Centralized policies such as a Windows Domain is much easier to manage than a hodgepodge of various desktops with no way to enforce policy.
The problem with a Windows desktops is that even with GPO, and after enforcing Group Policies, you still don't have any way of enforcing policies. Think about if for a while. You can get around anything under Windows, and often it isn't necessary to try very hard.
At the other extreme, try installing a real multi-user operating system in a real multi-user environment. They are totally different paradigms. With Group Policies, you are trying to use a user mode shell with system access to enforce policies. Under Linux, policies can be enforced through the operating system, and the file system it implements. The O/S with the low-level implemented security will always be more secure.
The problem the poster is having, is that you fundamentally can't compare Group Policies with the full security infrastructure of a multi-user operating system. They are two different things. They don't work the same way, and the difference can't be papered over with a cool shell interface.
Unfortunately, what you say seems to mirror my experience. You can trivially lock down a linux system so hard that no user will ever run something without permission. Under Windows, you can configure Group Policies all day long, and there are still a billion security holes.
One of my first attempts was locking down a Windows 95 machine. That was cracked in less than a couple of hours. After that, we tried Windows NT 4.0. With NT 4.0, we had to enable a feature that limited which executables would run on the machine, and after much testing, we made a list of 11 authorized executables and that was it. Unfortunately, 9 of the 11 were undocumented Windows internal programs. The newer versions of Windows got rid of the authorized executable list, created work-arounds (like ActiveX and RunDLL), and generally did other things that make it almost impossible to truly lock down a Windows machine. Windows XP Embedded stations have being taken over by viruses, even though the XP Embedded HD is mounted read only. It is truly difficult to configure a Windows machine to work on the shop floor, and to do it in such a way that the bored shop floor people can not mess anything up.
With Linux, you can mount an entire hard drive as read-only, and NFS mount the home directories. This makes it really tough to create permanent modifications to the system. If you want to be less draconian, one can easily restrict which executables the users have access to. Ubuntu even has switches that restrict the user's ability to mount new disk volumes to access floppy drives and USB keys. If you want to lock a computer down hard, it is much easier to accomplish the goal under Linux.
A properly configured Linux computer running a machine in real-time on the shop floor is impressive. If anything really goes wrong, just cycle the power. In practice, almost nothing does go wrong, and uptimes on the order of 6 months are achievable. Contrast this to Windows XP, which needs nightly reboots for virus updates, the machine starts stuttering when someone does a remote backup, and just about any user can find a way to entertain themselves by installing new desktop wallpapers...
The material in the design specification was essentially unobtanium. It couldn't be manufactured at all. Quietly, the manufacturing engineers developed a solution that almost met all of the design specifications, and this was an excellent compromise. Unfortunately, the design engineers couldn't be convinced to sign off on the design change because of quality procedure 15, and military qualification 7. However, the biggest reason the design engineers wouldn't sign off on the change was because of a supposedly critical but practically useless mandatory project requirement, like the missile must work when fired in -40 degree water from 20 feet under the polar ice shelf.
The manufacturing engineers decided that the "fire nuclear missile while under ice shelf function", probably wouldn't be used, so the modified material was actually just fine. They shipped the missiles, got paid, and everyone was happy. Until now, when someone tries to "fix" the original "fix".
This story has happened before and will happen again. Whenever you bump into a design that requires a part that "does not exist", watch out for the possibility that the part never did "exist". It could be that you are reading a "design" document, and not what manufacturing actually built. I've worked in manufacturing, and there are lots of stories about impossible to make designs that somehow got shipped.
Microsoft has said that Open Source is communist and Anti-American!
Yes, but after hundreds of billions of dollars in development expenses, this is the most expensive open source software in the world.
The world financial meltdown must count as the world's most expensive software bug ever. If this software had been widely accessible and better understood sooner, the macro-economic consequences of the activities relating to this software might have been better understood much sooner. The crisis might have been avoidable.
The world financial meltdown is a consequence of "security through obscurity," or in simpler terms: "What you don't know, won't hurt you."
I think you are on the right track. They have built something not all that different from an induction motor. Essentially, there are two different types of electrodes. The first pair is in the solution, and a current is going to be established. Given this is water, the current will be established by lining up positively charged hydrogen sides of the atoms with the negative electrode (the electron source). The negatively charged oxygen side of the water molecule will line up with the positive electrode (electron sink). A small current will flow, forcing the water to move. Initially, this motion won't be circular, to get circular motion you need a second set of electrodes.
The second set of electrodes develops a electro-static field. This set of electrodes isn't actually in contact with the water. The current carrying water molecules will try to align with this second set of electrodes, while carrying current. Of course, the fields are in two different directions, so static alignment isn't possible. As such, the motor starts to spin.
There are probably a bunch of other effects happening. Motors involving electro-statics tend to be very friction sensitive and fickle beasts. However, if you have a current, a dipole, and either a magnetic or an electro-static field, you will get torque. It is just in this case, the dipole is the water molecule, and as such the torque causes the water to spin.
For the poster wondering why the water at the center spins faster, the reason is likely due to the available torque and energy. Chances are the torque is proportional to the field strength (which is constant.) The water at the center has less distance to travel, and less drag. Thus it can spin faster. It would be quite unexpected for a liquid to behave like a solid, and travel at the same rotational rate at all diameters.
Finally, some posters have wondered how the motor is being powered. The power to spin the liquid is coming from the current flow between the two electrodes in the solution, and the corresponding voltage drop.
There's a school of thought that says the pirated music encourages more people to buy through album sales based on 'previews'. And yet the RIAA claim this sort of piracy decreases sales.
Advertising based on word of mouth is fickle for advertisers. If you have something good, then it works better than anything else. People trust friends. If your product sucks, then people still trust their friends, and won't touch your product.
Internet P2P programs like BitTorrent amplify this effect. Now, you can listen to something yourself, and figure out for yourself how much you like it. Thus P2P results in a dramatic decrease in control for advertisers. It is even more fickle than word of mouth.
If you have a poor product, but from a band with a good reputation, then you want to blitz market the product. Let no one listen to it in advance. Have it show up at stores in massive quantities the day of launch, and sell as much as you can on the first day. This way you can scam as many people as possible for first day sales. With some luck, this first day blitz will cover your costs, and everything will turn out OK. The movie industry specializes in this tactic.
P2P threatens to completely destabilize this advertising tactic. The record companies, which are really big advertisers, will not be happy about this loss of control. Even if P2P ultimately makes them more money, the record companies still won't be happy about the loss of control.
This could be fun. Here are some more suggestions:
- Welder - The little chips don't last long against a good arc welder.
- 600 VAC - Why stop at a wall outlet?
- Tesla Coil - 200 kV is better than 600 VAC
- Lightening Rod. Why stop at 200 kV?
- Oxy-acetylene Torch - higher temperatures
- Plasma Cutter - even higher temperatures
- NdYAG Laser - Etch your name into the remains of the flash chip.
- Chew Toy for Dog - Don't underestimate some of those canines, although USB keys might not be good for them.
- Log-Splitting Practice. How good are you at aiming that Axe?
- Place USB in Cement Footings of a building. Do the mob thing.
- Rock crusher
- Grinding Machine
- Wood chipper / pulper
- Cement kiln
- Blast Furnace
- Industrial Press - Terminator Style!
I'm pretty sure that some of these machines can destroy industrial quantities of USB keys, with little difficulty. Cement kilns and rock crushers can destroy just about anything. It would be interesting the see the resulting crushed rock in a piece of cement though. It would be colorful.
If I mount/home on a separate drive, (good to do when upgrading) the rest of the Linux file system fits nicely on a small SSD.
I would move/tmp to either a RAM disk or a hard drive. There is no point in having tmp files using up the lifespan of your SSD, especially after you just moved/home to extend its life. Also, you could move some of the stuff in/var to a hard drive or ramdisk. Good candidates might be/var/tmp and/var/log. Alternatively, you could just move the entire/var hierarchy to a hard drive.
FPGA's are really slow for some applications. This means a narrow window exists between an application that can be done on a micro-controller in software, and the equivalent application being done in an FPGA.
Many of my applications require long chains of counters and magnitude comparators. FPGA's seem to be particularly bad at implementing them. The estimates I use are:
A 50 MHz Schottky TTL counter can count at 25 MHz fairly easily.
A 100 MHz FPGA can count at 2-4 MHz, and
A 25 MHz micro-controller can count at 100 kHz.
Certain classes of PAL like devices can count quickly, but may be limited in how many bits they can put onto a single chip. For any given application, I try to figure out how quickly the counter chain needs to count at, and select appropriate solutions based on that. Often a micro-controller is a preferred solution, as so many different things can be done with it. Additionally, many modern devices have dedicated hardware that can remove the need for the FPGA entirely.
Counters also seem to be a particularly nasty corner case for FPGA devices. They are cheaply available as stock parts, so an FPGA based solution can be very expensive. Also, modern micro-controllers with built-in counter hardware are becoming very inexpensive. Sometimes cheaper than TTL devices.
Counters also have a very critical timing chain where one bit (the carry bit), must make it through every stage of the device inside one clock cycle. In an FPGA, every logic operation, interconnect, and cross-connect carries a timing penalty. FPGA's are different from PALs. In a PAL, every output bit can be used in any combinatorial operation on any output inside one clock cycle. In FPGA's, combinatorial delays can stack up very quickly, and significantly.
Researchers simulating new computer hardware designs love to use FPGA's, because expensive FPGA's can be used to simulate complex pieces of dedicated silicon. The delays in FPGA's accumulate in a similar scaled fashion as to what would happen in dedicated silicon. The resulting FPGA based solutions though, are roughly at one to two an orders of magnitudes worse than dedicated silicon, on a speed basis, a cost basis, and a speed/cost basis. This isn't really a problem if the simulation is only to verify functional correctness before one commits to "real" silicon, but it is a big problem if you are trying to build something with cheap hardware.
It's the one to two order of magnitude slower problem that always seems to mean that software solutions are always just a hair worse than FPGA based solutions. Every once in a while, a very FPGA friendly application does occur, and then the technology works well. It's just that it takes the "right" applications. Specifically, a register oriented application with a relatively small number of combinatorial inputs per register.
Well, I'm from Canada. I'm pretty sure everyone responsible for any large collection of computers is responsible for somehow "viewing or possessing child pornography." Seriously, you don't believe every single naked or sexual image of every girl on the internet is someone over 18?
- What if you have two teenagers on your network exchanging pictures of each other?
- How do you prevent any illegal activity from occurring on your network?
- How do you tell how old someone is? Can you tell the difference between a 16 year old and a 18 year old?
- Can you spot the illegal images in the millions of random GIF and JPEG images crashing around your network?
- Canadian law also covers written and drawn descriptions. How will you stop that?
- If the police suspect your company of having this stuff on your network, they will likely get a search warrant and seize all your computers (and backups.) How will your business deal with this?
Essentially, under Canadian law, almost everyone with a computer is guilty of possessing child pornography. If you have a run or own a big network of computers, you are guilty.
The scary thing is, in some countries the cops are starting to get really serious about this. In Britain, someone estimated the number of pedophiles at 60,000 to 70,000 people! That's 0.1% of the general population, and 0.2% of the men. Doesn't that number seem just a little bit too high?
If we just keep watering the definitions down, and expanding our terms of reference, then eventually we will be able to arrest everyone for something. It's always nice to start with the weird and strange people first. After all, those teen male computer hacker geeks, they always have naked images on their computers. At least one of them must have an image of a girl that is less than 18 years old!
Teenagers have already been charged in Canada under these laws. For the moment, computer professionals haven't been targeted by these laws, but it is only a matter of time. Once a computer professional is targeted, they will have a very hard time proving they are criminally innocent. In the mean time, their career, reputation and businesses will be completely destroyed. All it would take, is an accusation.
explain again how having... is supposed to give Microsoft an "edge" over Apple.
This is gorilla marketing. This story follows a five step plan:
1. Phone gets stolen from executive. Leak story to press.
2. The executive's teenage daughter is arrested by police. She became completely addicted to her father's new Zune phone. She comments: "The new Zune phone is amazingly cool." 3. Make billions of said phone.
4. ???? (Sell Zune Phones????)
5. Profit
At least, that is the story from the marketing department's internal emails...
On a more serious note, does anyone have any idea how Microsoft plans to generate significant sales of Microsoft phones? Being a late mover in a competitive marketplace with entrenched successful competition does not scream "Instant Sales" to me...
I'm not sure I agree with the person that moderated the parent as Flame Bait. Microsoft is a big enough target that it doesn't want to get sued over copyright violations. That was why Vista and Windows 7 have all that DRM crap. Now Microsoft wants to build a secure utility to transparently share files between people over the internet.
Has anyone ever built a secure file sharing utility over the internet that hasn't been abused in some way? Ever?
Sometimes it is just too easy to guess peoples passwords. People will share the potentially embarrassing items, whether it is an embarrassing picture, or a copyrighted song. Microsoft will log all this information. One enterprising teen could make all of your dirty laundry public knowledge.
You could try to nearly-hermeitcally seal the room
Activate the Halon fire suppression system and run...
Alternatively, simply adding a bunch of dry ice in an enclosed space below floor level will likely clear out any living organisms too. Be careful that you don't suffocate too.
Finally, there is always the old cyanide gas trick... No need to worry about suffocation with cyanide.
Safety Warning: Don't actually do any of the above.
Rats and mice are also different problems. If you have mice, cats are very effective. Mice will not even approach anywhere they think a cat lives. If you have rats, you will need a larger predator. At least a big cat, that you know will take out rats. Rats are much larger than mice.
I would consider lining everywhere there are cables with glue traps. That will catch anything that goes near the cables. Unfortunately, it could also be highly annoying. Line everywhere a cable enters or exits a small whole with steal wool. Mice are almost impossible to prevent entering a building, because they can move freely through such small entry points. They also seek out heat.
Finally, if the problem is rats, then it is much easier to block entry to the buildings. Rats are much larger than mice, so physical protection methods work better against rats. Be prepared to use concrete and steel solutions. Rats and squirrels can chew through wood. My experience is that rats will eat plastic much more readily than mice. Rats are much larger than mice, and are tougher to catch. Mouse traps are ineffective against rats. Consider sheathing your wiring in metal and/or concrete. Quick setting concrete is an easy way to plug oddly shaped holes. Metal conduit can be terminated with liquid tight fittings. Between the two solutions, you should be able to prevent mice and rats from either going through conduit, or going around conduit and exploiting holes in the building walls.
At least one hearing aid manufacturer deliberately does its DSP work in analog, to get the 1/30th the power with a minimal accuracy loss. The problem with digital filters (like DSPs) is that they use lots of power relative to analog filters. The fundamental design issue with what the author is suggesting, is that in digital the bits all have different weights. The difference between 2047+1=0 and 2047+1=2048 is huge, however in an ALU the difference is a single badly propagated carry bit.
Analog electronics doesn't suffer from the bit weighting problem. Thus a little bit off in analog is almost always just a little bit off. It makes a big difference. A scratched LP that is still playable, and a scratched DVD, with a much better sound quality, is unplayable. I deliberately picked DVDs for this example, as CDs have an exotic error-correction algorithm which makes them playable even if they are fairly damaged. As such, it isn't fair to compare a scratched CD to either a DVD or an LP. CDs are by far the least scratch sensitive media, but LPs which are analog, are more usable with scratches than most other digital media.
In any case, this is why some low power applications are implemented in analog and not digital.
Anyone hoping to pocket a percentage of $9,000,000 by giving a bunch of passwords to a bunch of people you don't know, and then assuming you won't get grassed out to the cops is likely making a major mistake.
If the criminal is smart, a better strategy might be to "give" the information away to the right group of people. This might give someone a smug sense of "revenge" against a former employer. Someone could short the stock in the stock market, or the theft could cover up some insider funny business. The initial criminal act may be different than what it appears.
Alternatively, the actual "inside" mastermind may actually be a victim too. Maybe someone conned an insider for information, or access to a laptop, and just sold the information. Maybe someone got hold of the backup tapes. This might actually a fairly low-value theft for the original criminal.
At one company I worked at, Excel was almost designated the company's standard word processor. You can write text in Excel. You can do tables and charts in Excel. You can do databases in Excel. Why use the other packages for anything else?
The scary thing is, the proposal kind of makes sense...
I once reprimanded a intern programming student for following the MSDN example code. There is something wrong with saying to someone: "I know you know that Microsoft example code does not work. You can tell by just looking at it."
There is something wrong about needing to teach young programmers to not follow the Microsoft documentation. Even after 10 years, Microsoft still hasn't went back and fixed the documentation either. They just created more API's with more documentation problems.
No good ever came from the Reply All button. It is like adding "Press this button to be fired" function to your corporate email system. You know someone is going to press the button, you know trouble will ensue, so why create the button?
To all the mods, please don't destroy all my Karma. I really do hate that Reply All button.
True, because they had to hunt and gather whereas we get our food from supermarkets.
I think most North Americans would be quite surprised at how fast civilization would fall apart if our supermarkets stopped magically "replentishing" themselves with food. The supply chain from the farm to the supermarket isn't very long, often only a few days, and usually less than one month.
The end of modern civilization is only a major global crisis away. We are already experiencing a global recession caused by a banking crisis that started in only one country.
Traditionally, you launch a technology like this into a sector willing to pay a heavy premium for the parts. If they were really able to produce capacitors with these specifications, one would think they would go after the ultra-exotic small volume applications first. These applications often provide huge amounts of funding, allowing you to debug your product and manufacturing techniques. Thus, initial commercialization is often to DARPA, or some other defense, military, electronic or specialized power conversion application. After you get your production processes down, then you go after the high-volume low-cost marketplace. Batteries are a fairly established, low cost market. Even if your product is good, it may be difficult to make enough product to meet demand.
For me, this patent begs the question: if you had capacitors this good, why are you selling them to compete against batteries?
I think your comment is bang on. Someone only has to generate software that fulfills the contract, and gets approved for use. It doesn't actually have to work correctly...
I think FORTRAN is going to stay the untouched leader in the fast and easy to program supercomputer / parallel mathematical simulations market. C is a very close second. Simply put, every supercomputer manufacturer makes sure its FORTRAN and C code is very quick, because that is what most of the customers use. I can't see Haskell competing in the supercomputer market / parallel processing market yet.
The next question is: Is Haskell going to be used in your next GUI based application that has a really processing intensive back end? Well functional programming is about finding a nice easy to program way to abstract and hopefully eliminate the serial aspects of your program. GUI code is about as serial as it gets. Users find it very confusing when programs pop up dialog boxes in parallel. This means a fundamental language-independent tension will exist in your program between the serial GUI code and the parallel back end code.
I don't think a good solution for the "easy parallel programming problem" is on the market yet. I would appreciate suggestions, because I am looking. I think Microsoft missed a big opportunity with.NET. Microsoft released a new programming architecture, but they omitted ground-up support for NUMA, supercomputing, and parallel processing architectures. The next big jump in computing will be the language that can smoothly harness various computing configurations ranging from single-core/single-computer to multi-core/multi-computer with no gaps in between.
Google may be the closest to the "parallel computing solution". They use Map-Reduce on server farms and then structure the program in a client/server architecture. Thus the client can have a procedural GUI, and the server can run parallel code. In Google's case the client is a web browser, and the server is a web server attached to a server farm. Maybe we just have to get used to client-server architectures for any programs running CPU intensive code.
Slashdot Mirror
The problem with a Windows desktops is that even with GPO, and after enforcing Group Policies, you still don't have any way of enforcing policies. Think about if for a while. You can get around anything under Windows, and often it isn't necessary to try very hard.
At the other extreme, try installing a real multi-user operating system in a real multi-user environment. They are totally different paradigms. With Group Policies, you are trying to use a user mode shell with system access to enforce policies. Under Linux, policies can be enforced through the operating system, and the file system it implements. The O/S with the low-level implemented security will always be more secure.
The problem the poster is having, is that you fundamentally can't compare Group Policies with the full security infrastructure of a multi-user operating system. They are two different things. They don't work the same way, and the difference can't be papered over with a cool shell interface.
Unfortunately, what you say seems to mirror my experience. You can trivially lock down a linux system so hard that no user will ever run something without permission. Under Windows, you can configure Group Policies all day long, and there are still a billion security holes.
One of my first attempts was locking down a Windows 95 machine. That was cracked in less than a couple of hours. After that, we tried Windows NT 4.0. With NT 4.0, we had to enable a feature that limited which executables would run on the machine, and after much testing, we made a list of 11 authorized executables and that was it. Unfortunately, 9 of the 11 were undocumented Windows internal programs. The newer versions of Windows got rid of the authorized executable list, created work-arounds (like ActiveX and RunDLL), and generally did other things that make it almost impossible to truly lock down a Windows machine. Windows XP Embedded stations have being taken over by viruses, even though the XP Embedded HD is mounted read only. It is truly difficult to configure a Windows machine to work on the shop floor, and to do it in such a way that the bored shop floor people can not mess anything up.
With Linux, you can mount an entire hard drive as read-only, and NFS mount the home directories. This makes it really tough to create permanent modifications to the system. If you want to be less draconian, one can easily restrict which executables the users have access to. Ubuntu even has switches that restrict the user's ability to mount new disk volumes to access floppy drives and USB keys. If you want to lock a computer down hard, it is much easier to accomplish the goal under Linux.
A properly configured Linux computer running a machine in real-time on the shop floor is impressive. If anything really goes wrong, just cycle the power. In practice, almost nothing does go wrong, and uptimes on the order of 6 months are achievable. Contrast this to Windows XP, which needs nightly reboots for virus updates, the machine starts stuttering when someone does a remote backup, and just about any user can find a way to entertain themselves by installing new desktop wallpapers ...
The material in the design specification was essentially unobtanium. It couldn't be manufactured at all. Quietly, the manufacturing engineers developed a solution that almost met all of the design specifications, and this was an excellent compromise. Unfortunately, the design engineers couldn't be convinced to sign off on the design change because of quality procedure 15, and military qualification 7. However, the biggest reason the design engineers wouldn't sign off on the change was because of a supposedly critical but practically useless mandatory project requirement, like the missile must work when fired in -40 degree water from 20 feet under the polar ice shelf.
The manufacturing engineers decided that the "fire nuclear missile while under ice shelf function", probably wouldn't be used, so the modified material was actually just fine. They shipped the missiles, got paid, and everyone was happy. Until now, when someone tries to "fix" the original "fix".
This story has happened before and will happen again. Whenever you bump into a design that requires a part that "does not exist", watch out for the possibility that the part never did "exist". It could be that you are reading a "design" document, and not what manufacturing actually built. I've worked in manufacturing, and there are lots of stories about impossible to make designs that somehow got shipped.
Yes, but after hundreds of billions of dollars in development expenses, this is the most expensive open source software in the world.
The world financial meltdown must count as the world's most expensive software bug ever. If this software had been widely accessible and better understood sooner, the macro-economic consequences of the activities relating to this software might have been better understood much sooner. The crisis might have been avoidable.
The world financial meltdown is a consequence of "security through obscurity," or in simpler terms: "What you don't know, won't hurt you."
I think you are on the right track. They have built something not all that different from an induction motor. Essentially, there are two different types of electrodes. The first pair is in the solution, and a current is going to be established. Given this is water, the current will be established by lining up positively charged hydrogen sides of the atoms with the negative electrode (the electron source). The negatively charged oxygen side of the water molecule will line up with the positive electrode (electron sink). A small current will flow, forcing the water to move. Initially, this motion won't be circular, to get circular motion you need a second set of electrodes.
The second set of electrodes develops a electro-static field. This set of electrodes isn't actually in contact with the water. The current carrying water molecules will try to align with this second set of electrodes, while carrying current. Of course, the fields are in two different directions, so static alignment isn't possible. As such, the motor starts to spin.
There are probably a bunch of other effects happening. Motors involving electro-statics tend to be very friction sensitive and fickle beasts. However, if you have a current, a dipole, and either a magnetic or an electro-static field, you will get torque. It is just in this case, the dipole is the water molecule, and as such the torque causes the water to spin.
For the poster wondering why the water at the center spins faster, the reason is likely due to the available torque and energy. Chances are the torque is proportional to the field strength (which is constant.) The water at the center has less distance to travel, and less drag. Thus it can spin faster. It would be quite unexpected for a liquid to behave like a solid, and travel at the same rotational rate at all diameters.
Finally, some posters have wondered how the motor is being powered. The power to spin the liquid is coming from the current flow between the two electrodes in the solution, and the corresponding voltage drop.
Advertising based on word of mouth is fickle for advertisers. If you have something good, then it works better than anything else. People trust friends. If your product sucks, then people still trust their friends, and won't touch your product.
Internet P2P programs like BitTorrent amplify this effect. Now, you can listen to something yourself, and figure out for yourself how much you like it. Thus P2P results in a dramatic decrease in control for advertisers. It is even more fickle than word of mouth.
If you have a poor product, but from a band with a good reputation, then you want to blitz market the product. Let no one listen to it in advance. Have it show up at stores in massive quantities the day of launch, and sell as much as you can on the first day. This way you can scam as many people as possible for first day sales. With some luck, this first day blitz will cover your costs, and everything will turn out OK. The movie industry specializes in this tactic.
P2P threatens to completely destabilize this advertising tactic. The record companies, which are really big advertisers, will not be happy about this loss of control. Even if P2P ultimately makes them more money, the record companies still won't be happy about the loss of control.
This could be fun. Here are some more suggestions:
- Welder - The little chips don't last long against a good arc welder.
- 600 VAC - Why stop at a wall outlet?
- Tesla Coil - 200 kV is better than 600 VAC
- Lightening Rod. Why stop at 200 kV?
- Oxy-acetylene Torch - higher temperatures
- Plasma Cutter - even higher temperatures
- NdYAG Laser - Etch your name into the remains of the flash chip.
- Chew Toy for Dog - Don't underestimate some of those canines, although USB keys might not be good for them.
- Log-Splitting Practice. How good are you at aiming that Axe?
- Place USB in Cement Footings of a building. Do the mob thing.
- Rock crusher
- Grinding Machine
- Wood chipper / pulper
- Cement kiln
- Blast Furnace
- Industrial Press - Terminator Style!
I'm pretty sure that some of these machines can destroy industrial quantities of USB keys, with little difficulty. Cement kilns and rock crushers can destroy just about anything. It would be interesting the see the resulting crushed rock in a piece of cement though. It would be colorful.
I would move /tmp to either a RAM disk or a hard drive. There is no point in having tmp files using up the lifespan of your SSD, especially after you just moved /home to extend its life. Also, you could move some of the stuff in /var to a hard drive or ramdisk. Good candidates might be /var/tmp and /var/log. Alternatively, you could just move the entire /var hierarchy to a hard drive.
FPGA's are really slow for some applications. This means a narrow window exists between an application that can be done on a micro-controller in software, and the equivalent application being done in an FPGA.
Many of my applications require long chains of counters and magnitude comparators. FPGA's seem to be particularly bad at implementing them. The estimates I use are:
A 50 MHz Schottky TTL counter can count at 25 MHz fairly easily.
A 100 MHz FPGA can count at 2-4 MHz, and
A 25 MHz micro-controller can count at 100 kHz.
Certain classes of PAL like devices can count quickly, but may be limited in how many bits they can put onto a single chip. For any given application, I try to figure out how quickly the counter chain needs to count at, and select appropriate solutions based on that. Often a micro-controller is a preferred solution, as so many different things can be done with it. Additionally, many modern devices have dedicated hardware that can remove the need for the FPGA entirely.
Counters also seem to be a particularly nasty corner case for FPGA devices. They are cheaply available as stock parts, so an FPGA based solution can be very expensive. Also, modern micro-controllers with built-in counter hardware are becoming very inexpensive. Sometimes cheaper than TTL devices.
Counters also have a very critical timing chain where one bit (the carry bit), must make it through every stage of the device inside one clock cycle. In an FPGA, every logic operation, interconnect, and cross-connect carries a timing penalty. FPGA's are different from PALs. In a PAL, every output bit can be used in any combinatorial operation on any output inside one clock cycle. In FPGA's, combinatorial delays can stack up very quickly, and significantly.
Researchers simulating new computer hardware designs love to use FPGA's, because expensive FPGA's can be used to simulate complex pieces of dedicated silicon. The delays in FPGA's accumulate in a similar scaled fashion as to what would happen in dedicated silicon. The resulting FPGA based solutions though, are roughly at one to two an orders of magnitudes worse than dedicated silicon, on a speed basis, a cost basis, and a speed/cost basis. This isn't really a problem if the simulation is only to verify functional correctness before one commits to "real" silicon, but it is a big problem if you are trying to build something with cheap hardware.
It's the one to two order of magnitude slower problem that always seems to mean that software solutions are always just a hair worse than FPGA based solutions. Every once in a while, a very FPGA friendly application does occur, and then the technology works well. It's just that it takes the "right" applications. Specifically, a register oriented application with a relatively small number of combinatorial inputs per register.
Well, I'm from Canada. I'm pretty sure everyone responsible for any large collection of computers is responsible for somehow "viewing or possessing child pornography." Seriously, you don't believe every single naked or sexual image of every girl on the internet is someone over 18?
- What if you have two teenagers on your network exchanging pictures of each other?
- How do you prevent any illegal activity from occurring on your network?
- How do you tell how old someone is? Can you tell the difference between a 16 year old and a 18 year old?
- Can you spot the illegal images in the millions of random GIF and JPEG images crashing around your network?
- Canadian law also covers written and drawn descriptions. How will you stop that?
- If the police suspect your company of having this stuff on your network, they will likely get a search warrant and seize all your computers (and backups.) How will your business deal with this?
Essentially, under Canadian law, almost everyone with a computer is guilty of possessing child pornography. If you have a run or own a big network of computers, you are guilty.
The scary thing is, in some countries the cops are starting to get really serious about this. In Britain, someone estimated the number of pedophiles at 60,000 to 70,000 people! That's 0.1% of the general population, and 0.2% of the men. Doesn't that number seem just a little bit too high?
If we just keep watering the definitions down, and expanding our terms of reference, then eventually we will be able to arrest everyone for something. It's always nice to start with the weird and strange people first. After all, those teen male computer hacker geeks, they always have naked images on their computers. At least one of them must have an image of a girl that is less than 18 years old!
Teenagers have already been charged in Canada under these laws. For the moment, computer professionals haven't been targeted by these laws, but it is only a matter of time. Once a computer professional is targeted, they will have a very hard time proving they are criminally innocent. In the mean time, their career, reputation and businesses will be completely destroyed. All it would take, is an accusation.
This is gorilla marketing. This story follows a five step plan:
1. Phone gets stolen from executive. Leak story to press.
2. The executive's teenage daughter is arrested by police. She became completely addicted to her father's new Zune phone. She comments: "The new Zune phone is amazingly cool."
3. Make billions of said phone.
4. ???? (Sell Zune Phones????)
5. Profit
At least, that is the story from the marketing department's internal emails ...
On a more serious note, does anyone have any idea how Microsoft plans to generate significant sales of Microsoft phones? Being a late mover in a competitive marketplace with entrenched successful competition does not scream "Instant Sales" to me ...
I'm not sure I agree with the person that moderated the parent as Flame Bait. Microsoft is a big enough target that it doesn't want to get sued over copyright violations. That was why Vista and Windows 7 have all that DRM crap. Now Microsoft wants to build a secure utility to transparently share files between people over the internet.
Has anyone ever built a secure file sharing utility over the internet that hasn't been abused in some way? Ever?
Sometimes it is just too easy to guess peoples passwords. People will share the potentially embarrassing items, whether it is an embarrassing picture, or a copyrighted song. Microsoft will log all this information. One enterprising teen could make all of your dirty laundry public knowledge.
Activate the Halon fire suppression system and run ...
Alternatively, simply adding a bunch of dry ice in an enclosed space below floor level will likely clear out any living organisms too. Be careful that you don't suffocate too.
Finally, there is always the old cyanide gas trick ... No need to worry about suffocation with cyanide.
Rats and mice are also different problems. If you have mice, cats are very effective. Mice will not even approach anywhere they think a cat lives. If you have rats, you will need a larger predator. At least a big cat, that you know will take out rats. Rats are much larger than mice.
I would consider lining everywhere there are cables with glue traps. That will catch anything that goes near the cables. Unfortunately, it could also be highly annoying. Line everywhere a cable enters or exits a small whole with steal wool. Mice are almost impossible to prevent entering a building, because they can move freely through such small entry points. They also seek out heat.
Finally, if the problem is rats, then it is much easier to block entry to the buildings. Rats are much larger than mice, so physical protection methods work better against rats. Be prepared to use concrete and steel solutions. Rats and squirrels can chew through wood. My experience is that rats will eat plastic much more readily than mice. Rats are much larger than mice, and are tougher to catch. Mouse traps are ineffective against rats. Consider sheathing your wiring in metal and/or concrete. Quick setting concrete is an easy way to plug oddly shaped holes. Metal conduit can be terminated with liquid tight fittings. Between the two solutions, you should be able to prevent mice and rats from either going through conduit, or going around conduit and exploiting holes in the building walls.
At least one hearing aid manufacturer deliberately does its DSP work in analog, to get the 1/30th the power with a minimal accuracy loss. The problem with digital filters (like DSPs) is that they use lots of power relative to analog filters. The fundamental design issue with what the author is suggesting, is that in digital the bits all have different weights. The difference between 2047+1=0 and 2047+1=2048 is huge, however in an ALU the difference is a single badly propagated carry bit.
Analog electronics doesn't suffer from the bit weighting problem. Thus a little bit off in analog is almost always just a little bit off. It makes a big difference. A scratched LP that is still playable, and a scratched DVD, with a much better sound quality, is unplayable. I deliberately picked DVDs for this example, as CDs have an exotic error-correction algorithm which makes them playable even if they are fairly damaged. As such, it isn't fair to compare a scratched CD to either a DVD or an LP. CDs are by far the least scratch sensitive media, but LPs which are analog, are more usable with scratches than most other digital media.
In any case, this is why some low power applications are implemented in analog and not digital.
Anyone hoping to pocket a percentage of $9,000,000 by giving a bunch of passwords to a bunch of people you don't know, and then assuming you won't get grassed out to the cops is likely making a major mistake.
If the criminal is smart, a better strategy might be to "give" the information away to the right group of people. This might give someone a smug sense of "revenge" against a former employer. Someone could short the stock in the stock market, or the theft could cover up some insider funny business. The initial criminal act may be different than what it appears.
Alternatively, the actual "inside" mastermind may actually be a victim too. Maybe someone conned an insider for information, or access to a laptop, and just sold the information. Maybe someone got hold of the backup tapes. This might actually a fairly low-value theft for the original criminal.
This makes my heart leap for joy.
Seriously, I can't tell you how good this makes me feel. I'm going to cry ...
It is like seeing the first day of spring, after you have been in jail for far, far, too long.
At one company I worked at, Excel was almost designated the company's standard word processor. You can write text in Excel. You can do tables and charts in Excel. You can do databases in Excel. Why use the other packages for anything else?
The scary thing is, the proposal kind of makes sense ...
Maybe, someone should start selling black hole insurance policies. The payout is $1,000,000 if your house gets swallowed by a black hole.
Quick! Protect yourself, your family, your house, in the event an evil LHC black hole swallows them. The premium is $100/month ...
I once reprimanded a intern programming student for following the MSDN example code. There is something wrong with saying to someone: "I know you know that Microsoft example code does not work. You can tell by just looking at it."
There is something wrong about needing to teach young programmers to not follow the Microsoft documentation. Even after 10 years, Microsoft still hasn't went back and fixed the documentation either. They just created more API's with more documentation problems.
To all the mods, please don't destroy all my Karma. I really do hate that Reply All button.
I think most North Americans would be quite surprised at how fast civilization would fall apart if our supermarkets stopped magically "replentishing" themselves with food. The supply chain from the farm to the supermarket isn't very long, often only a few days, and usually less than one month.
The end of modern civilization is only a major global crisis away. We are already experiencing a global recession caused by a banking crisis that started in only one country.
Traditionally, you launch a technology like this into a sector willing to pay a heavy premium for the parts. If they were really able to produce capacitors with these specifications, one would think they would go after the ultra-exotic small volume applications first. These applications often provide huge amounts of funding, allowing you to debug your product and manufacturing techniques. Thus, initial commercialization is often to DARPA, or some other defense, military, electronic or specialized power conversion application. After you get your production processes down, then you go after the high-volume low-cost marketplace. Batteries are a fairly established, low cost market. Even if your product is good, it may be difficult to make enough product to meet demand.
For me, this patent begs the question: if you had capacitors this good, why are you selling them to compete against batteries?
I think your comment is bang on. Someone only has to generate software that fulfills the contract, and gets approved for use. It doesn't actually have to work correctly ...
I think FORTRAN is going to stay the untouched leader in the fast and easy to program supercomputer / parallel mathematical simulations market. C is a very close second. Simply put, every supercomputer manufacturer makes sure its FORTRAN and C code is very quick, because that is what most of the customers use. I can't see Haskell competing in the supercomputer market / parallel processing market yet.
The next question is: Is Haskell going to be used in your next GUI based application that has a really processing intensive back end? Well functional programming is about finding a nice easy to program way to abstract and hopefully eliminate the serial aspects of your program. GUI code is about as serial as it gets. Users find it very confusing when programs pop up dialog boxes in parallel. This means a fundamental language-independent tension will exist in your program between the serial GUI code and the parallel back end code.
I don't think a good solution for the "easy parallel programming problem" is on the market yet. I would appreciate suggestions, because I am looking. I think Microsoft missed a big opportunity with .NET. Microsoft released a new programming architecture, but they omitted ground-up support for NUMA, supercomputing, and parallel processing architectures. The next big jump in computing will be the language that can smoothly harness various computing configurations ranging from single-core/single-computer to multi-core/multi-computer with no gaps in between.
Google may be the closest to the "parallel computing solution". They use Map-Reduce on server farms and then structure the program in a client/server architecture. Thus the client can have a procedural GUI, and the server can run parallel code. In Google's case the client is a web browser, and the server is a web server attached to a server farm. Maybe we just have to get used to client-server architectures for any programs running CPU intensive code.