Ok, I think I get your point: It cannot be proved that the class loader will not accept an illegal class. I agree to that part.
My point is that the class loader (and the Java sandbox) must be bullet-proof, and that part should be proved to be correct. The class loader is a "first line of defence". What must be proved here is that the class loader cannot itself cannot be penetrated.
Did you just walk out of an undergrad Computer Science class?;)
No - I have a colleague, that used to work for ESA. Proving that software really works as it's supposed to do, is something you do, when you cannot afford the consequence of failure (eg. loss of human life, or perhaps even worse: bad PR...).
Now, my question for you folks: Could Java afford a failure in this area, or should SUN do what it takes?
I am a bit tired of the aproach "Let's just see if it works!". That aproach works well on an old car, but it does not work well on the linch-pin of one of the most important technologies today!
Why not do what it takes: Prove that it will work, and prove that it cannot be broken!
Could Java/SUN afford a major flaw in the Java sandbox/class loader...? I think not!
I think the price tag is quite important for the Nintendo 'Revolution'. So is the precision of the handle: Is it is not precise enough, it cannot be used for many games.
Ok, I imagine that a gyroscope is quite expensive. It will require either a high-speed wheel of some sort, bearings and detectors. It is also quite delicate, and to imagine a gyroscope that should survive a kid dropping it on a floor...? An optic gyscope perhaps, but then the price tag would be quite different...
Furthermore, a gyroscope (mechanical or optical) will only detect twists and turns. It will not detect movements along the three axes, which is something the controller can detect.
And then there is the two small devices that needs to be placed with some distance, one on each side of the TV/monitor....
But... What if these two small devices were IR diodes, and the controller contained a small camera for motion detection....?
It would then be able to calculate twists, turns and movements along the axes, by looking at the diodes, and calculate the movements! It woul certainly give the wanted precision in some of the directions (depending of the IR camera resolution, of course)
Another thing that could hint some IR detector on the controller, is the 'window' on the end that points in the direction of the TV/monitor. I do not see an IR receiver on the 'Revolution'...
I think the price tag is quite important for the Nintendo 'Revolution'. So is the precision of the handle: Is it is not precise enough, it cannot be used for many games.
Ok, I imagine that a gyroscope is quite expensive. It will require a high-speed wheel of some sort, bearings and detectors. It is also quite delicate, and to imagine a gyroscope that should survive a kid dropping it on a floor...? I didn't think so.
Furthermore, a gyroscope will only detect twists and turns. It will not detect movements along the three axes, which is something the controller can detect.
And then there is the two small devices that needs to be placed with some distance, one on each side of the TV/monitor....
What if these two small devices were IR diodes, and the controller contained a small camera for motion detection....?
It would then be able to calculate twists, turns and movements along the axes, by looking at the diodes, and calculate the movements! It woul certainly give the wanted precision in some of the directions (depending of the IR camera resolution, of course)
Another thing that could hint some IR detector on the controller, is the 'window' on the end that points in the direction of the TV/monitor. I do not see an IR receiver on the 'Revolution'...
Google's 'Summer of Code' http://code.google.com/summerofcode.html looks like an appealing idea. How about following up on that, eg. by donating 100.000$ to a company, that in return for the publicity, will organize, administer, publish, hand out, whatever it takes, to distribute 100.000$ to selected Open Source Projects programmers. Next year, another company would be selected the honerable job.
Slashdot Mirror
Ok, I think I get your point: It cannot be proved that the class loader will not accept an illegal class. I agree to that part.
My point is that the class loader (and the Java sandbox) must be bullet-proof, and that part should be proved to be correct. The class loader is a "first line of defence". What must be proved here is that the class loader cannot itself cannot be penetrated.
Did you just walk out of an undergrad Computer Science class? ;)
No - I have a colleague, that used to work for ESA. Proving that software really works as it's supposed to do, is something you do, when you cannot afford the consequence of failure (eg. loss of human life, or perhaps even worse: bad PR...).
Now, my question for you folks: Could Java afford a failure in this area, or should SUN do what it takes?
I am a bit tired of the aproach "Let's just see if it works!". That aproach works well on an old car, but it does not work well on the linch-pin of one of the most important technologies today!
Why not do what it takes: Prove that it will work, and prove that it cannot be broken!
Could Java/SUN afford a major flaw in the Java sandbox/class loader...? I think not!
It's simply brilliant - I just can't wait to try it out!
Here are my reasons for liking it:
Bring it on - I'm ready to order, and I cannot wait to play Conkers Bad Fur Day again!! :-)
Ok, I imagine that a gyroscope is quite expensive. It will require either a high-speed wheel of some sort, bearings and detectors. It is also quite delicate, and to imagine a gyroscope that should survive a kid dropping it on a floor...? An optic gyscope perhaps, but then the price tag would be quite different...
Furthermore, a gyroscope (mechanical or optical) will only detect twists and turns. It will not detect movements along the three axes, which is something the controller can detect.
And then there is the two small devices that needs to be placed with some distance, one on each side of the TV/monitor....
But... What if these two small devices were IR diodes, and the controller contained a small camera for motion detection....?
It would then be able to calculate twists, turns and movements along the axes, by looking at the diodes, and calculate the movements! It woul certainly give the wanted precision in some of the directions (depending of the IR camera resolution, of course)
Another thing that could hint some IR detector on the controller, is the 'window' on the end that points in the direction of the TV/monitor. I do not see an IR receiver on the 'Revolution'...
Comments anyone...?
I think the price tag is quite important for the Nintendo 'Revolution'. So is the precision of the handle: Is it is not precise enough, it cannot be used for many games.
Ok, I imagine that a gyroscope is quite expensive. It will require a high-speed wheel of some sort, bearings and detectors. It is also quite delicate, and to imagine a gyroscope that should survive a kid dropping it on a floor...? I didn't think so.
Furthermore, a gyroscope will only detect twists and turns. It will not detect movements along the three axes, which is something the controller can detect.
And then there is the two small devices that needs to be placed with some distance, one on each side of the TV/monitor....
What if these two small devices were IR diodes, and the controller contained a small camera for motion detection....?
It would then be able to calculate twists, turns and movements along the axes, by looking at the diodes, and calculate the movements! It woul certainly give the wanted precision in some of the directions (depending of the IR camera resolution, of course)
Another thing that could hint some IR detector on the controller, is the 'window' on the end that points in the direction of the TV/monitor. I do not see an IR receiver on the 'Revolution'...
Comments anyone...?
How can the version without the illegal bundling be optional?!?
Google's 'Summer of Code' http://code.google.com/summerofcode.html looks like an appealing idea. How about following up on that, eg. by donating 100.000$ to a company, that in return for the publicity, will organize, administer, publish, hand out, whatever it takes, to distribute 100.000$ to selected Open Source Projects programmers. Next year, another company would be selected the honerable job.