Liquid nitrogen is a lot of fun. You can stick your finger into it without being harmed. You only get hurt by it if you drink it or if you touch water that has been in contact with it. (Water ice cooled by liquid nitrogen is fairly nasty. It sticks.)
Space may be cold but it is a wonderful insulator. The only way to lose heat in space is to radiate it off. Keeping a superconductor cold enough would be a challenge.
The client can only predict the outcome of a task if it knows all the variables. The solution is to ensure that you can only succeed in the game by constantly interacting with other players.
A computer that plays a game must be able to determine whether a given position in the game is better than another given position. It is very hard for the computer to find the best course of action when an important asset for a player is "the yellow player owes me a big favor"...
Athlon sucks for kernel compilation time. The Pentium-II sucked worse, but the Pentium-III has a 256-bit path to second level cache, and gcc/loves/ to have about 128kiB to 256kiB of very fast cache.
The Athlons 64-bit path to second level cache cannot compete. It would be nice if AMD optimized their processors for gcc, but honestly I think SpecWeb and Quake are more important benchmarks for them.
Washington Post probably got a hint from someone... US propaganda has been working for a while to make China look like the next big threat to the US. It seems that the US has to define itself as the shining knight in opposition to the evil monsters, and after the cold war there has been a lack of evil monsters.
I expect relations between China and the US to deteriorate over the next few years. Soon there'll be another cold war -- that's what people in the US know and are comfortable with.
Actually, if you do the math on a hydrogen bomb explosion, it turns out that the radius within which radiation is instantly lethal is significantly smaller than the radius within which heat is instantly lethal. In other words, if you die instantly from the radiation, you would have died anyway at most a few seconds later when the heat wave arrived.
Further away from the explosion the most dangerous thing is the wind force. A shelter could certainly help there. Of course you might die a few years later from cancer, but then again, you might not.
From the Slashdot posts it seems that a lot of California's power at this moment is provided by hydroelectric power plants. It sounds as if those power plants are in use over large periods of the day, possibly all the time, and furthermore they are being drained so they will be unable to produce later in the year.
In that case the hydroelectric plants are not being used efficiently. Wind power would count fully in that scenario -- they would allow the hydroelectric plants to provide more peak power.
At the current interest rate, wind power is profitable in Denmark if you can get a good location for the windmills and you get paid around 7c/kWh.
Unfortunately the western part of Denmark (where most of the good locations are) and the rest of the Northern European market has an overcapacity at the moment, so electricity on the spot market is cheap. On the other hand, the eastern part of Denmark and the Scandinavian market it connects to is fairly close to the capacity limit, so the wind power would do a lot more good there. There are no connections of any size worth mentioning between the two markets. (I have a beautiful explanation why, but unfortunately there isn't enough room in the margin.)
Make the transmission lines DC and run them in the ground instead of up on poles. That seems to make most people happy. And yes, the technology is there to do it at least with 100kV lines. 400kV are AFAIK still a challenge, but it's being worked on.
Oh and while the initial cost is higher, there is less maintenance.
I can tell you how fast you were going. The IC3 trains which you were no doubt using has a top speed of 180km/h. Since you happened to travel on some of the most used tracks in Denmark, you were doing close to that for a large part of the journey. Racing the train is quite a challenge these days.
In the tunnel under Storebaelt you were doing 140km/h. The last part of the journey was on lower quality tracks, typical speeds 100-120km/h.
By the way, the Lyntog which stops at fewer stations can make the journey in ~3h15m.
Benny
Re:Costantly use a conversion table?
on
E=MC
·
· Score: 1
"Code reuse" is easy to sell to managers and the advantages seem obvious to anyone. There is a serious lack of studies proving that extra work done to facilitate code reuse ends up paying itself back on average.
Using object oriented design to structure programs can be a great help. Design is a very personal topic and no set design system can fit everyone. I have seen people who would design things using a functional model, essentially using a scientific method to solve problems. Others, perhaps the author of the article, think primarily in terms of the work that the program needs to do, and are productive that way. Still others are most productive when they design things as collections of objects that interact.
A given programming language tends to fit a specific design paradigm best, but there is nothing stopping you from designing using one paradigm and using a programming language intended for a different one. It is just a matter of knowing what you are doing. The author uses relational database tools as an example. Yes, it sucks to use a relational database if you design object oriented, but since other people already developed the methods needed, it is a SMOP.
So far there have been quite a few cases where the GPL has come close to being tested in court. In every case the alleged GPL-offender has chosen to settle in some way.
It seems to me that the reason for those settlements is not just the fear of losing a court battle, but instead the fear of bad publicity. It is easy to do the whole "Giant corporation against small idealist entrepreneur" thing.
StarOffice is still a Word Processor, but when I first learned to use it I discovered that it creates documents with much more structure than Word does. If you are careful, you can make documents almost as structured as LaTeX documents are.
The three big problems are 1) You cannot actually view the raw codes, 2) It is not possible to use external tools on the documents, and 3) The print output sucks compared to LaTeX.
The first two are fixed by using XML in OpenOffice 6. The third will be fixed when someone does a better renderer. Note that since the document format is XML, you can potentially use any XML renderer on the documents.
Now if only they got printing working on Unix in the beta...
An FSF-EU would arbitrarily exclude countries such as Norway and most of Eastern Europe. It would be contributing to the growing split between the EU and the rest of Europe. Europe has a lot of common history, there is no point in splitting it along the arbitrary borders of EU.
Whether this FSF-Europe is a good thing or not remains to be seen, but an FSF-EU would be terrible.
Re:Steril genetic plants
on
Golden Rice
·
· Score: 1
Sterility does not help. Genes pass from one organism to another across species. Especially among bacteria, but also between plants.
I didn't believe that when I first heard it, but I was researched it and it's an accepted fact.
Re:GM food is not a good idea yet
on
Golden Rice
·
· Score: 1
There was a series of articles in the Danish periodical for engineers, Ingeniøren. It is describing how direct current is the future, and how effect-electronics makes it possible. It is possible to do a DC-to-DC transformation with very little loss.
CO and CO2 are fairly rare in air (especially the former). Increasing the CO level by 5% or 10% compared to the natural level does not change the overall composition of the air much.
If you pour as much (absolute amount, not relative) nitrogen into the air as we now pour CO2 and CO, the atmosphere will maybe consist of 71% nitrogen instead of 70%. Noone will notice.
The water level stays the same when floating ice melts. The ice under the water line admittedly weighs less than a corresponding amount of water, but that is exactly offset by the fact that some of the ice is above the water line.
Try it out at home, it's a fairly easy experiment.
That's a lot of heat used for melting ice instead of being measured as a temperature increase. Which makes the measure temperature increases even more alarming.
Assume I am a user on a system that only has emacs and I really love xemacs. Then I install xemacs using my regular account, "amorsen".
I discover that others need xemacs too, and that they have started using the one I installed. Then I grin evilly and change the xemacs executable to fork() a little X event watching program before it exec()'s the actual xemacs. This way I can gather passwords from unsuspecting users.
Software installation must be done by trusted users, and noone else. Anything else would require an operating system that actually protected users from the programs they run. Such an OS would not be called Unix.
Slashdot Mirror
Liquid nitrogen is a lot of fun. You can stick your finger into it without being harmed. You only get hurt by it if you drink it or if you touch water that has been in contact with it. (Water ice cooled by liquid nitrogen is fairly nasty. It sticks.)
Space may be cold but it is a wonderful insulator. The only way to lose heat in space is to radiate it off. Keeping a superconductor cold enough would be a challenge.
The client can only predict the outcome of a task if it knows all the variables. The solution is to ensure that you can only succeed in the game by constantly interacting with other players. A computer that plays a game must be able to determine whether a given position in the game is better than another given position. It is very hard for the computer to find the best course of action when an important asset for a player is "the yellow player owes me a big favor"...
Athlon sucks for kernel compilation time. The Pentium-II sucked worse, but the Pentium-III has a 256-bit path to second level cache, and gcc /loves/ to have about 128kiB to 256kiB of very fast cache.
The Athlons 64-bit path to second level cache cannot compete. It would be nice if AMD optimized their processors for gcc, but honestly I think SpecWeb and Quake are more important benchmarks for them.
Washington Post probably got a hint from someone... US propaganda has been working for a while to make China look like the next big threat to the US. It seems that the US has to define itself as the shining knight in opposition to the evil monsters, and after the cold war there has been a lack of evil monsters.
I expect relations between China and the US to deteriorate over the next few years. Soon there'll be another cold war -- that's what people in the US know and are comfortable with.
Actually, if you do the math on a hydrogen bomb explosion, it turns out that the radius within which radiation is instantly lethal is significantly smaller than the radius within which heat is instantly lethal. In other words, if you die instantly from the radiation, you would have died anyway at most a few seconds later when the heat wave arrived.
Further away from the explosion the most dangerous thing is the wind force. A shelter could certainly help there. Of course you might die a few years later from cancer, but then again, you might not.
From the Slashdot posts it seems that a lot of California's power at this moment is provided by hydroelectric power plants. It sounds as if those power plants are in use over large periods of the day, possibly all the time, and furthermore they are being drained so they will be unable to produce later in the year.
In that case the hydroelectric plants are not being used efficiently. Wind power would count fully in that scenario -- they would allow the hydroelectric plants to provide more peak power.
At the current interest rate, wind power is profitable in Denmark if you can get a good location for the windmills and you get paid around 7c/kWh.
Unfortunately the western part of Denmark (where most of the good locations are) and the rest of the Northern European market has an overcapacity at the moment, so electricity on the spot market is cheap. On the other hand, the eastern part of Denmark and the Scandinavian market it connects to is fairly close to the capacity limit, so the wind power would do a lot more good there. There are no connections of any size worth mentioning between the two markets. (I have a beautiful explanation why, but unfortunately there isn't enough room in the margin.)
Benny
Make the transmission lines DC and run them in the ground instead of up on poles. That seems to make most people happy. And yes, the technology is there to do it at least with 100kV lines. 400kV are AFAIK still a challenge, but it's being worked on.
Oh and while the initial cost is higher, there is less maintenance.
Benny
I can tell you how fast you were going. The IC3 trains which you were no doubt using has a top speed of 180km/h. Since you happened to travel on some of the most used tracks in Denmark, you were doing close to that for a large part of the journey. Racing the train is quite a challenge these days.
In the tunnel under Storebaelt you were doing 140km/h. The last part of the journey was on lower quality tracks, typical speeds 100-120km/h.
By the way, the Lyntog which stops at fewer stations can make the journey in ~3h15m.
Benny
Ahem, a pound = 2.2 Kg
That's what NASA thought...
"Code reuse" is easy to sell to managers and the advantages seem obvious to anyone. There is a serious lack of studies proving that extra work done to facilitate code reuse ends up paying itself back on average.
Using object oriented design to structure programs can be a great help. Design is a very personal topic and no set design system can fit everyone. I have seen people who would design things using a functional model, essentially using a scientific method to solve problems. Others, perhaps the author of the article, think primarily in terms of the work that the program needs to do, and are productive that way. Still others are most productive when they design things as collections of objects that interact.
A given programming language tends to fit a specific design paradigm best, but there is nothing stopping you from designing using one paradigm and using a programming language intended for a different one. It is just a matter of knowing what you are doing. The author uses relational database tools as an example. Yes, it sucks to use a relational database if you design object oriented, but since other people already developed the methods needed, it is a SMOP.
Netfuture has an article about Golden Rice
If nVidia released specifications for their cards, there would probably be a driver. They don't, so there isn't.
So far there have been quite a few cases where the GPL has come close to being tested in court. In every case the alleged GPL-offender has chosen to settle in some way.
It seems to me that the reason for those settlements is not just the fear of losing a court battle, but instead the fear of bad publicity. It is easy to do the whole "Giant corporation against small idealist entrepreneur" thing.
RedHat 7 versions of nightly builds are available. There are also builds for RH6 there, and those work on at least Suse 6.4.
Benny
StarOffice is still a Word Processor, but when I first learned to use it I discovered that it creates documents with much more structure than Word does. If you are careful, you can make documents almost as structured as LaTeX documents are.
The three big problems are 1) You cannot actually view the raw codes, 2) It is not possible to use external tools on the documents, and 3) The print output sucks compared to LaTeX.
The first two are fixed by using XML in OpenOffice 6. The third will be fixed when someone does a better renderer. Note that since the document format is XML, you can potentially use any XML renderer on the documents.
Now if only they got printing working on Unix in the beta...
An FSF-EU would arbitrarily exclude countries such as Norway and most of Eastern Europe. It would be contributing to the growing split between the EU and the rest of Europe. Europe has a lot of common history, there is no point in splitting it along the arbitrary borders of EU.
Whether this FSF-Europe is a good thing or not remains to be seen, but an FSF-EU would be terrible.
Sterility does not help. Genes pass from one organism to another across species. Especially among bacteria, but also between plants.
I didn't believe that when I first heard it, but I was researched it and it's an accepted fact.
Pennicillin doesn't reproduce.
CO and CO2 are fairly rare in air (especially the former). Increasing the CO level by 5% or 10% compared to the natural level does not change the overall composition of the air much.
If you pour as much (absolute amount, not relative) nitrogen into the air as we now pour CO2 and CO, the atmosphere will maybe consist of 71% nitrogen instead of 70%. Noone will notice.
Benny
The water level stays the same when floating ice melts. The ice under the water line admittedly weighs less than a corresponding amount of water, but that is exactly offset by the fact that some of the ice is above the water line.
Try it out at home, it's a fairly easy experiment.
Benny
That's a lot of heat used for melting ice instead of being measured as a temperature increase. Which makes the measure temperature increases even more alarming.
Benny
Assume I am a user on a system that only has emacs and I really love xemacs. Then I install xemacs using my regular account, "amorsen".
I discover that others need xemacs too, and that they have started using the one I installed. Then I grin evilly and change the xemacs executable to fork() a little X event watching program before it exec()'s the actual xemacs. This way I can gather passwords from unsuspecting users.
Software installation must be done by trusted users, and noone else. Anything else would require an operating system that actually protected users from the programs they run. Such an OS would not be called Unix.
Benny
It's not hard to reverse a patch.
Benny