1/2 of our virus infections are a'la IE and browsing email sites, like yahoo and hotmail. We had to block them for that reason.
I'm glad linus did this
on
Linus on DRM
·
· Score: 2, Interesting
I've taken this position from day 1 (as soon as I was able to comprehend it)
Linux needs some DRM infrastructure, though it's use is to be discouraged. It is not tech's place to play politics. Linux will need DRM to be used everywhere, as it can today as soon as MS makes it availible for the studios to use.
I never said it was hard. Getting changes to made to the kernel is the hard part though, particularly if you are proprietary...
For exes, you have to setup a stack and heap for the program, so there's still a _little_ work to be done, but it would rule. You could boot Windows in under 10 minutes this way!
Execute in Place (EIP)- currently, your system will copy the program to RAM. Here, you'd copy everything from volatile ram to Non-volatile ram - a rather wasteful operation don't you think?
This is not just for exe's but for datafiles as well...
Well the cell-recognition is the problem. To detect 640x480, you're need at least 1280x960 on the webcam.
But if you could do that, at 16 colors, you'd have 1/2 of 10base-t speed, and that's assuming 1 flash/sec. I think there's room for at least 5 a second, since most webcams these days do 30fps.
It's be good for phsyical isolation of networks and other devices. (Camera phones?)
Well at the last place I worked, I worked for R&D, and we were always at odds with IT. They wouldn't let us put this one computer on the network because it was leased, and not under our configuration control. But it was in the same room as a computer that was. They even faced each other... and then my idea was born. You can even fan-out to multiple networks or fan-in. I think it'd be damn cool to walkinto a room full of flashing monitors.
It could wind up looking like the screens in the matrix, where you could see: "That was a UDP packet, and this is an ARP request...";-)
Now spammers can see if you're really fat or not. I can also show them my schlong, so they won't 1) tell me how to make it longer and 2) tell me how I can increase my breast size.
To be able to position a webcam from computer A in front of a monitor from computer B, and reverse it. So each camera is looking at each other's monitor.
Then I want to display crap on the screen which then gets interpreted as data (Imagine a 4x4 checkerboard, black=0, white=1, so each screen displays 16 bits at a time)
Now use this to bridge two networks.
Questions: How many cells can be fit on a monitor? How fast can you change/read the data? Ideally if your webcam is 320x200, you could get 64kbits per flash. If you can use 4 colors instead of two, you're upto ISDN speeds...
Kinda makes sense... Scorpius=male scorpion... A collar that stings you.. Yeah, I think it's a great idea! Though I think it's a better name for the technology or project name...
We do it everywhere, and a whole freaking lot when we're younger too. (Thomas the Train anyone?)
We like to ascribe organic behavioral characteristics to silicon-based machines? No way. I can fix just about any problem on a PC, but usually it's just faster to laydown the OS or application that is no longer working.
The causes of these 'mysterious' problem are either faulty memory, or corrupt files on the disk. Even corrupt datafiles can cause programs to go awry, not just corrupt libraries and exes.
I've seen a common problem on windows - that it "forgets" about the network. THe files are there, everything should be working, but it just doesn't. This problem can be fixed by extracting the DLLS, and VXDs out of the cabs that are associated with networking.
While I said it "forgot" networking, it didn't really. A DLL failed to load because it got corrupt on disk which prevented the stack from loading completely. Once refreshed with the uncorrupt one, it works fine. I suspect all the quirks of PCs can be traced back to corruption on the file system or a bad flip-flop in RAM.
If PCs developed personalityies, then Computer Science would cease to exist, and wou'd all have to get degrees in neuroscience!
This is true for *EVERYTHING* today. If it exists today, then it's stable. I can think of no process that has not completed a single cycle. If my assertion is correct, then there are no unstable systems.
The only possible exception that comes to mind is the expansion of the universe, but current data indicates that we will forever expand, and never collapse. That is the only acyslic thing in the universe. (Known to me)
I call stabilities "reinforcing harmonics". It's a pattern of behavior that doesn't lead to it's own destruction. (Not that it won't ever be destroyed, but it won't destroy itself) Life is one such example (you don't exist and die out, you exist and propigate - any life that is life has since stablized), self-regulating stars are another. But stars are only self-regulating for a while - while they have fuel enough to hold a stable reaction - then they begin to die.
This implies existance a 2nd order stability, stability of stabilities. The longer you last, the higher your 2nd order stability. This follows the first rule - if you're existing today, then you've got a decent 2nd order stability.
So to recap, if it wasn't stable, it'd be gone by now. The only way for that to be incorrect is for a massive amount of stored energy (or lack of) that would disrupt the reinforcing harmonics
I can just see it now, both sides duking it out in a NASA meeting room. One side says it's not a problem. The other says it'snot. Eventually, they decide that frying people alive is cheaper and less of an inconvience than sending up more air on a rocket and sending up another shuttle shortly there after.
When an airplane is defective, they don't continue the flight until it is repaired. I've been on flights where they landed and made us switch planes because a problem was discovered. I would never expect them to finish the flight. If they did, and we crashed, then that's negligence.
So yes, I do expect that.
Furthermore, if we could transfer passengers in mid flight, I'd expect that too. But don't start comparing an atmosphereic docking to a space docking.
Knowing about a problem and doing nothing about it IS negligence.
We can be critical without villanizing NASA. Were there easy measures to be taken that could have prevented their deaths? Absolutely. Did they know of these measures? Yeap, as a matter of fact they declined to implement them even they cost NOTHING.
Being dangerous is one thing, it comes with the territory. Being stupid is a completely different matter, and there is no excuse for it at NASA.
"risking another shuttle and an untrained crew" I'd hope that by now NASA wouls have enough poeple to put together an ad-hoc shuttle flight. True, it would take some time to get it to the lauch pad and everything set up, but by then the crew should be in place.
If NASA hasn't come up with such a simple contingentcy plan and crew by 2003, that too is negligence.
Transer crew to the ISS, and send up another shuttle full of air and repair/dismantle equipment. Throw it into orbit and if it can't be repaired, bring it down bits at a time for recyling in another shuttle. Or leave it up there for spare parts.
They KNOW the shuttle was damaged before re-entry. They declined to have it properly inspected before re-entry. Now they try to make it appear that it's some big mystery.
Suffice to say, having any damamge (her than a few heat tiles) should bar it from re-entry. I'm no NASA expert (one wonders if they exist these days...) but I know that wind damage, particularly a deforemed wing's leading edge surface would cause exccessive friction what would result in heat build up enough to liquify metal.
I don't know why there isn't a nelgigence suit against them. With all the offers they declined, it's pretty clear cut.
Patent litigation. Developing patents then sueing people for using them is going to be the next real business. Forget innovating, we can sue people and get quicker rests at much higher profit margins!
Someone will then patent a "patent trial" and then put an end to it all. (And not a good thing either - it'll be the end of innovation in America)
Speed is one factor among many. Any respectible BC knows that databases must provide 4 things (ACID): A)atomicity C)consistantcy I)Isolation D)Durablilty
Caching data in RAM greatly detracts from D and C. I'd never be a memeber of a bank that used this kind of RAM cache database, because I could wake up and none of my deposits would be there if the server went down before the nightly flush.
In addition relational databases' biggest boom is that they use SQL, relatively standard and easy to use.
You should be able to achive better results in a properly tuned database engine and properly designed schema.
Most databases are NOT tuned. Being a member of PostgreSQL's mailing lists, I can tell you that all benchmarks (at least those run against PostgreSQL) are run against the default, very conservative config.
Slashdot Mirror
1/2 of our virus infections are a'la IE and browsing email sites, like yahoo and hotmail. We had to block them for that reason.
I've taken this position from day 1 (as soon as I was able to comprehend it)
Linux needs some DRM infrastructure, though it's use is to be discouraged. It is not tech's place to play politics. Linux will need DRM to be used everywhere, as it can today as soon as MS makes it availible for the studios to use.
I stand behind Linus here.
Developers! Developers! Developers! Developers!
What do developer's like to use? GPL/LGPL baby!
Yes, Linux is very close to this, but it can be done better still if the filesystem is known to be of this sort.
I never said it was hard. Getting changes to made to the kernel is the hard part though, particularly if you are proprietary...
For exes, you have to setup a stack and heap for the program, so there's still a _little_ work to be done, but it would rule. You could boot Windows in under 10 minutes this way!
Execute in Place (EIP)- currently, your system will copy the program to RAM. Here, you'd copy everything from volatile ram to Non-volatile ram - a rather wasteful operation don't you think?
This is not just for exe's but for datafiles as well...
No, that since I have a shlong, I have no need for breast enlargement...
Shhhhhh!
But see, this requires no fiber!
Well the cell-recognition is the problem. To detect 640x480, you're need at least 1280x960 on the webcam.
But if you could do that, at 16 colors, you'd have 1/2 of 10base-t speed, and that's assuming 1 flash/sec. I think there's room for at least 5 a second, since most webcams these days do 30fps.
It's be good for phsyical isolation of networks and other devices. (Camera phones?)
Well at the last place I worked, I worked for R&D, and we were always at odds with IT. They wouldn't let us put this one computer on the network because it was leased, and not under our configuration control. But it was in the same room as a computer that was. They even faced each other... and then my idea was born. You can even fan-out to multiple networks or fan-in. I think it'd be damn cool to walkinto a room full of flashing monitors.
;-)
It could wind up looking like the screens in the matrix, where you could see: "That was a UDP packet, and this is an ARP request..."
Now spammers can see if you're really fat or not. I can also show them my schlong, so they won't 1) tell me how to make it longer and 2) tell me how I can increase my breast size.
To be able to position a webcam from computer A in front of a monitor from computer B, and reverse it. So each camera is looking at each other's monitor.
Then I want to display crap on the screen which then gets interpreted as data (Imagine a 4x4 checkerboard, black=0, white=1, so each screen displays 16 bits at a time)
Now use this to bridge two networks.
Questions: How many cells can be fit on a monitor?
How fast can you change/read the data?
Ideally if your webcam is 320x200, you could get 64kbits per flash. If you can use 4 colors instead of two, you're upto ISDN speeds...
Kinda makes sense... Scorpius=male scorpion... A collar that stings you.. Yeah, I think it's a great idea! Though I think it's a better name for the technology or project name...
Some schools require that the parents sign all homework that gets turned in!
;-)
This way, you can still do an inferior job.
What's next: radio controlled dog collars that shock you until your homework is done, or if you cut class.
We do it everywhere, and a whole freaking lot when we're younger too. (Thomas the Train anyone?)
We like to ascribe organic behavioral characteristics to silicon-based machines? No way. I can fix just about any problem on a PC, but usually it's just faster to laydown the OS or application that is no longer working.
The causes of these 'mysterious' problem are either faulty memory, or corrupt files on the disk. Even corrupt datafiles can cause programs to go awry, not just corrupt libraries and exes.
I've seen a common problem on windows - that it "forgets" about the network. THe files are there, everything should be working, but it just doesn't. This problem can be fixed by extracting the DLLS, and VXDs out of the cabs that are associated with networking.
While I said it "forgot" networking, it didn't really. A DLL failed to load because it got corrupt on disk which prevented the stack from loading completely. Once refreshed with the uncorrupt one, it works fine. I suspect all the quirks of PCs can be traced back to corruption on the file system or a bad flip-flop in RAM.
If PCs developed personalityies, then Computer Science would cease to exist, and wou'd all have to get degrees in neuroscience!
This is true for *EVERYTHING* today. If it exists today, then it's stable. I can think of no process that has not completed a single cycle. If my assertion is correct, then there are no unstable systems.
The only possible exception that comes to mind is the expansion of the universe, but current data indicates that we will forever expand, and never collapse. That is the only acyslic thing in the universe. (Known to me)
I call stabilities "reinforcing harmonics". It's a pattern of behavior that doesn't lead to it's own destruction. (Not that it won't ever be destroyed, but it won't destroy itself) Life is one such example (you don't exist and die out, you exist and propigate - any life that is life has since stablized), self-regulating stars are another. But stars are only self-regulating for a while - while they have fuel enough to hold a stable reaction - then they begin to die.
This implies existance a 2nd order stability, stability of stabilities. The longer you last, the higher your 2nd order stability. This follows the first rule - if you're existing today, then you've got a decent 2nd order stability.
So to recap, if it wasn't stable, it'd be gone by now. The only way for that to be incorrect is for a massive amount of stored energy (or lack of) that would disrupt the reinforcing harmonics
I can just see it now, both sides duking it out in a NASA meeting room. One side says it's not a problem. The other says it'snot. Eventually, they decide that frying people alive is cheaper and less of an inconvience than sending up more air on a rocket and sending up another shuttle shortly there after.
When an airplane is defective, they don't continue the flight until it is repaired. I've been on flights where they landed and made us switch planes because a problem was discovered. I would never expect them to finish the flight. If they did, and we crashed, then that's negligence.
So yes, I do expect that.
Furthermore, if we could transfer passengers in mid flight, I'd expect that too. But don't start comparing an atmosphereic docking to a space docking.
Knowing about a problem and doing nothing about it IS negligence.
We can be critical without villanizing NASA. Were there easy measures to be taken that could have prevented their deaths? Absolutely. Did they know of these measures? Yeap, as a matter of fact they declined to implement them even they cost NOTHING.
Being dangerous is one thing, it comes with the territory. Being stupid is a completely different matter, and there is no excuse for it at NASA.
Yes they did. They had offers from the military to use satalites to inspect it. Why they did not just get out and look is beyond me.
I'm not acting like an expert, just someone who has been following this apearently a lot closer that you.
"risking another shuttle and an untrained crew"
I'd hope that by now NASA wouls have enough poeple to put together an ad-hoc shuttle flight. True, it would take some time to get it to the lauch pad and everything set up, but by then the crew should be in place.
If NASA hasn't come up with such a simple contingentcy plan and crew by 2003, that too is negligence.
Transer crew to the ISS, and send up another shuttle full of air and repair/dismantle equipment. Throw it into orbit and if it can't be repaired, bring it down bits at a time for recyling in another shuttle. Or leave it up there for spare parts.
They KNOW the shuttle was damaged before re-entry. They declined to have it properly inspected before re-entry. Now they try to make it appear that it's some big mystery.
Suffice to say, having any damamge (her than a few heat tiles) should bar it from re-entry. I'm no NASA expert (one wonders if they exist these days...) but I know that wind damage, particularly a deforemed wing's leading edge surface would cause exccessive friction what would result in heat build up enough to liquify metal.
I don't know why there isn't a nelgigence suit against them. With all the offers they declined, it's pretty clear cut.
Patent litigation. Developing patents then sueing people for using them is going to be the next real business. Forget innovating, we can sue people and get quicker rests at much higher profit margins!
Someone will then patent a "patent trial" and then put an end to it all. (And not a good thing either - it'll be the end of innovation in America)
Speed is one factor among many. Any respectible BC knows that databases must provide 4 things (ACID):
A)atomicity
C)consistantcy
I)Isolation
D)Durablilty
Caching data in RAM greatly detracts from D and C.
I'd never be a memeber of a bank that used this kind of RAM cache database, because I could wake up and none of my deposits would be there if the server went down before the nightly flush.
In addition relational databases' biggest boom is that they use SQL, relatively standard and easy to use.
You should be able to achive better results in a properly tuned database engine and properly designed schema.
Most databases are NOT tuned. Being a member of PostgreSQL's mailing lists, I can tell you that all benchmarks (at least those run against PostgreSQL) are run against the default, very conservative config.