Your comments are close to two decades over. Today its all objects which defers the issues yet another step away from reality. I agree with your other comments.
Any problem in computer science can be solved with one additional layer of indirection. But that usually will create another problem. -- David Wheeler (of ILLIAC fame, not the others)
What does this say about Safenet's crypto cards that they are trying to push into banking situations? Their crypto card is something like 4 times the price of Sun's.
I've been looking into this since 2001. The biggest real problem is detecting the pulsars. The free space signal loss is on the order of -400db. The math is much harder than dealing with GPS and you have to find the easy way to figure out tick counts. Other than that, its workable. Modern GPS receivers do have methods to remove pulsar noise from the signals they are watching. Defining a coordinate system will be a mess as well but that could lead to a reasonable way to define things for all planets as well. Nearly everything in space rotates and most rotate the same direction in about the same plane so you can define a north and south. The real problem is how do you define the origin of Longitude? I suspect the best way is define it by the highest point on the planet but that leads to problems when the highest point is at the unstable edge of an active volcano. What do you use as a reference for a solar system or a galaxy?
You do it just like GPS receivers get their pseudo range. You start by guessing what time it is and you use that to guess your position then you refine your idea of the current time which gives a better position. Once you have a rough idea of where you are in time and space, you can figure out how many ticks the pulsar has produced since your epoch and feed that back into the Kalman Filter to refine your fix. If you wanted this to work for all time, your Kalman filter would need a parameter for drift of each pulsar's clock. This is not much different than how a GPS receiver deals with Doppler shift from the sats and many use it to tweak the delta X,Y,Z parameters as well. It only gets truly messy when you have to figure in the wobble of a planet that is orbiting a wobbling star. The Navstar GPS system broadcast factors that are linked to Moon, Jupiter and Saturn already. A GPS system can get by with as little as 16 factors in its Kalman filter but a near space pulsar based one would need at least 40 factors of at least 3d polar coordinates.
It turns out that the 70 year extension is worse for artists since the recording companies can now bill the artist more in preservation charges and other made up funny charges. I suspect it will be very difficult to find a single artists that this will help.
Re:Java is safe, mysql is safe...
on
Oracle Buys Sun
·
· Score: 1
I've been running lots of the new T1 and some T2 hardware. Most of it is slower than Sparc IIIi on my work loads. The T1000 with its hardware acceleration of SSL means that its faster than 2 year old x86 hardware only for the 1st 2 to 3k and then its game over for its speed. If you need an example of the new hardware, just look at the sun shop... that site is so slow I wonder if they found some Timex Sinclares to run it.
I like Sparc hardware and have been using it for more than a decade. I used to be sun fanboy. I use it on most of my internet facing gear since it can not run code on a overflowed stack since it uses a dedicated hardware stack that can't ever provide code to the execution engine. Too bad sun has given up on any real development. I figure a re-taped IIIi on 65 or 45nm would blow away the T1 and T2 for most average Unix workloads.
Assume you can hack a major supermarket chain's pin pads. On the 1st day you try everyones code with 0001. On average you will lockout a small number of customers (who lock them selves out all the time anyway). Yet you have one in 10000 odds of hitting any given card and considering that 10,000 to a few million cards have gone through in a day, thats not bad odds. If you don't use 0001, 0002 and use things more people are likely to take like 1074 (oct 74) or 4321 you will find that your hit rate is far higher.
The arcing that some people report when bulbs die is a result of the resistance inside the tube going so high that the easier path is around the bulb. Most manufactures claim their bulbs will produce about 80% of their new light at 2000 hours and that is close for the better bulbs if you discount the 1st 10 hours. My light meter claims the cheap bulbs lose more light output faster. I've also noticed that the current goes up as the cheaper bulbs age as well. One thing to consider is that when you look at a 100W bulb and a 50W bulb, you don't perceive the 50W bulb as being half the brightness thanks to the nonlinear way the eye sees light. Light meters are cheap, power meters are cheap and its easy to do these tests.
If you replace a 100W with a 25W CFL, your not getting the same light. My kitchen has 4 lights and with CFLs, I have to have bulbs in all 4 of them or else it seems dark. Before it would be fine with 2 60W bulbs for a total load at at the power station of about 128VA but 4 CFLs have a grid load of 156VA. Which is more energy efficient? CFLs have a nasty habit of being bright in their 1st 10 hours and then dropping off to about 80% after their 1st 1000 hours. Very old CFLs will produce less light per watt than and incandescent bulb and they can do that for thousands of hours.
That assumes that the unused power has some place to go. if you take a 26 VA load, your grid must provide 26 W even if the device only uses 13W as 26W goes in but 13 comes out leaving the device to use 13. If you have other things that don't care about the funny power (like old style light bulbs), its not a problem. If you end up with 99% of the light bulbs on the grid clicking on at 90V, the generator is going to be unhappy ramping that voltage up and then the current is going to increase a massive amount when all the CFLs decide to turn on for a fraction of a cycle.
Right now the power factor losses just end up heating up the wires.
Most power factor meters just measure lead and lag and give you a number which is useful for motors. The odd loading of the CFL (and LEDs) means its power factor may be much worse than you have seen. There are reports of power factors of the CFLs sold in Australia can be below.20. They are also replacing all the meters here with ones that compensate for bad power factor. Someone has since nice graphs from his oscilloscope where he demonstrates just how bad it is.
I find it nearly impossible to explain power factor to most people so I've been explaining it as an electricity grid efficiency factor. That 100W old school bulb will waste about 7 VA in the grid here yet its replacement CFL will waste nearly 14 VA.
Throw in a few more a) the fire risk when they reach end of life b) their failure modes can often cause them to draw more than 100 watts which means you can't put them in a lamp that can only deal with a 60W heat load c) they produce UV d) recycle-ability e) production energy costs f) mercury content
The UV issue may be very rare but the bulbs are supposed to be made out of leaded glass (which makes them nonrecyclable in most parts of the world) and if the lead content is too low, they leak UV and then make ozone which means even don't ever break even with the green house gas argument.
There are very few people in the world who can properly recycle the office tube florescent bulbs and most of them just end up in landfills. In Australia less than 1% of the bulbs end up recycled. I've heard that Germany is considering banning CFLs due to mercury concerns as they have start finding out the mercury problem is bigger than they thought it would be.
We keep hearing about 5mg of mercury yet references on the web show that people who actually recover mercury can often get double that out of them. Either they cracked alchemy or someone is lying about mercury content.
The US Navy have an under water research station in Guam that sends lots of data back to Hawaii. They used to ask for massive increases in capacity every few years so someone decided to send thousands of fibers that direction. With the newer long range amps, you don't need any under sea repeaters from Hawaii to Guam so you can get massive bandwidth there.
Its also along the route where you run fiber from the US to Japan and Asia.
But Conroy played the part and he must be punished. I for one will donate over a thousand dollars to anyone who opposes him in the next election unless I find a bigger threat. I don't want this crud getting this far again.
Slashdot Mirror
Which brakes? Most rains have several sets... and 400 meters on a train is sort of like 9 feet at high way speeds in a modern car.
Your comments are close to two decades over. Today its all objects which defers the issues yet another step away from reality.
I agree with your other comments.
Any problem in computer science can be solved with one additional layer of indirection. But that usually will create another problem. -- David Wheeler (of ILLIAC fame, not the others)
What does this say about Safenet's crypto cards that they are trying to push into banking situations? Their crypto card is something like 4 times the price of Sun's.
Your doctor should have a big thick book that describes most drugs and the ratios of common and even very rare side effects.
The interlaced switch and loop is called Duffs Device.
The 1st footprint went away within a few hours of its imprint since it was at the bottom of the only ladder to the LM.
I've been looking into this since 2001. The biggest real problem is detecting the pulsars. The free space signal loss is on the order of -400db. The math is much harder than dealing with GPS and you have to find the easy way to figure out tick counts. Other than that, its workable. Modern GPS receivers do have methods to remove pulsar noise from the signals they are watching. Defining a coordinate system will be a mess as well but that could lead to a reasonable way to define things for all planets as well. Nearly everything in space rotates and most rotate the same direction in about the same plane so you can define a north and south. The real problem is how do you define the origin of Longitude? I suspect the best way is define it by the highest point on the planet but that leads to problems when the highest point is at the unstable edge of an active volcano. What do you use as a reference for a solar system or a galaxy?
You do it just like GPS receivers get their pseudo range. You start by guessing what time it is and you use that to guess your position then you refine your idea of the current time which gives a better position. Once you have a rough idea of where you are in time and space, you can figure out how many ticks the pulsar has produced since your epoch and feed that back into the Kalman Filter to refine your fix. If you wanted this to work for all time, your Kalman filter would need a parameter for drift of each pulsar's clock. This is not much different than how a GPS receiver deals with Doppler shift from the sats and many use it to tweak the delta X,Y,Z parameters as well. It only gets truly messy when you have to figure in the wobble of a planet that is orbiting a wobbling star. The Navstar GPS system broadcast factors that are linked to Moon, Jupiter and Saturn already. A GPS system can get by with as little as 16 factors in its Kalman filter but a near space pulsar based one would need at least 40 factors of at least 3d polar coordinates.
Someone had to make the Highlander II for the Trek universe. Now we know who that man was.
The tires on your car will have a speed rating so yes it is a valid argument.
Was it cheaper when they did it all with paper files?
So are you going to trademark your kid?
It turns out that the 70 year extension is worse for artists since the recording companies can now bill the artist more in preservation charges and other made up funny charges. I suspect it will be very difficult to find a single artists that this will help.
I've been running lots of the new T1 and some T2 hardware. Most of it is slower than Sparc IIIi on my work loads. The T1000 with its hardware acceleration of SSL means that its faster than 2 year old x86 hardware only for the 1st 2 to 3k and then its game over for its speed. If you need an example of the new hardware, just look at the sun shop... that site is so slow I wonder if they found some Timex Sinclares to run it.
I like Sparc hardware and have been using it for more than a decade. I used to be sun fanboy. I use it on most of my internet facing gear since it can not run code on a overflowed stack since it uses a dedicated hardware stack that can't ever provide code to the execution engine. Too bad sun has given up on any real development. I figure a re-taped IIIi on 65 or 45nm would blow away the T1 and T2 for most average Unix workloads.
google "CFL end of life"
Assume you can hack a major supermarket chain's pin pads. On the 1st day you try everyones code with 0001. On average you will lockout a small number of customers (who lock them selves out all the time anyway). Yet you have one in 10000 odds of hitting any given card and considering that 10,000 to a few million cards have gone through in a day, thats not bad odds. If you don't use 0001, 0002 and use things more people are likely to take like 1074 (oct 74) or 4321 you will find that your hit rate is far higher.
The arcing that some people report when bulbs die is a result of the resistance inside the tube going so high that the easier path is around the bulb. Most manufactures claim their bulbs will produce about 80% of their new light at 2000 hours and that is close for the better bulbs if you discount the 1st 10 hours. My light meter claims the cheap bulbs lose more light output faster. I've also noticed that the current goes up as the cheaper bulbs age as well. One thing to consider is that when you look at a 100W bulb and a 50W bulb, you don't perceive the 50W bulb as being half the brightness thanks to the nonlinear way the eye sees light. Light meters are cheap, power meters are cheap and its easy to do these tests.
If you replace a 100W with a 25W CFL, your not getting the same light. My kitchen has 4 lights and with CFLs, I have to have bulbs in all 4 of them or else it seems dark. Before it would be fine with 2 60W bulbs for a total load at at the power station of about 128VA but 4 CFLs have a grid load of 156VA. Which is more energy efficient? CFLs have a nasty habit of being bright in their 1st 10 hours and then dropping off to about 80% after their 1st 1000 hours. Very old CFLs will produce less light per watt than and incandescent bulb and they can do that for thousands of hours.
That assumes that the unused power has some place to go. if you take a 26 VA load, your grid must provide 26 W even if the device only uses 13W as 26W goes in but 13 comes out leaving the device to use 13. If you have other things that don't care about the funny power (like old style light bulbs), its not a problem. If you end up with 99% of the light bulbs on the grid clicking on at 90V, the generator is going to be unhappy ramping that voltage up and then the current is going to increase a massive amount when all the CFLs decide to turn on for a fraction of a cycle.
Right now the power factor losses just end up heating up the wires.
Most power factor meters just measure lead and lag and give you a number which is useful for motors. The odd loading of the CFL (and LEDs) means its power factor may be much worse than you have seen. There are reports of power factors of the CFLs sold in Australia can be below .20. They are also replacing all the meters here with ones that compensate for bad power factor. Someone has since nice graphs from his oscilloscope where he demonstrates just how bad it is.
I find it nearly impossible to explain power factor to most people so I've been explaining it as an electricity grid efficiency factor. That 100W old school bulb will waste about 7 VA in the grid here yet its replacement CFL will waste nearly 14 VA.
Throw in a few more
a) the fire risk when they reach end of life
b) their failure modes can often cause them to draw more than 100 watts which means you can't put them in a lamp that can only deal with a 60W heat load
c) they produce UV
d) recycle-ability
e) production energy costs
f) mercury content
The UV issue may be very rare but the bulbs are supposed to be made out of leaded glass (which makes them nonrecyclable in most parts of the world) and if the lead content is too low, they leak UV and then make ozone which means even don't ever break even with the green house gas argument.
There are very few people in the world who can properly recycle the office tube florescent bulbs and most of them just end up in landfills. In Australia less than 1% of the bulbs end up recycled. I've heard that Germany is considering banning CFLs due to mercury concerns as they have start finding out the mercury problem is bigger than they thought it would be.
We keep hearing about 5mg of mercury yet references on the web show that people who actually recover mercury can often get double that out of them. Either they cracked alchemy or someone is lying about mercury content.
The could have gone with land only based repeaters but they didn't. Someone could duplicate their link for a small fraction of the cash they spent.
The US Navy have an under water research station in Guam that sends lots of data back to Hawaii. They used to ask for massive increases in capacity every few years so someone decided to send thousands of fibers that direction. With the newer long range amps, you don't need any under sea repeaters from Hawaii to Guam so you can get massive bandwidth there.
Its also along the route where you run fiber from the US to Japan and Asia.
But Conroy played the part and he must be punished. I for one will donate over a thousand dollars to anyone who opposes him in the next election unless I find a bigger threat. I don't want this crud getting this far again.
And I don't think its over yet.
You don't do backups, you do redundancy. If your IT model depends on one server doing its thing right, you have already gone down the path the doom.