Yes, but don't you feel that's a bit unfair to your employees and investors? Arguably the most valuable thing about going public is that is gives liquidity to your shareholders, i.e. it allows the people that helped build your company to earn some profit from their hard work / investment.
turn the car off. Just turn the ignition and pull the keys.
Definitely do not turn it all the way off and "pull the keys". If you do this, you may engage the steering wheel lock feature, and the driver could try to swerve and find the wheel locked in the turned position. If you do chose to turn off the car, just turn it one click to kill the engine. But as has been mentioned above, simply applying the brakes firmly should always be effective, and failing that, neutral is a good option. Neither of these have the steering wheel lock danger.
According to wardsauto.com, 13M cars and trucks were sold in 2011. At a cost of $200 each, that means it would cost $2.6B per year to add these cameras to every vehicle. Even if this would eliminate all 200 of the backup-related deaths each year (which it obviously wouldn't), that would mean spending $13M per life saved. This is far higher than the figure used in most engineering projects; i.e. this is not a good return per dollar on safety, and there are much more cost-effective ways to spend this money.
This box may be easy to hack, but unless they were particularly negligent in the design of their security, it won't be with a man-in-the-middle attack on the phone line communication like you've described. If they encrypt this communication using SSL, a man-in-the-middle attack will not be possible. Briefly, here's how SSL prevents such attacks:
1. Client generates a random session key. 2. Client encrypts the session key with the server's public key and transmits the encrypted key to the server. 3. Server decrypts the received session key using the server's private key. 4. Client and server send data back and forth, encrypting using the session key.
Since only the server has the private key, it is not possble for a man in the middle to recover the session key and intercept communications or impersonate the server. The man in the middle can of course alter the encrypted stream as it relays between the client and server, but doing so will corrupt the data and be noticed by the client and server.
That said, there are other attack vectors that are more plausible. The movies are on the client device, and although they are almost certainly encrypted, the decryption key must exist somewhere on the device (so it can play the movies), so once users discover where/how this key is stored, they will have unrestricted access to the content.
The problem is that a video driver needs to know very specific low-level details about the card's design in order to achieve maximum performance, which means the driver's source code will give a lot of information about the card's architecture. This would make the driver source a very useful tool to someone who wanted to reverse engineer a video card.
You're right that this is a bad design, but I think you're placing the blame in the wrong place. Games need admin priviledges to install because they often need to update system-level components such as DirectX, which requires admin priveledges.
It would be a great if an application running as user could make a system call saying "I need DirectX 9.5", and this would cause the OS to obtain the update, either over the internet or from the game installation CD (after verifying a signature). This would be the best of both worlds; it would allow user-level apps to update system components without letting them directly manipulate system directories.
I don't think it was meant as a threat. I think the logic is that if it becomes available for free, they need to release it for sale ASAP because if they wait a lot of people who would have paid for it will P2P download it for free instead.
Yes, that's certainly true if you are in a climate where you are heating your house/office. However, keep in mind that the opposite is true if you are cooling your house. In this case you are paying for the inefficiency twice; once for the computer, and then again in higher air conditioning costs to remove the excess heat.
Great analogy with renting phones. More and more often, the main driver behind getting a PC is internet access, which makes the PC very much like a telephone to many consumers, in the sense that it is just a device to access an online service.
It is interesting how that model failed for telephones. I think the main problem was that telephones became so cheap, and so many new options were added (cordless, integrated answering machines, etc.). I think this model will fail for similar reasons; as computers continue to get cheaper, consumers won't like the idea of paying a continuous fee for something they could buy outright for a hundred dollars or so.
I am quite tired of hearing this argument. Are you suggesting we should ignore all the lesser crimes to make sure the biggies get maximum attention? If someone steals your car, should the response be, "How can we spend time on that case when we have rapes and murders to prosecute?"
To be clear, I do not agree with what the RIAA is doing. But please stop basing arguments (and modding them to +5) on this poor logic. Say "it's not a crime, it's a civil matter" and I will fully agree with you. But when you compare copyright infringement to murder you are actually weakening your own position by implying that it is a crime, just not one that the courts should spend their precious time on.
I have to admit that I think your methods are a bit extreme, but your core concept is a good one.
I wonder though how practical it can possibly be. Do your children ever go over to their friends' houses? If so, I doubt many parents are as restrictive as you, so your children will certainly be exposed to some of the things you are trying to keep from them. Or do you restrict your children from leaving your home, in which case I have to worry about the lasting damage you are doing to their social skills and ability to interact with others.
I think $20M is a small price to pay for that cool of a place in history, whatever the eventual "market price" for a spaceflight turns out to be. Being the 3rd civilian in space is a bit more of an achievement than owning the 3rd Tivo. The whole concept of value changes when you're worth $100M. Would you rather:
a) Have $100M in the bank to spend on houses, cars, etc., or, b) Go into space for a couple days and then have $80M to spend on houses, cars, etc.
To me that would be a pretty easy call, but that's just me.
I agree with you in principle. However, I think you are missing an important point about the marriage of DRM and open source: It's not possible. In a DRM system, somewhere there is code that says:
if ( licenseIsValid() ) {
playContent(); }
If this is open source, I can simply comment out the validation and recompile, and no more DRM.
Okay since you worked in the industry you might be able to answer this question:
Why isn't the system set up so that when a major generator goes down and the other generators can't support the additional load, some part of the grid is immediately blacked out to reduce the load? That way only a small portion of your customers get blacked out instead of an entire region of the country.
I assume that as part of your software testing, you guys run the code through a simulator, right? In other words you run a test so that the software is receiving the exact sensor signals it would be receiving during an actual atmospheric entry/landing, and you can observe the output to make sure it does the right things at the right time.
What I don't understand is why such testing doesn't uncover almost all of these types of problems. It seems like you'd be able to watch the output and say, "Whoa, parachute detatched at altitute 2000m, that's bad," for example.
Is such testing not done, or is it that these unit problems are outside the scope of such tests (i.e. the actual physical sensors send feet when meters were sent by the simulated sensors)?
I know -- I saw the UID; you've been here a lot longer than I have. I also recognized your name from many comments that I've read. It was just a cheap ploy to try to get some karma.
I do agree with you of course. When I first started reading slashdot I'd see a headline like "Microsoft bribes Congressman not to use Linux" and get very angry at the injustice in this country, only to read the article and learn that the headline was a pretty gross exaggeration. Now I take it all with a grain of salt.
You forgot "Do I have to set the Evit Bit before driving recklessly?" That one never seems to get old...
Re:Well...
on
Ageism in IT?
·
· Score: 3, Insightful
From reading through this discussion I'm surprised no one has said this yet: The answer is that a healthy organization needs BOTH young developers and experienced developers.
The most obvious reason for this is the continuing life of an organization; if you have only older workers you are screwed when they all decide to retire, and if you hire only young developers you will watch in frustration as they make the mistakes that a more experienced developer could have forseen and avoided.
The ideal ratio is a trickier issue and is probably the source of the perceived ageism. A department that is already "top-heavy" might be motivated to hire some younger developers in order to balance things out.
Take a look at The Case for Mars by Robert Zubrin (ISBN: 0684835509). You are right that a base on the Moon is a poor starting point to get to Mars -- despite the fact that the Moon is much closer than Mars, it actually takes almost as much delta V (change in velocity, which translates to amount of fuel) to get to the Moon as it does to Mars, so the detour to the Moon (in going to Mars) is very counterproductive.
Mars has an atmosphere (albeit a thin one), water, soil that is likely to be very fertile, compounds that can be combined to make rocket fuel, and other things that are necessary for building a nearly self-sufficient base in a reasonable timeframe. The Moon has none of these. Zubrin even presents a pretty good case for why a Mars base could become profitable at some point in the future.
I don't think it will require a bytecode change. I don't know the details of how it works, but my guess is that the compiler will simply insert the casts for you. For example, when you say:
List<Foo> list = new ArrayList<Foo>(); list.add( myFoo ); Foo otherFoo = list.get( 0 );
The compiler will interpret the second and third lines as if they said:
It's a pretty elegant way to incorporate type-safe generics without imposing any changes on the VM. I think the compiler changes would be relatively easy to implement -- nothing more than inserting casts at the appropriate places.
Slashdot Mirror
Yes, but don't you feel that's a bit unfair to your employees and investors? Arguably the most valuable thing about going public is that is gives liquidity to your shareholders, i.e. it allows the people that helped build your company to earn some profit from their hard work / investment.
turn the car off. Just turn the ignition and pull the keys.
Definitely do not turn it all the way off and "pull the keys". If you do this, you may engage the steering wheel lock feature, and the driver could try to swerve and find the wheel locked in the turned position. If you do chose to turn off the car, just turn it one click to kill the engine. But as has been mentioned above, simply applying the brakes firmly should always be effective, and failing that, neutral is a good option. Neither of these have the steering wheel lock danger.
I've seen a talk on Focus Fusion and it sounds like a very interesting concept; would love to hear from experts whether the idea has any real promise.
According to wardsauto.com, 13M cars and trucks were sold in 2011. At a cost of $200 each, that means it would cost $2.6B per year to add these cameras to every vehicle. Even if this would eliminate all 200 of the backup-related deaths each year (which it obviously wouldn't), that would mean spending $13M per life saved. This is far higher than the figure used in most engineering projects; i.e. this is not a good return per dollar on safety, and there are much more cost-effective ways to spend this money.
That said, to point out a counter-example,IBM seems to have changed its tune as well without some sort of outspoken visionary.
Dude, give Lou Gerstner some credit. While he may not be as publicly "outspoken" as Jobs, he clearly had vision.
http://en.wikipedia.org/wiki/Louis_V._Gerstner%2C_Jr.
I don't understand why they can't do something like this:
http://en.wikipedia.org/wiki/Rogers_Centre/
It only takes 20 minutes to retract the roof, so it could be left closed until just before launch.
This box may be easy to hack, but unless they were particularly negligent in the design of their security, it won't be with a man-in-the-middle attack on the phone line communication like you've described. If they encrypt this communication using SSL, a man-in-the-middle attack will not be possible. Briefly, here's how SSL prevents such attacks:
1. Client generates a random session key.
2. Client encrypts the session key with the server's public key and transmits the encrypted key to the server.
3. Server decrypts the received session key using the server's private key.
4. Client and server send data back and forth, encrypting using the session key.
Since only the server has the private key, it is not possble for a man in the middle to recover the session key and intercept communications or impersonate the server. The man in the middle can of course alter the encrypted stream as it relays between the client and server, but doing so will corrupt the data and be noticed by the client and server.
That said, there are other attack vectors that are more plausible. The movies are on the client device, and although they are almost certainly encrypted, the decryption key must exist somewhere on the device (so it can play the movies), so once users discover where/how this key is stored, they will have unrestricted access to the content.
The problem is that a video driver needs to know very specific low-level details about the card's design in order to achieve maximum performance, which means the driver's source code will give a lot of information about the card's architecture. This would make the driver source a very useful tool to someone who wanted to reverse engineer a video card.
You're right that this is a bad design, but I think you're placing the blame in the wrong place. Games need admin priviledges to install because they often need to update system-level components such as DirectX, which requires admin priveledges.
It would be a great if an application running as user could make a system call saying "I need DirectX 9.5", and this would cause the OS to obtain the update, either over the internet or from the game installation CD (after verifying a signature). This would be the best of both worlds; it would allow user-level apps to update system components without letting them directly manipulate system directories.
I don't think it was meant as a threat. I think the logic is that if it becomes available for free, they need to release it for sale ASAP because if they wait a lot of people who would have paid for it will P2P download it for free instead.
Yes, that's certainly true if you are in a climate where you are heating your house/office. However, keep in mind that the opposite is true if you are cooling your house. In this case you are paying for the inefficiency twice; once for the computer, and then again in higher air conditioning costs to remove the excess heat.
Great analogy with renting phones. More and more often, the main driver behind getting a PC is internet access, which makes the PC very much like a telephone to many consumers, in the sense that it is just a device to access an online service.
It is interesting how that model failed for telephones. I think the main problem was that telephones became so cheap, and so many new options were added (cordless, integrated answering machines, etc.). I think this model will fail for similar reasons; as computers continue to get cheaper, consumers won't like the idea of paying a continuous fee for something they could buy outright for a hundred dollars or so.
I am quite tired of hearing this argument. Are you suggesting we should ignore all the lesser crimes to make sure the biggies get maximum attention? If someone steals your car, should the response be, "How can we spend time on that case when we have rapes and murders to prosecute?"
To be clear, I do not agree with what the RIAA is doing. But please stop basing arguments (and modding them to +5) on this poor logic. Say "it's not a crime, it's a civil matter" and I will fully agree with you. But when you compare copyright infringement to murder you are actually weakening your own position by implying that it is a crime, just not one that the courts should spend their precious time on.
If you are interested (as I was) in hearing more about the SOHO recovery, here is a link with some good info:
y
http://sohowww.nascom.nasa.gov/operations/Recover
I have to admit that I think your methods are a bit extreme, but your core concept is a good one.
I wonder though how practical it can possibly be. Do your children ever go over to their friends' houses? If so, I doubt many parents are as restrictive as you, so your children will certainly be exposed to some of the things you are trying to keep from them. Or do you restrict your children from leaving your home, in which case I have to worry about the lasting damage you are doing to their social skills and ability to interact with others.
I think $20M is a small price to pay for that cool of a place in history, whatever the eventual "market price" for a spaceflight turns out to be. Being the 3rd civilian in space is a bit more of an achievement than owning the 3rd Tivo. The whole concept of value changes when you're worth $100M. Would you rather:
a) Have $100M in the bank to spend on houses, cars, etc., or,
b) Go into space for a couple days and then have $80M to spend on houses, cars, etc.
To me that would be a pretty easy call, but that's just me.
I agree with you in principle. However, I think you are missing an important point about the marriage of DRM and open source: It's not possible. In a DRM system, somewhere there is code that says:
if ( licenseIsValid() ) {
playContent();
}
If this is open source, I can simply comment out the validation and recompile, and no more DRM.
Okay since you worked in the industry you might be able to answer this question:
Why isn't the system set up so that when a major generator goes down and the other generators can't support the additional load, some part of the grid is immediately blacked out to reduce the load? That way only a small portion of your customers get blacked out instead of an entire region of the country.
I assume that as part of your software testing, you guys run the code through a simulator, right? In other words you run a test so that the software is receiving the exact sensor signals it would be receiving during an actual atmospheric entry/landing, and you can observe the output to make sure it does the right things at the right time.
What I don't understand is why such testing doesn't uncover almost all of these types of problems. It seems like you'd be able to watch the output and say, "Whoa, parachute detatched at altitute 2000m, that's bad," for example.
Is such testing not done, or is it that these unit problems are outside the scope of such tests (i.e. the actual physical sensors send feet when meters were sent by the simulated sensors)?
I know -- I saw the UID; you've been here a lot longer than I have. I also recognized your name from many comments that I've read. It was just a cheap ploy to try to get some karma.
I do agree with you of course. When I first started reading slashdot I'd see a headline like "Microsoft bribes Congressman not to use Linux" and get very angry at the injustice in this country, only to read the article and learn that the headline was a pretty gross exaggeration. Now I take it all with a grain of salt.
Do the /. editors actually do any fact checking before they post???
I guess you're new here. Welcome to slashdot.
You forgot "Do I have to set the Evit Bit before driving recklessly?" That one never seems to get old...
From reading through this discussion I'm surprised no one has said this yet: The answer is that a healthy organization needs BOTH young developers and experienced developers.
The most obvious reason for this is the continuing life of an organization; if you have only older workers you are screwed when they all decide to retire, and if you hire only young developers you will watch in frustration as they make the mistakes that a more experienced developer could have forseen and avoided.
The ideal ratio is a trickier issue and is probably the source of the perceived ageism. A department that is already "top-heavy" might be motivated to hire some younger developers in order to balance things out.
Take a look at The Case for Mars by Robert Zubrin (ISBN: 0684835509). You are right that a base on the Moon is a poor starting point to get to Mars -- despite the fact that the Moon is much closer than Mars, it actually takes almost as much delta V (change in velocity, which translates to amount of fuel) to get to the Moon as it does to Mars, so the detour to the Moon (in going to Mars) is very counterproductive.
Mars has an atmosphere (albeit a thin one), water, soil that is likely to be very fertile, compounds that can be combined to make rocket fuel, and other things that are necessary for building a nearly self-sufficient base in a reasonable timeframe. The Moon has none of these. Zubrin even presents a pretty good case for why a Mars base could become profitable at some point in the future.
I don't think it will require a bytecode change. I don't know the details of how it works, but my guess is that the compiler will simply insert the casts for you. For example, when you say:
List<Foo> list = new ArrayList<Foo>();
list.add( myFoo );
Foo otherFoo = list.get( 0 );
The compiler will interpret the second and third lines as if they said:
list.add( (Foo)myFoo );
Foo otherFoo = (Foo)list.get( 0 );
It's a pretty elegant way to incorporate type-safe generics without imposing any changes on the VM. I think the compiler changes would be relatively easy to implement -- nothing more than inserting casts at the appropriate places.