IANAL, of course, but I have some experience in these areas.
1. Can I do it with Linux today (GPL2) and tomorrow (GPL3)?
Yes, of course. However if GPL licensed binaries are distributed with it, you must provide source for those binaries for a period of no less than 3 years longer than you distribute the device or application. This, of course, applies to GPL licensed binaries even if they are running on a BSD kernel. This provision applies whether you have modified the code or not. You can rely on a third party to distribute the source code. However, if that third party ceases distribution of the source before the 3 years are up, you are still obligated to distribute the source.
So, yes, if you distribute a linux kernel that you downloaded from "xyzzyplugh.com" you can direct people to "xyzzyplugh.com" for sources. But if "xyzzyplugh.com" goes out of business, you still need to be able to provide the sources.
The same is true if you use a BSD kernel, but include the GPL application "spork" embedded in the device. You need to be able to provide the sources to "spork" regardless of where you obtained the "spork" binary and regardless of whether the "spork" sources are available elsewhere.
2. Can I statically link the code with Linux libraries? (My own experience shows that dynamic linking is too much to bear.)
No. If you statically link the Linux libraries, or any GPL licensed library, then you are obligated to license your application under the GPL. If you statically link a LGPL licensed binary, your application can remain closed source. However you must make source for the LGPL licensed library available, again for at least 3 years longer than you distribute the application.
3. Can I obfuscate my code (e.g. encode it)?
Source or object code? In either case the answer is "of course" for all licenses. But if you are under GPL or LGPL, you need to make the means to decrypt it available to anyone who wants it.
4. Could I be forced to publish this code by some 3-d party?
No, but you could be required to cease distribution and potentially be penalized in either civil or criminal court for any copyright violations you commit. This could apply regardless of the licence of the OS kernel you use. Be sure to understand the license of any software or libraries that you distribute.
5. Am I correct that programming in and selling BSD-based boxes won't raise any of the above problems?
No, you are not. Not all code in a BSD distribution is necessarily licensed under the BSD. Nor are you protected from anyone who claims that their code was improperly included in a BSD licensed application. Worse yet, there are potential patent violations in just about any device you might want to sell.
I only have one follow-on question. Why are you so afraid of releasing your sources? What's the worst that could happen?
It's not the $2000 each in legal contributions that these folks got from the RIAA that buys votes. The minuscule legal contributions are not the problem. It's the additional few million that is laundered through the parties, other candidates, other donors and the PACs that buys influence. That, and the under the table "perks" that lawmakers get. I'm sure many of the people who got gobs of RIAA money aren't on this list.
This would drive radio stations to playing non-RIAA music, accelerating the RIAA's demise.
But that's not how it would work. RIAA claims it has the right to collect royalties for non-RIAA artists, and would certainly write its contracts with radio stations in such a way that they could either pay a flat fee, or a per song (any song, RIAA or not) fee.
Any radio station refusing to pay on the grounds that they don't play RIAA affiliated artists would be taken to court and would need to provide records of everything they have played in order to prove that they haven't. That will cost them more than agreeing to pay the protection money.
I would be nice if someone in the justice system had the balls to stand up to them. We used to have laws against extortion. Then again we used to have laws against bribery. I guess they are both "protected speech" these days.
There's a very simple formula for determining whether a relationship is socially acceptable in the USA. Half your age plus seven. Simple, easy, and no sexism involved.
If the younger person is under 18 take half the age of the older person and add seven. Here are the the results.
It is acceptable for a 20 year old to have a sexual relationship with a 17 year old.
It is not acceptable for a 20 year old to have a sexual relationship with a 15 year old. But it is acceptable for a 16 year old to have a relationship with a 15 year old.
If two 14 year olds want to play doctor that's fine.
It is unacceptable for anyone of any age to have sex with a 13 year old.
Sure, there's still a dividing line, but it beats the current system in some states where overzealous prosecutors will arrest a boy on his 18th birthday for having sex with his 17 year old girlfriend. And at least this way once a relationship is legal, it's always legal.
"The formula" works for adults as well, but we can't and shouldn't regulate the sexual behviour of consenting adults. However, if you're 40 and dating a 20 year old, it's going to raise eyebrows. Dating a 27 year old, not so much.
Been using the fluorescent light bulbs for a few years now, and while they use less power, the do cost more, and unlike what is promised on the package, they need replacement as often as the incandescent kind, in my experience, so probably a wash in terms of lifetime cost.
I think there's probably something wrong with how you are using them. Enclosed fixtures? Recessed lighting?
I've got of order 20 compact fluorescents in the house. Since January, I have replaced 2. That's consistent with a 4 year lifespan. All of the ones that needed replacement were in enclosed fixtures. In open fixtures, I don't think I have ever replaced one that failed. I have replaced older ones that are getting dim or that don't match the lighting color of the newer ones.
The are more expensive than incandescent, but in California they are heavily subsidised. The local drugstore has GE 60W equivalents at 3 for $3.99. The incandescents are 3 for $2.99.
"Very low pollution. Most FABs emit water cleaner than they take it in. LEDs can be produced lead free, and indium arsenic levels are exceptionally low."
What about levels of Cerium, Terbium, and Erbium? It has to come from somewhere and go to somewhere..... And where are they putting the waste that they are extracting from the water to make it cleaner?
Everyone keeps confusing bandwidth with latency. The bandwidth is the amount of time it takes the data to travel its own length. The latency is the time it takes for the data to get from source to destination.
At 0.85 mach, a A380 travels its own length in about 1/4 second. So the bandwidth of a A380 is 6720 Pbps. You only need 881,832 A380s to maintain that bandwidth over a 20000 mile course. How to get a 150 ton payload onto or off of an A380 in 0.25 seconds is left as an excercise for the network engineer.
Hmmm, let's see: Let's have maximum capacity DVD's at 9GB and for the sake of this exercise let's say the station wagon's capacity is 1000 DVDs so we have 9000GB moving around. Let's say the 20,000 mile distance will be covered at top speed (breaking speed limits in all states) at 100miles/h that results in 200 hours of deliverance time so:
You are mixing up latency with bandwidth. The latency (round trip time) of the connection here is 400 hours. The bandwidth (i.e. data rate) is the amount of data divided by the time it takes for the data to travel its own length.
At 100 mph, a station wagon will travel its length in 0.14 seconds. So the bandwidth of a stationwagon packed with 9000 GB of data is about 550 Tbps.
Given a train of station wagons running at 100mph, you could sustain that. Of course with 1440000000 ms ping times, I wouldn't try playing Battlefield 2 over that connection.
Seriously, the distinction is important. If you included transit time when calculating bandwidth, the theoretical maximum bandwidth for a 12,000 bit packet on a 20,000 mile path would be 112 kbps.
China is testing 75 million students for 2 million university slots. The UK is testing 5 million students for a million university slots. The difference between the 80th percentile and the 97th percentile is pretty significant, and has very little to do with the quality of the primary and secondary education systems in either country.
But please ignore this, and proceed being alarmed. It's certainly easier than thinking.
If the government isn't going to pay 100% of the cost of the research, yes. My last research project was funded at 80% of cost. Where do they expect the other 20% to come from if we can't profit from our research?
Consider yourself lucky. You've hit the gravy train. In Astronomy, you're lucky if you can cover 50% of a research project off of one grant. The typical annual NSF astronomy research grant is just barely enough to cover a grad student's salary plus overhead. A typical non-faculty research scientist needs 5 or more active grants to cover their own salary because if there is more than 20% salary support for a non-student scientist the grant will get shot down for spending too much on "senior personnel."
That's one reason for the link in my signature.
And no, I don't believe software should be patentable. Nor should human genes or cell lines, or anything else capable of reproducing itself. Propogating a clipping from a rose is not an intelectual property violation.
Your DRAM in a PC is essentially an L3 cache. Your disk, an L4. With todays CPU's hitting 90%+ L1 cache hits, and 85% L2 cache hits what they've done is double the speed of 15% of your cache misses. BFD. Net overall system performance increase is maybe 5% depending upon your application.
A little math helps a lot.
Let's assume 1 cycle L1 and a 2 cycle access to a 64-byte/line L2 cache. So an L1 cache miss costs 8 cycles. So if you had 90% hit rate on L1, and 100% hit on L2, your processor will spend 8*0.1/(8*0.1+1*0.9) or 47% of its time waiting on cache misses. (Hyperthreading helps this a bit because the other thread might be able to work at this point). To simplify the next step, we'll say that this averages to an even 2 cycles per access.
Now lets say that the miss rate on L2 is 15% and that a line fill costs you 100 cycles. So the processor spends 100*0.15/(100*0.15+2*0.85) or 90% of its time waiting for main memory. I know some people don't want to hear it, but this is VERY VERY TYPICAL for most applications. Now double the memory speed. You haven't really changed the access time, but you have improved the time to get the second eight words chunk. So now maybe you have cut that 100 cycles to 75 cycles. You've increased the speed of the application by 0.9*25% or 22.5%
Every programmer should know an assembly language. It's a good start, but it's not nearly enough.
They should also learn when not to use it.
If they are programming for a cached memory architecture, they should know that reordering memory accesses can be more important than using assembly.
If they use GCC should know how to use the full spectrum of inline assembler syntax. (i.e. let the compiler allocate the registers).
If they program for x86 and compatibles, they should understand why their optimal Pentium 4 code isn't optimal Pentium M code.
They need to know optimal programming isn't just about learning tricks. Some trick are good (i.e. Carmack's inverse square root). Some are pointless, dangerous, or both (most examples of recursion, XOR swap).
Learning PPC assembly isn't going to help them figure out PIC programming. Learning PIC programming might help them learn PPC assembly.
If it's that good an optimization, put it in the compiler. Everyone should learn to modify the code generation of their favorite open source compiler.
If it might kill someone, do it right even if that means its not optimized.
Having your job classified as exempt from FLSA laws carries with it a certain status, though. Employees like to be "salaried," and not have to fill out an hourly timesheet, even if filling out a timesheet means the occasional opportunity for overtime.
Having never seen a company that would actually let an employee fill in overtime on their timesheet, I wouldn't whether it's better to be hourly or salaried. There seems to be little difference in practice.
Regardless of the law, at most companies it seems understood that overtime will be compensated with comp time, and that the comp time will never actually materialize. Complaints about the system may be given at your exit interview.
Um...to be clear, radiation includes "light", IR and UV components as well.
I don't know the full effects of increased higher-energy photon radiation hitting earth's surface would be, but it seems like most would be converted to lower-energy IR upon hitting the ground. I.e. -- it should, cause some rise in temperature. Whether or not it is significant, I can't say, but I don't believe it to be inconsequential.
Ok. Let me do a little math in my head. The sun emits about 1% of its radiation at wavelengths shorter than 2500 angstroms, and 16% at wavelengths above 13000 angstroms. Lets say the current UV transmission is 0, and that the average IR transmission is 50%. The visible albedo of the earth is about 0.39, so of the light hitting the earth 55.5% actually makes it to the ground.
Lets assume that the atmosphere becomes transparent to UV and the UV albedo is zero. Now the fraction of the sunlight hitting the earth is 56.5%. So the temperature will increase by about the 1/4 power of 56.5/55.5 which is 0.45%. Multiply by the average temperature of the earth (about 275K) and you get 1.2C.
That's definitely not inconsequential. It's also definitely not what's happening. (Sorry to get your hopes up) That's because the ozone hole only covers a small fraction of the lighted area of the earth and even under the hole the atmosphere isn't UV transparent. Looking around at some resources I see that at its worst, atmospheric transmission between 2000 and 2500 angstroms increased by 5%. So the absorbed flux didn't go from 55.5% to 56.5%, it went from 55.5% to 55.54%. Unfortunately that only translates to a mere 0.05C, which is inconsequential.
I won't repeat the calculation for man's power usage. Let's just say 0.04% of the sunlight hitting the earth is enormous compared to how much power people use.
1) the cosmic ray particles are involved in cloud formation which is a massive blocking agent for total radiant solar energy.
Call me back when you learn some science. The earth's magnetic field does not signifcantly shield against cosmic rays. Cosmic rays do not come from the sun.
2) the southern hemisphere of jupiter is regional?
The upwelling zone around 37 degrees south is certainly regional.
3) I believe NASA has just reported finding enough water ice to flood the planet to an average depth of 10 to 20 meters down there in the southern regions using subsurface radar.
And what does this have to do with our discussion? Absolutely nothing.
4) residence time is shorter eh? Didn't realize plants and plankton preferred breathing in methane to co2.
Again, learn some science. Maybe you haven't noticed, but methane is very reactive in the presence of molecular oxygen. It turns into water and CO2 relatively quickly.
5) So now you're blaming ancient ice ages on modern people! Must be some sort of reverse causality thing going on here.
Read it again, idjit. I'm saying that the existence of ice ages does not have any bearing on whether we are altering the climate.
Hate to mention this to a SETI true believer but reality sucks.
I'm a scientist, asshat. I don't "believe" anything about SETI. There is a possibility that extraterrestrial intelligences exist. I think that it is worth the effort to look, even if the results are negative. Belief has nothing to do with it. You know nothing of any work I have done regarding the development of intelligence or any estimates I may have made of the number of communicating civilizations that may exist in the Galaxy.
And I think I know a bit more about the history of the solar system than you do. Then again, so do most fifth graders.
Slashdot Mirror
Not until after you feed it this kitten.
So, yes, if you distribute a linux kernel that you downloaded from "xyzzyplugh.com" you can direct people to "xyzzyplugh.com" for sources. But if "xyzzyplugh.com" goes out of business, you still need to be able to provide the sources.
The same is true if you use a BSD kernel, but include the GPL application "spork" embedded in the device. You need to be able to provide the sources to "spork" regardless of where you obtained the "spork" binary and regardless of whether the "spork" sources are available elsewhere.
No. If you statically link the Linux libraries, or any GPL licensed library, then you are obligated to license your application under the GPL. If you statically link a LGPL licensed binary, your application can remain closed source. However you must make source for the LGPL licensed library available, again for at least 3 years longer than you distribute the application. Source or object code? In either case the answer is "of course" for all licenses. But if you are under GPL or LGPL, you need to make the means to decrypt it available to anyone who wants it. No, but you could be required to cease distribution and potentially be penalized in either civil or criminal court for any copyright violations you commit. This could apply regardless of the licence of the OS kernel you use. Be sure to understand the license of any software or libraries that you distribute.Flamebait? Gee, I should know better than to make jokes.
Or is Linus turning into a replica of RMS, only without the manners?
It's not the $2000 each in legal contributions that these folks got from the RIAA that buys votes. The minuscule legal contributions are not the problem. It's the additional few million that is laundered through the parties, other candidates, other donors and the PACs that buys influence. That, and the under the table "perks" that lawmakers get. I'm sure many of the people who got gobs of RIAA money aren't on this list.
But that's not how it would work. RIAA claims it has the right to collect royalties for non-RIAA artists, and would certainly write its contracts with radio stations in such a way that they could either pay a flat fee, or a per song (any song, RIAA or not) fee.
Any radio station refusing to pay on the grounds that they don't play RIAA affiliated artists would be taken to court and would need to provide records of everything they have played in order to prove that they haven't. That will cost them more than agreeing to pay the protection money.
I would be nice if someone in the justice system had the balls to stand up to them. We used to have laws against extortion. Then again we used to have laws against bribery. I guess they are both "protected speech" these days.
If the younger person is under 18 take half the age of the older person and add seven. Here are the the results.
- It is acceptable for a 20 year old to have a sexual relationship with a 17 year old.
- It is not acceptable for a 20 year old to have a sexual relationship with a 15 year old. But it is acceptable for a 16 year old to have a relationship with a 15 year old.
- If two 14 year olds want to play doctor that's fine.
- It is unacceptable for anyone of any age to have sex with a 13 year old.
Sure, there's still a dividing line, but it beats the current system in some states where overzealous prosecutors will arrest a boy on his 18th birthday for having sex with his 17 year old girlfriend. And at least this way once a relationship is legal, it's always legal."The formula" works for adults as well, but we can't and shouldn't regulate the sexual behviour of consenting adults. However, if you're 40 and dating a 20 year old, it's going to raise eyebrows. Dating a 27 year old, not so much.
The village has already burned to the ground except for they mayor's house. The fire is heading towards the grain silo. Which should we save?
I think there's probably something wrong with how you are using them. Enclosed fixtures? Recessed lighting?
I've got of order 20 compact fluorescents in the house. Since January, I have replaced 2. That's consistent with a 4 year lifespan. All of the ones that needed replacement were in enclosed fixtures. In open fixtures, I don't think I have ever replaced one that failed. I have replaced older ones that are getting dim or that don't match the lighting color of the newer ones.
The are more expensive than incandescent, but in California they are heavily subsidised. The local drugstore has GE 60W equivalents at 3 for $3.99. The incandescents are 3 for $2.99.
What about levels of Cerium, Terbium, and Erbium? It has to come from somewhere and go to somewhere..... And where are they putting the waste that they are extracting from the water to make it cleaner?
At 0.85 mach, a A380 travels its own length in about 1/4 second. So the bandwidth of a A380 is 6720 Pbps. You only need 881,832 A380s to maintain that bandwidth over a 20000 mile course. How to get a 150 ton payload onto or off of an A380 in 0.25 seconds is left as an excercise for the network engineer.
You are mixing up latency with bandwidth. The latency (round trip time) of the connection here is 400 hours. The bandwidth (i.e. data rate) is the amount of data divided by the time it takes for the data to travel its own length.
At 100 mph, a station wagon will travel its length in 0.14 seconds. So the bandwidth of a stationwagon packed with 9000 GB of data is about 550 Tbps.
Given a train of station wagons running at 100mph, you could sustain that. Of course with 1440000000 ms ping times, I wouldn't try playing Battlefield 2 over that connection.
Seriously, the distinction is important. If you included transit time when calculating bandwidth, the theoretical maximum bandwidth for a 12,000 bit packet on a 20,000 mile path would be 112 kbps.
I think you should be limited to saying it for no more than 5 years.
But please ignore this, and proceed being alarmed. It's certainly easier than thinking.
Frankly, I'll wait until someone come up with a way to play Wii games on my Gamecube. IMHO, there's no use spending a cent on last decades technology.
Someone listening to my typing could match my timing well enough to get in if they also knew the password.
I know people like to think that awful acts like this are something new. As it turns out massacres aren't new.
Consider yourself lucky. You've hit the gravy train. In Astronomy, you're lucky if you can cover 50% of a research project off of one grant. The typical annual NSF astronomy research grant is just barely enough to cover a grad student's salary plus overhead. A typical non-faculty research scientist needs 5 or more active grants to cover their own salary because if there is more than 20% salary support for a non-student scientist the grant will get shot down for spending too much on "senior personnel."
That's one reason for the link in my signature.
And no, I don't believe software should be patentable. Nor should human genes or cell lines, or anything else capable of reproducing itself. Propogating a clipping from a rose is not an intelectual property violation.
Except you've forgotten that those 15% of the memory accesses take 90% of the time.
A little math helps a lot.
Let's assume 1 cycle L1 and a 2 cycle access to a 64-byte/line L2 cache. So an L1 cache miss costs 8 cycles. So if you had 90% hit rate on L1, and 100% hit on L2, your processor will spend 8*0.1/(8*0.1+1*0.9) or 47% of its time waiting on cache misses. (Hyperthreading helps this a bit because the other thread might be able to work at this point). To simplify the next step, we'll say that this averages to an even 2 cycles per access.
Now lets say that the miss rate on L2 is 15% and that a line fill costs you 100 cycles. So the processor spends 100*0.15/(100*0.15+2*0.85) or 90% of its time waiting for main memory. I know some people don't want to hear it, but this is VERY VERY TYPICAL for most applications. Now double the memory speed. You haven't really changed the access time, but you have improved the time to get the second eight words chunk. So now maybe you have cut that 100 cycles to 75 cycles. You've increased the speed of the application by 0.9*25% or 22.5%
Eric
Every programmer should know an assembly language. It's a good start, but it's not nearly enough.
Regardless of the law, at most companies it seems understood that overtime will be compensated with comp time, and that the comp time will never actually materialize. Complaints about the system may be given at your exit interview.
Lets assume that the atmosphere becomes transparent to UV and the UV albedo is zero. Now the fraction of the sunlight hitting the earth is 56.5%. So the temperature will increase by about the 1/4 power of 56.5/55.5 which is 0.45%. Multiply by the average temperature of the earth (about 275K) and you get 1.2C.
That's definitely not inconsequential. It's also definitely not what's happening. (Sorry to get your hopes up) That's because the ozone hole only covers a small fraction of the lighted area of the earth and even under the hole the atmosphere isn't UV transparent. Looking around at some resources I see that at its worst, atmospheric transmission between 2000 and 2500 angstroms increased by 5%. So the absorbed flux didn't go from 55.5% to 56.5%, it went from 55.5% to 55.54%. Unfortunately that only translates to a mere 0.05C, which is inconsequential.
I won't repeat the calculation for man's power usage. Let's just say 0.04% of the sunlight hitting the earth is enormous compared to how much power people use.
Call me back when you learn some science. The earth's magnetic field does not signifcantly shield against cosmic rays. Cosmic rays do not come from the sun.
The upwelling zone around 37 degrees south is certainly regional. And what does this have to do with our discussion? Absolutely nothing. Again, learn some science. Maybe you haven't noticed, but methane is very reactive in the presence of molecular oxygen. It turns into water and CO2 relatively quickly. Read it again, idjit. I'm saying that the existence of ice ages does not have any bearing on whether we are altering the climate. I'm a scientist, asshat. I don't "believe" anything about SETI. There is a possibility that extraterrestrial intelligences exist. I think that it is worth the effort to look, even if the results are negative. Belief has nothing to do with it. You know nothing of any work I have done regarding the development of intelligence or any estimates I may have made of the number of communicating civilizations that may exist in the Galaxy.And I think I know a bit more about the history of the solar system than you do. Then again, so do most fifth graders.