> but with a distributed setup small increases in speed on the client can have dramatic increases in the processing power. if there are 1,000,000 keys processed a day, a 1% speedup = 10,000 more keys.
No. A 1% speedup isn't "dramatic." Things start getting dramatic when speed triples. Think of it this way. A data warehouse with a one terabyte array can save 10 gigs with 1% compression. Holy cow! That's like 3000 mp3s. But people with 1TB arrays don't care about 10 gigs.
You time is better spent recruiting cycles than developers.
Easy. Send the client a sata packet with known characteristics every now and then and check the results. This technique should work fine for seti.
The situation is a bit different with distributed.net. If you cannot make fake positive packets, you cannot have security. However, if you can generate fake positive packets, then you can trap all evil clients.
> to accelerate a bunch of electrons to.0001% of the speed of light requires the output of a medium sized generator and large vacuum enclosure.
Bullshit. Example: a gun.
.0001%c = 1000 ft/s, about the speed of a bullet. A bullet weighs about 10 grams. Pb is #82, 207 g/mol. So 6.2e23 * (10 / 207) * 82 = 2.4e24 electrons. Bam! 380k coulombs! 'Course my estimate is conservative because of the protons and neutrons. Presumably it would take a lot less energy to accelerate an electron only bullet. Ratio is 3600:1, so figure 1.4G coulombs.
More useful speeds (670 mph is slow) need correspondingly larger guns... Laugh now.
> Sometimes light can be made to travel through a medium faster than it's natural rate.
No. When entering a slower medium (higher index of refraction) the photons will slow down instantaneously. Otherwise windows would glow blue. Other stuff can go faster than the photons though (for n > 1.0). We can't make sound go faster than the speed of sound, but airplanes can. Likewise neutrons can go faster than light in water.
Is it possible to make an encryption system that supports one-way transmission (broadcast) to a hostile client? DIVX machines could negotiate over a phone line. Without this ability can't the media be replayed, bit-for-bit? Is protection futile?
I'd say keep the keyboards attached for the following reasons: o Some motherborads need it (press F1 to continue) o Only $5 each o Useful for correcting stupid mistakes (ifconfig eth0 down)
It's suprising how many problems you can solve "blind." Need to know if ethernet if up? (ifconfig eth0 | grep UP) || eject/dev/cdrom
I wrote a preemptive thread switcher in about 250 lines of C. It ran a 68hc11. I didn't have any sync primitives, but you could just use globals and suspend interrupts. I'm sure it could be shrunk to 175 with little trouble. Is it an os?
As far as I can tell the 68hc11 has no test-and-set instruction (!!!).Is this true?
> Progress in developing the Internet is easier than progress in any other field. The Internet is 'contrived' and entirely invented; as such it is not constrained by physical or natural boundaries.
Wouldn't it be neat to have a centralized database that would collect the hashes of various spams. Email clients could query the database to see if a message was spam before presenting it to the user. When a user receives spam, just forward it to the database and it would be blocked for everyone else. 'Course its probably been patented already.
> Your missing one _very_ important issue. The network that the files were being shared on was the property of the school.
The school didn't intercept data flowing over the lan with tcpdump. They guessed passwords to gain access to a students hard disk! My disk, my property.
> No, but if you leave your door wide open and put a sign outside your door saying "Come on in!" and someone does and notices you breaking the law, you are still accountable.
What if the door is locked and a key is under the mat? I think this situation is analogous to the password in a readme file.
I don't want to hear what JP has to say. The lawsuits, the lies, the overall sleaziness of his character--this material does not interest me. Interview someone else instead. CmdrTaco?
> My point is that I'm sure lots of people find this practice unfair. Which it isn't, if you believe in the value of a free market economy. That's all...
Yeah, but we're all threatening to not buy coke, not force gov't intervention.
Hey! I opted out of a harder spanish class. I dislike foriegn languages, they're not a subject that interests me. By taking an easier class I was able to spend more time doing fun things like programming and robotics. I might have learned less spanish, but I learned more overall.
I made a small bow out of my technics kit. It shot bamboo skewers with a rubber band. It had enough force to travel across the room and stick in the drywall.
Um, no. First of all, a the voltage regulator will maintain a constant voltage to the chip regardless of current consumption ( within some bounds). Hence the name "voltage regulator." The max clock speed for a chip is derived from the time it takes to swich the state of one of its transistors. This switching time arises from the charging and discharging of the inherently capacitive traces and mosfet gates. Slapping more voltage across the chip charges these caps faster, reducing the switching time. I refer you to _Principles_of_CMOS_VLSI_Design_ by Neil H. E. Weste and Kamran Eshraghian for further details.
Dave Patterson is now a professor here at Berkeley. He's a great lecturer, one of the best I've had. I took a lower-div architecture class from him last year. It was great being able to learn the rationale behind certain features in the systems he designed, eg. sparc reg windows. He was on the core development teams for riad, sparc, and mips, among other things. Another person who probably belongs on your first list is William Kahan of IEEE 754 fame. He is responsible for the floating point used on every modern processor. Unlike Patterson though, you don't want to take any of his courses. The last course he taught he failed a large majority of the students. His contribution was a lot more important than basic though.
I have the server if you want. P3 550, 256 megs ram, ethernet to the backbone, Linux. I haven't thought of something useful to do with it yet, this may be it.
Nikkel?? Maybe you mean "nickel" or "nickle," either is permissible.
I've never heard of the nickel - Titanium interaction you discribe. The reason titanium should be avoided in tools like screwdriver blades is that Ti is very soft. Most Ti alloys cannot be hardened above Rc 45 or so. Screwdrivers probably are hardened to Rc 55 or 60.
Slashdot Mirror
> but with a distributed setup small increases in speed on the client can have dramatic increases in the processing power. if there are 1,000,000 keys processed a day, a 1% speedup = 10,000 more keys.
No. A 1% speedup isn't "dramatic." Things start getting dramatic when speed triples. Think of it this way. A data warehouse with a one terabyte array can save 10 gigs with 1% compression. Holy cow! That's like 3000 mp3s. But people with 1TB arrays don't care about 10 gigs.
You time is better spent recruiting cycles than developers.
Easy. Send the client a sata packet with known characteristics every now and then and check the results. This technique should work fine for seti.
The situation is a bit different with distributed.net. If you cannot make fake positive packets, you cannot have security. However, if you can generate fake positive packets, then you can trap all evil clients.
> to accelerate a bunch of electrons to .0001% of the speed of light requires the output of a medium sized generator and large vacuum enclosure.
.0001%c = 1000 ft/s, about the speed of a bullet. A bullet weighs about 10 grams. Pb is #82, 207 g/mol. So 6.2e23 * (10 / 207) * 82 = 2.4e24 electrons. Bam! 380k coulombs! 'Course my estimate is conservative because of the protons and neutrons. Presumably it would take a lot less energy to accelerate an electron only bullet. Ratio is 3600:1, so figure 1.4G coulombs.
Bullshit.
Example: a gun.
More useful speeds (670 mph is slow) need correspondingly larger guns... Laugh now.
> Sometimes light can be made to travel through a medium faster than it's natural rate.
No. When entering a slower medium (higher index of refraction) the photons will slow down instantaneously. Otherwise windows would glow blue. Other stuff can go faster than the photons though (for n > 1.0). We can't make sound go faster than the speed of sound, but airplanes can. Likewise neutrons can go faster than light in water.
> aprox 19 Mbytes worth of data is stuck in the wire. Talk about latency!
Optical delay lines. He he.
Is it possible to make an encryption system that supports one-way transmission (broadcast) to a hostile client? DIVX machines could negotiate over a phone line. Without this ability can't the media be replayed, bit-for-bit? Is protection futile?
tsrif ?
I'd say keep the keyboards attached for the following reasons:
/dev/cdrom
o Some motherborads need it (press F1 to continue)
o Only $5 each
o Useful for correcting stupid mistakes (ifconfig eth0 down)
It's suprising how many problems you can solve "blind." Need to know if ethernet if up?
(ifconfig eth0 | grep UP) || eject
Ryan
I wrote a preemptive thread switcher in about 250 lines of C. It ran a 68hc11. I didn't have any sync primitives, but you could just use globals and suspend interrupts. I'm sure it could be shrunk to 175 with little trouble. Is it an os?
As far as I can tell the 68hc11 has no test-and-set instruction (!!!).Is this true?
> Progress in developing the Internet is easier than progress in any other field. The Internet is 'contrived' and entirely invented; as such it is not constrained by physical or natural boundaries.
Try designing a chip. You have no idea.
Wouldn't it be neat to have a centralized database that would collect the hashes of various spams. Email clients could query the database to see if a message was spam before presenting it to the user. When a user receives spam, just forward it to the database and it would be blocked for everyone else. 'Course its probably been patented already.
> Your missing one _very_ important issue. The network that the files were being shared on was the property of the school.
The school didn't intercept data flowing over the lan with tcpdump. They guessed passwords to gain access to a students hard disk! My disk, my property.
> No, but if you leave your door wide open and put a sign outside your door saying "Come on in!" and someone does and notices you breaking the law, you are still accountable.
What if the door is locked and a key is under the mat? I think this situation is analogous to the password in a readme file.
>Yes, i'm being nitpicky and anal. But there are worse things to be (for example, redundant).
I'd rather hear something good twice than something worthless once. That's why I didn't preview...
I don't want to hear what JP has to say. The lawsuits, the lies, the overall sleaziness of his character--this material does not interest me. Interview someone else instead. CmdrTaco?
> Is sex with statues legal?
Statutory rape, IIRC
> What exactly is the big deal about speeds in the GHz? May be a mile stone, but it's only logical progression.
So is Y2K. Look out, Moore's law says that your processor will revert to DC in about 18 months.
Ryan
> Actually with an infinite number of monk...
A finite amount of time is required though, even with an infinite amount of monkeys. Remeber that typewriters spew forth text in corpuscular units.
Ryan
> My point is that I'm sure lots of people find this practice unfair. Which it isn't, if you believe in the value of a free market economy. That's all...
Yeah, but we're all threatening to not buy coke, not force gov't intervention.
Hey! I opted out of a harder spanish class. I dislike foriegn languages, they're not a subject that interests me. By taking an easier class I was able to spend more time doing fun things like programming and robotics. I might have learned less spanish, but I learned more overall.
800 on the math sat? That's 95th percentile. Whoop-de-do.
I made a small bow out of my technics kit. It shot bamboo skewers with a rubber band. It had enough force to travel across the room and stick in the drywall.
Um, no. First of all, a the voltage regulator will maintain a constant voltage to the chip regardless of current consumption ( within some bounds). Hence the name "voltage regulator." The max clock speed for a chip is derived from the time it takes to swich the state of one of its transistors. This switching time arises from the charging and discharging of the inherently capacitive traces and mosfet gates. Slapping more voltage across the chip charges these caps faster, reducing the switching time. I refer you to _Principles_of_CMOS_VLSI_Design_ by Neil H. E. Weste and Kamran Eshraghian for further details.
Dave Patterson is now a professor here at Berkeley. He's a great lecturer, one of the best I've had. I took a lower-div architecture class from him last year. It was great being able to learn the rationale behind certain features in the systems he designed, eg. sparc reg windows. He was on the core development teams for riad, sparc, and mips, among other things. Another person who probably belongs on your first list is William Kahan of IEEE 754 fame. He is responsible for the floating point used on every modern processor. Unlike Patterson though, you don't want to take any of his courses. The last course he taught he failed a large majority of the students. His contribution was a lot more important than basic though.
I have the server if you want. P3 550, 256 megs ram, ethernet to the backbone, Linux. I haven't thought of something useful to do with it yet, this may be it.
Nikkel?? Maybe you mean "nickel" or "nickle," either is permissible.
I've never heard of the nickel - Titanium interaction you discribe. The reason titanium should be avoided in tools like screwdriver blades is that Ti is very soft. Most Ti alloys cannot be hardened above Rc 45 or so. Screwdrivers probably are hardened to Rc 55 or 60.