Isn't this just a self fulfilling prophecy? I mean, everybody will just be trying to download all the porn they can before the internet goes down which will intrinsically shutdown the internet.
If you work with formally trained engineers (mechanical, electrical, civil, chemical, aerospace...) and see how the good ones in those fields go about their daily work, you will find that the truly good software engineers work much more like them than the people who just code.
In the end, Engineering is a process. The good ones, in any field, work and produce similarly. That's not to say formal engineering training automatically makes you great. But, you do get more exposure to relevant technical issues.
>> Vehicle dynamics modelling was performed on PC's.
Were they using ADAMS? The people I knew that did vehicle dynamics used ADAMS on UNIX boxes.
>> No shit:-) In Addition, another trick was making sense of the data in a timely manner, which was my main objective. Just think about qualifying:
I was thinking more for the engineers who designed the cars, not just the ones supporting the race. The race engineers seem to be looking for instant gratification for the race weekend, while the car designers seem to have to take account of the data as a whole.
Your point about the driver input is missed by many fans, I think. It's clear that Schumacher has superior driving skills, but it also seems like he gives feedback that's useful to the engineers as well. Plus, he gets to work with engineers that understand what he's telling them.
F1 tires do not have treads. They have grooves. BIG difference. Treads are used to siphon water away as to make it safe to drive in the wet. The groove tires were the stupid idea of the FIA to try to slow the cars down by reducing the size of the contact patch. This was never going to work as any tire engineer would have told you the tire manufacturer would simply kick the compounds up a notch to compensate for the grooves. Had the FIA been listening to the teams, they would have reduced the effects of the aerodynamics and gone back to wider slick tires, not narrower grooved tires.
Also, the transfer of technology from racing cars to street cars is overstated. The composites industry was ready to supply the Big 3 back in the 70s, only to be turned away. They didn't make race cars with structural composites until the late 80s. A lot of the stuff you see in high tech racing cars now has been used for a long time in the aerospace industry.
F1 is to the point of being extravagant for the sake of being extravagant. Not particularly interesting. If you want to see some real racing at the higest level, two wheels is where it's at.
CFD and vehicle dynamics modelling would require some real computing power. I think HP supplies Linux clusters to Williams.
A microcontroller with a 128MB compact flash card could probably log all the data they need from a race.
Plus, collecting data is pretty easy. The real trick is making sense of the data.
It's always amusing how CEOs justify their compensation because they claim the success of the company is tightly coupled to him or her. But, when accounting irregularities arise, they start blaming every and any body else.
They could simply randomize source and destinationIP address that are traced, i.e. 734.234.332.984. I've seen it done before when a certain semiconductor chip company was tracing network patterns in attempts to come up with better protocols and hardware. I don't think they got very far, though. Plus, the actual payload data may not be that important for this type of study, just headers.
Debian installers for the most part are crap. You can't do md or lvm with the debian installer (though, I think the sarge beta installer has lvm). The best debian distro for newbies would have to bo Knoppix. As a long time user of RedHat, I have quickly gotten even lazier with apt. So, I'm willing to put up with the crappy installer in favor of apt.
From my limited experience, it appears that servlets simply wrap cgi functionality into a nice little package. Why not consider using cgicc (http://www.gnu.org/software/cgicc/cgicc.html) for C++ applications.
The dollar isn't worth anything without the American government's backing. Who's got the most dollars? American corporations are not about the free market place. They use the American government to leverage their position. Maybe we should go back to the gold standard.
Ever tried stepping through code with a debugger that's full of macros? Not fun, unless you like printf's. As far as Troll Tech using macros for their signal/slots, I don't care as someone using their toolkit. One less abstraction to worry about.
First of all, NFS is widely used because it's been around forever. People know what to expect, good and bad. Those willing to use it know how to deal with the bad. Just as developers want the most stable platform possible, the sys admins would rather deal with a stable, well tested, well documented filesystem that they have to deploy. No need to re-invent the wheel. I mean, if GPL is the reason for lack of deployment, what's the excuse for the spread of Linux and all the useful GNU tools?
Secondly, the GPL doens't require that you accept its terms. You may also choose not to distribute the changes you make. Plus, if you know all the "tips and tricks," you are free to re-invent the wheel and distribute it (or not) under the terms of your choice.
Your analogy claims that if you were to take some GPL software, modify it on your own computer, use your modifications to do something really useful (i.e. make tons of money), then because of the GPL you have to tell people what you did to the software. This is simply not the case. This is only the case if you chose to distribute your binaries.
What the GPL does is essentially forces you to list your references. When you publish the results of a research project, you are ethically required to acknowledge your references, or ultimately, nobody takes you or your work seriously. If you choose not to publish your work, then nobody cares. Surely, you are not arguing that derived works should be credited as original?
Slashdot Mirror
Isn't this just a self fulfilling prophecy? I mean, everybody will just be trying to download all the porn they can before the internet goes down which will intrinsically shutdown the internet.
If you work with formally trained engineers (mechanical, electrical, civil, chemical, aerospace...) and see how the good ones in those fields go about their daily work, you will find that the truly good software engineers work much more like them than the people who just code. In the end, Engineering is a process. The good ones, in any field, work and produce similarly. That's not to say formal engineering training automatically makes you great. But, you do get more exposure to relevant technical issues.
Big deal... try something harder. ;-)
http://avdil.gtri.gatech.edu/AUVS/IARCLaunchPoint. html
Franklin help founded a country... this guy is just a turban wearing commie. ;-)
>> Vehicle dynamics modelling was performed on PC's.
:-) In Addition, another trick was making sense of the data in a timely manner, which was my main objective. Just think about qualifying:
Were they using ADAMS? The people I knew that did vehicle dynamics used ADAMS on UNIX boxes.
>> No shit
I was thinking more for the engineers who designed the cars, not just the ones supporting the race. The race engineers seem to be looking for instant gratification for the race weekend, while the car designers seem to have to take account of the data as a whole.
Your point about the driver input is missed by many fans, I think. It's clear that Schumacher has superior driving skills, but it also seems like he gives feedback that's useful to the engineers as well. Plus, he gets to work with engineers that understand what he's telling them.
F1 tires do not have treads. They have grooves. BIG difference. Treads are used to siphon water away as to make it safe to drive in the wet. The groove tires were the stupid idea of the FIA to try to slow the cars down by reducing the size of the contact patch. This was never going to work as any tire engineer would have told you the tire manufacturer would simply kick the compounds up a notch to compensate for the grooves. Had the FIA been listening to the teams, they would have reduced the effects of the aerodynamics and gone back to wider slick tires, not narrower grooved tires. Also, the transfer of technology from racing cars to street cars is overstated. The composites industry was ready to supply the Big 3 back in the 70s, only to be turned away. They didn't make race cars with structural composites until the late 80s. A lot of the stuff you see in high tech racing cars now has been used for a long time in the aerospace industry. F1 is to the point of being extravagant for the sake of being extravagant. Not particularly interesting. If you want to see some real racing at the higest level, two wheels is where it's at.
CFD and vehicle dynamics modelling would require some real computing power. I think HP supplies Linux clusters to Williams. A microcontroller with a 128MB compact flash card could probably log all the data they need from a race. Plus, collecting data is pretty easy. The real trick is making sense of the data.
It's always amusing how CEOs justify their compensation because they claim the success of the company is tightly coupled to him or her. But, when accounting irregularities arise, they start blaming every and any body else.
They could simply randomize source and destinationIP address that are traced, i.e. 734.234.332.984. I've seen it done before when a certain semiconductor chip company was tracing network patterns in attempts to come up with better protocols and hardware. I don't think they got very far, though. Plus, the actual payload data may not be that important for this type of study, just headers.
Debian installers for the most part are crap. You can't do md or lvm with the debian installer (though, I think the sarge beta installer has lvm). The best debian distro for newbies would have to bo Knoppix. As a long time user of RedHat, I have quickly gotten even lazier with apt. So, I'm willing to put up with the crappy installer in favor of apt.
Auto pilot? There's already such a beast. It's called a train. Parallel Parking Project
Irregardless...
From my limited experience, it appears that servlets simply wrap cgi functionality into a nice little package. Why not consider using cgicc (http://www.gnu.org/software/cgicc/cgicc.html) for C++ applications.
EISA bus.
Because too many "hackers" don't seem to consider the cost and risks of upgrading, or in the manufacturing world, "retooling."
The dollar isn't worth anything without the American government's backing. Who's got the most dollars? American corporations are not about the free market place. They use the American government to leverage their position. Maybe we should go back to the gold standard.
Ever tried stepping through code with a debugger that's full of macros? Not fun, unless you like printf's. As far as Troll Tech using macros for their signal/slots, I don't care as someone using their toolkit. One less abstraction to worry about.
First of all, NFS is widely used because it's been around forever. People know what to expect, good and bad. Those willing to use it know how to deal with the bad. Just as developers want the most stable platform possible, the sys admins would rather deal with a stable, well tested, well documented filesystem that they have to deploy. No need to re-invent the wheel. I mean, if GPL is the reason for lack of deployment, what's the excuse for the spread of Linux and all the useful GNU tools?
Secondly, the GPL doens't require that you accept its terms. You may also choose not to distribute the changes you make. Plus, if you know all the "tips and tricks," you are free to re-invent the wheel and distribute it (or not) under the terms of your choice.
Your analogy claims that if you were to take some GPL software, modify it on your own computer, use your modifications to do something really useful (i.e. make tons of money), then because of the GPL you have to tell people what you did to the software. This is simply not the case. This is only the case if you chose to distribute your binaries.
What the GPL does is essentially forces you to list your references. When you publish the results of a research project, you are ethically required to acknowledge your references, or ultimately, nobody takes you or your work seriously. If you choose not to publish your work, then nobody cares. Surely, you are not arguing that derived works should be credited as original?