data packet. Everyone settled on an old-style TTY packet, and then experimented with parity, stop bits, and baud rate. Hardware flow control wasn't properly defined either. Intel made an 8251 chip that would chop-off messages mid-byte if the hardware flow-control lines were used.
The American's and the British had it in WWII on the bombers. Before that, I'm sure someone can find an example of a planes from the 20's, and maybe even from WWI that had an omni-directional gun turret.
A big issue in the early fighters was that they couldn't figure out how to fire forward. The ammunition was unreliable, and if you tried to fire through the propeller blades, it would shoot off the propellers. In the early years, this was solved by adding "wedges" to the props such that if the ammo hit the prop, it wouldn't destroy the propeller.
The early aircraft would have also investigated alternative designs, including omni-directional gun turrets, and sending a guy up in the air with a rifle or sub-machine gun.
Replacing X is a big project. Sometimes it takes a while to generate something good.
I'm not sure anyone has a good model for handling rewrites of massive projects. The experiences of KDE 4.0 and Gnome 3.0 come to mind. Eventually, they were better, but it takes some time with a massive upgrade like that.
The other issue is that User's often have a very good idea of what they don't like. However, bulimic criticism does not help to refine a software product. It just splits the ecosystem. Ultimately the user's need to use their computer, and the new software just isn't ready. So the developer's and user's go in different directions.
Closed source isn't the solution either: with Windows 8, Microsoft split it's ecosystem. Windows 10 hasn't fixed the split (yet).
I'm pretty sure the Pi's Ethernet network interface is stable. I have been managing a fleet of them.
There is a whole list of other potential problems. Did you remember to opto-isolate your I/O? and place RC snubbers or free-wheeling diodes on the loads? I have seen many an embedded project go wrong with inductive kick-back problems.
I did notice the operating system tends to freeze when accessing the SD card. I wasn't clear if that was because the SD card layer was doing a lock on the operating system, or it was just a lock on my application, which was waiting for I/O. For one generation, we added an automatic reboot cron-job to reboot the Pi's periodically. This was to overcome some issues with USB WiFi communications. I don't think this is needed with the latest Raspberry Pi's.
On the whole, the Raspberry Pi has been one of the best embedded platforms that I have worked with. I can give pages of creative bugs/issues with some of the other industrial micro-controller products. If the Raspberry Pi had a fast disk (SATA) interface and built-in WiFi, I would be very happy.
When waves interact they generate an interference pattern. The interference pattern will have larger up and downs than the original waves. However, it might not be obvious to a ship that they are in such an interference pattern, as locally the sea looks like waves. Depending on the path of the boat, shortly before the peak of the interference pattern, you may see an unusually quiet region. I could see how it would get a mariner confused.
What I never understood was if rogue waves were caused by interference patterns, or something completely different like under-ocean mudslides. If this research works out, maybe we will know the answer.
I suspect that the NSA and the FBI already have the dirt on the field of candidates. Whichever one is going to win, they will be pointed in the correct direction and be on the side of the NSA and FBI at the end of the campaign.
The cynic in me suspects the ghost of J. Edgar Hoover is still alive and well in Washington. I'm not sure the CIA dirty tricks squads died with Watergate either. With all the information the NSA collects, the dirty tricks squads could be alive and well...
I'm not clear on this either. Why would someone want to build a large-area device, like a super-capacitor, on top of an IC? given that the cost per unit surface area of a modern IC is astronomical?
The question is particularly complex, as we are talking about super-capacitors. Super-capacitors usually have terrible AC characteristics. If we were talking about high-performance ceramic capacitors, then the answer would be to improve the AC performance of the power distribution net, or for RF communications.
Alternatively, the answer could be to power a portable device. However, a typical capacitor has such a large surface area that it is a roll (or a stack) of many layers. To power the portable device, the surface area of the chip would have to be very large relative to its power consumption. This would be an unusual combination, as power consumption is often a function of die surface-area.
In concept, any device will have an application. However, I'm not clear on what the application for this device would be...
The devil is really in the details with the x86 series. Intel has claimed whatever suited them over the years. As such, the 8088 was a 16-bit processor, even though it had an 8-bit data bus. The 386 was a 32-bit processor, even though the 386SX chip had a 16-bit data bus. The Pentium processor was 64-bit processor, because it had a 64-bit data bus, however the first generation Pentium processors only had a 32-bit ALU.
Modern processors almost defy description by data bus width. There are so many DRAM and cache-line width tricks that modern processors defy description based on simple bus-width numbers. For instance, how do you describe dual or quad channel memory controller, that can independently fetch different pages on different data buses? and loads memory 4-bits at a time to fill a cache line? and works with or without ECC?
Since the advent of RISC processors, in general, a 64-bit processor means a processor with a 64-bit ALU. This keeps things simple.
My expectation is that LENR will be used for heat production. If you want gigawatt-scale electricity production, then the options are fission reactors or high-temperature fusion. The problem with using a low temperature process for thermal electric power generation is that thermal cycles perform better at higher temperature differentials. In terms of using LENR for electricity generation, this means that the practical issues of turning the heat back into electricity wipe out the benefits of using LENR to make surplus energy.
If the goal of the reactor is simply to make heat, then LENR is a much more feasible option. It is likely that LENR systems can generate enough heat to be a useful electric-heating device. Detailed cost-benefit analysis will depend on the application. There are enough electric-heating applications in existence that LENR almost certainly will find a niche market (and possibly a main-stream market.)
The Canadian multi-lingual keyboard is a disaster. It changes the shape of some of the most important keys on the keyboard, like ENTER, SHIFT and Backslash. Any touch typist familiar with a standard US keyboard, can't type on the multi-lingual keyboard. However, to support French, about 1/2 of the keyboards in the marketplace are Canadian multi-lingual. Nobody can type on anything, unless they get the right kind of keyboard,, or type slowly.
All this would be somewhat great if Canadian multi-lingual keyboard was multi-lingual. However, it isn't. I don't even think it has all the French letters either. Once you get into other languages, even romantic languages, the keyboard doesn't have the right keys.
The Canadian multi-lingual keyboard is the classic Canadian compromise. It makes no one happy, and no one completely upset.
An approximation used in telecommunications is that the required bandwidth for a communications signal is at least the baudrate. As such, 5.3 Gb/s throughput router would need a little over 5.3 GHz of bandwidth to make the physical layer work. Pretty sure that's not happening with a 5 GHz router...
If you want to run a protocol layer on top of that, given a shared communications medium and randomly located stations, much more bandwidth is required.
The first few episodes of Doctor Who were so bad they almost got the series cancelled. Don't start a new Doctor Who fan watching anything before the first Dalek series on Skaro. That series was what made Doctor Who the second biggest British TV franchise ever.
For a modern audience, I would recommend starting at the reboot with the new series and Russell T Davies and Christopher Ecceleston. That was the point at which Doctor Who became interesting again for a new generation.
People make mistakes. A huge percentage of accidents occur to drivers aged 25-65 (60%) on a clear day (74%). Driverless cars will make an impact by going after the dominant sources of accidents.
Personally, I suspect that partially interactive and/or assistive technologies will be deployed first. That way the care doesn't have to handle every case properly, particularly winter and icy conditions.
If the job was only about securing data, then security professional's would recommend destroying the data. The military has been known to do exactly this. Destroying the data creates the ultimate security.
What makes security people into security professionals, is that the professionals can design systems that allow authorized activities happen smoothly while simultaneously keeping out the bad guys. That is a much harder task than simply securing the data against unauthorized access. It requires the professional to focus on the balance between usability, security and profit.
Slashdot Mirror
Python's not as bad as it could be. I think it's a fine language for other people to program in.
That's faint praise. People said similar things about COBOL. You could always read what someone else wrote ...
You don't want your long-running processes to have root privileges. It's a massive security hole. Many of the Linux daemons for server use run on less than root privileges (Apache, MySQL, etc).
A better approach would have been to have a group that had the ability to make processes run after logout. That would be a security improvement, since you could then determine which users had the rights to have persistent processes.
This is change overturns about 40 years of Linux/Unix computer history. The concept of nohup is used everywhere in Linux server land, and breaking that programming idiom will have significant ramifications.
The issue with Edison and Galileo is that they tend to be crippled platforms. It's been a while since I last looked at them, but I always ask the same questions. Does it have Ethernet? WiFi? 5V/3.3V serial port? Bunch of pins for miscellaneous I/O? SATA port or SD card? Enough RAM to run a standard version of Linux? HDMI and keyboard ports for testing?
The latest Raspberry Pi has most of that. However, there always seems to be a catch with Intel products. Intel doesn't want the Edison or Galileo to compete in the PC market, and as a result, they are crippled platforms.
Birds fly towards your car because it boosts lift and windspeed. The car generates a huge aerodynamic wake, both in front and behind the vehicle. The birds are taking advantage of this fast moving air. It is a clever piece of aerodynamics. However, the margin of failure on the manoeuvre is frighteningly small.
We're expecting these CS graduates to go fourth and do something, ...
Historically, universities were about perpetuating knowledge and the advancement of knowledge. Apprenticeships and professional programs are where people learn do to something practical. Universities were the hallowed halls of pure learning.
In the beginning, no one foresaw that a pure math specialization would have huge practical use. Some of the greats in computer science never thought their work would ever see use outside of the math department. If memory serves, Bool was extremely pleased that Boolean Algebra existed as a theoretical exercise that would never have a practical application.
Typically spacecraft software is hand verified for its correctness -- the main reason it is so expensive.
On a space program, the teams are highly experienced at verifying software.
RS-232 has done quite well.
However, it was a horribly defined standard. They defined voltage levels and pins, but nothing about the data packet. Everyone settled on an old-style TTY packet, and then experimented with parity, stop bits, and baud rate. Hardware flow control wasn't properly defined either. Intel made an 8251 chip that would chop-off messages mid-byte if the hardware flow-control lines were used.
The only way a gun protects you is if the assassin misses the first time.
And, in the midst of the confusion, you correctly identify, shoot and hit the assassin before he tries again.
Of course, shooting and hitting the assassin assumes your marksmanship with a handgun is better than his, when he might have a rifle.
The American's and the British had it in WWII on the bombers. Before that, I'm sure someone can find an example of a planes from the 20's, and maybe even from WWI that had an omni-directional gun turret.
A big issue in the early fighters was that they couldn't figure out how to fire forward. The ammunition was unreliable, and if you tried to fire through the propeller blades, it would shoot off the propellers. In the early years, this was solved by adding "wedges" to the props such that if the ammo hit the prop, it wouldn't destroy the propeller.
The early aircraft would have also investigated alternative designs, including omni-directional gun turrets, and sending a guy up in the air with a rifle or sub-machine gun.
Replacing X is a big project. Sometimes it takes a while to generate something good.
I'm not sure anyone has a good model for handling rewrites of massive projects. The experiences of KDE 4.0 and Gnome 3.0 come to mind. Eventually, they were better, but it takes some time with a massive upgrade like that.
The other issue is that User's often have a very good idea of what they don't like. However, bulimic criticism does not help to refine a software product. It just splits the ecosystem. Ultimately the user's need to use their computer, and the new software just isn't ready. So the developer's and user's go in different directions.
Closed source isn't the solution either: with Windows 8, Microsoft split it's ecosystem. Windows 10 hasn't fixed the split (yet).
I'm pretty sure the Pi's Ethernet network interface is stable. I have been managing a fleet of them.
There is a whole list of other potential problems. Did you remember to opto-isolate your I/O? and place RC snubbers or free-wheeling diodes on the loads? I have seen many an embedded project go wrong with inductive kick-back problems.
I did notice the operating system tends to freeze when accessing the SD card. I wasn't clear if that was because the SD card layer was doing a lock on the operating system, or it was just a lock on my application, which was waiting for I/O. For one generation, we added an automatic reboot cron-job to reboot the Pi's periodically. This was to overcome some issues with USB WiFi communications. I don't think this is needed with the latest Raspberry Pi's.
On the whole, the Raspberry Pi has been one of the best embedded platforms that I have worked with. I can give pages of creative bugs/issues with some of the other industrial micro-controller products. If the Raspberry Pi had a fast disk (SATA) interface and built-in WiFi, I would be very happy.
When waves interact they generate an interference pattern. The interference pattern will have larger up and downs than the original waves. However, it might not be obvious to a ship that they are in such an interference pattern, as locally the sea looks like waves. Depending on the path of the boat, shortly before the peak of the interference pattern, you may see an unusually quiet region. I could see how it would get a mariner confused.
What I never understood was if rogue waves were caused by interference patterns, or something completely different like under-ocean mudslides. If this research works out, maybe we will know the answer.
Yes, especially after watching Doctor Who. :-)
I suspect that the NSA and the FBI already have the dirt on the field of candidates. Whichever one is going to win, they will be pointed in the correct direction and be on the side of the NSA and FBI at the end of the campaign.
The cynic in me suspects the ghost of J. Edgar Hoover is still alive and well in Washington. I'm not sure the CIA dirty tricks squads died with Watergate either. With all the information the NSA collects, the dirty tricks squads could be alive and well ...
I'm not clear on this either. Why would someone want to build a large-area device, like a super-capacitor, on top of an IC? given that the cost per unit surface area of a modern IC is astronomical?
The question is particularly complex, as we are talking about super-capacitors. Super-capacitors usually have terrible AC characteristics. If we were talking about high-performance ceramic capacitors, then the answer would be to improve the AC performance of the power distribution net, or for RF communications.
Alternatively, the answer could be to power a portable device. However, a typical capacitor has such a large surface area that it is a roll (or a stack) of many layers. To power the portable device, the surface area of the chip would have to be very large relative to its power consumption. This would be an unusual combination, as power consumption is often a function of die surface-area.
In concept, any device will have an application. However, I'm not clear on what the application for this device would be ...
The devil is really in the details with the x86 series. Intel has claimed whatever suited them over the years. As such, the 8088 was a 16-bit processor, even though it had an 8-bit data bus. The 386 was a 32-bit processor, even though the 386SX chip had a 16-bit data bus. The Pentium processor was 64-bit processor, because it had a 64-bit data bus, however the first generation Pentium processors only had a 32-bit ALU.
Modern processors almost defy description by data bus width. There are so many DRAM and cache-line width tricks that modern processors defy description based on simple bus-width numbers. For instance, how do you describe dual or quad channel memory controller, that can independently fetch different pages on different data buses? and loads memory 4-bits at a time to fill a cache line? and works with or without ECC?
Since the advent of RISC processors, in general, a 64-bit processor means a processor with a 64-bit ALU. This keeps things simple.
The groklaw coverage was so good. I know that PJ closed the site down. Did anything ever spring up in its place?
My expectation is that LENR will be used for heat production. If you want gigawatt-scale electricity production, then the options are fission reactors or high-temperature fusion. The problem with using a low temperature process for thermal electric power generation is that thermal cycles perform better at higher temperature differentials. In terms of using LENR for electricity generation, this means that the practical issues of turning the heat back into electricity wipe out the benefits of using LENR to make surplus energy.
If the goal of the reactor is simply to make heat, then LENR is a much more feasible option. It is likely that LENR systems can generate enough heat to be a useful electric-heating device. Detailed cost-benefit analysis will depend on the application. There are enough electric-heating applications in existence that LENR almost certainly will find a niche market (and possibly a main-stream market.)
The Canadian multi-lingual keyboard is a disaster. It changes the shape of some of the most important keys on the keyboard, like ENTER, SHIFT and Backslash. Any touch typist familiar with a standard US keyboard, can't type on the multi-lingual keyboard. However, to support French, about 1/2 of the keyboards in the marketplace are Canadian multi-lingual. Nobody can type on anything, unless they get the right kind of keyboard,, or type slowly.
All this would be somewhat great if Canadian multi-lingual keyboard was multi-lingual. However, it isn't. I don't even think it has all the French letters either. Once you get into other languages, even romantic languages, the keyboard doesn't have the right keys.
The Canadian multi-lingual keyboard is the classic Canadian compromise. It makes no one happy, and no one completely upset.
An approximation used in telecommunications is that the required bandwidth for a communications signal is at least the baudrate. As such, 5.3 Gb/s throughput router would need a little over 5.3 GHz of bandwidth to make the physical layer work. Pretty sure that's not happening with a 5 GHz router ...
If you want to run a protocol layer on top of that, given a shared communications medium and randomly located stations, much more bandwidth is required.
For the readers not in the loop, please ...
The first few episodes of Doctor Who were so bad they almost got the series cancelled. Don't start a new Doctor Who fan watching anything before the first Dalek series on Skaro. That series was what made Doctor Who the second biggest British TV franchise ever.
For a modern audience, I would recommend starting at the reboot with the new series and Russell T Davies and Christopher Ecceleston. That was the point at which Doctor Who became interesting again for a new generation.
People make mistakes. A huge percentage of accidents occur to drivers aged 25-65 (60%) on a clear day (74%). Driverless cars will make an impact by going after the dominant sources of accidents.
Personally, I suspect that partially interactive and/or assistive technologies will be deployed first. That way the care doesn't have to handle every case properly, particularly winter and icy conditions.
With Darth Vader, why would you need much of a fighter screen?
Only a young Jedi using the force could successfully mount an attack against a Death Star ...
If the job was only about securing data, then security professional's would recommend destroying the data. The military has been known to do exactly this. Destroying the data creates the ultimate security.
What makes security people into security professionals, is that the professionals can design systems that allow authorized activities happen smoothly while simultaneously keeping out the bad guys. That is a much harder task than simply securing the data against unauthorized access. It requires the professional to focus on the balance between usability, security and profit.