This is prolly a good book and all but get real people, computers are just tools and the audience this book was intended for knows this.
Really? How many programmer's really know what P and NP mean and imply? Not all are computer scientists.
How many programmer's know that there is no general algorithm to detect infinite loops in a program? (It can be done for a specific program on specific hardware, but takes an enormous amount of memory. Basically you store the entire system state at each step and check for a repeat.)
I wouldn't say a lifetime, but the market is so big that even with Fab 30 AMD can only supply somwhere between 30 and 40% of the PC market. So, it will be quite a while before AMD "beats" Intel, requiring adding 1 or 2 more Fabs. the important thing is that they remain a viable and profitable competitor with the 20-30% marketshare capability as a threat to Intel.
Assume the code being patented is contained in my body. Legally, the patent would make it illegal for me to be alive, because my body would be using the patented gene to make proteins. What if they patented a blood protein. I could now no longer donate blood because the use of the gene is patented by some corporation, and that corporation must be compensated for that donation. So, it seems far fetched, but that is what the letter of the law says.
It could also be argued that because the gene is part of my body, it is my property, therefore granting exclusive rights to use that gene to someone amounts to the illegal seizure of my property. The key point here is that patenting genetic code discovered through the sequencing of human DNA implies the human DNA came from a person or group of people. The only argument that needs to be made is that a person legally owns their own DNA. Once that assertion is made, there is no justification for patents on human DNA sequences.
The only problem is that this argument might work for human DNA, but it won't invalidate other animal or plant DNA patents.
Note: this only applies to copyright law. Patent law on the other hand is a completely different issue although, i believe it should follow the same rules to some extent. I believe patents on genetic codes that occur in nature should be unenforceable since they are simply fact, and regardless of the effort required to decode the fact that my Y chromosome has the sequence hcctgaaggth should not be patentable.
The problem is that under current copyright law facts are not owned by anyone even if you did the work to acquire those facts they are nto owned by you. Using the original poster's example of rainfall data.
If he goe out and collects the data and publishes it to a database, copyright law would say he owns the data base, and protect that ownership. The thing is that he doesn't own the facts in the database. Some one else could use the facts in the database (i.e. each piece of rainfall data) and put it into there own compilation. And, the original owner will have no legal protection. While ethically the original source of the facts must be sited, there original author does not have to be compensated for the use of the data in another work.
This is a good thing. Facts should always be public domain regardless of who did the work to get those facts. In the US this has been established by legal precedent and copyright law. The proposed change to the law would increase the costs of research and development because suddenly every researcher would have to pay for basic facts, and could not legally compile there own database without either paying the original source of the facts, or collecting those same facts all over again.
I don't think the implicaions of this law are understood completely. Current copyright law through various legal precedents grants copyright protection to the format of a collection of data. the classic example is a phone book. It also only applies to the exact organization if that organization is not obvious.
The classic example is a phonebook. A phonebook is a collection of data i.e. names, phone numbers, and addresses. Organized in alphabetical order. As it turns out under current copyright law this has minimal protection. Alphabetical ordering is obvious, and the rest of the directory is information which by law is publlic domain and not protected by copyright.
A law protecting databases and their content could easily extend to a copyright on information. Basically, a database should be covered just like a phone book. Any content in the database would be owned by the creator of that content, but any information would have to continue to be public domain.
Basically, this means that the databases of internet search engines can be extracted and reorganized into a new database, simply because URL and page titles are information and therefore are not and should not be protected.
Dastardly
P.S. Arguably a page title could be considered the property of the creator of the original, but the URL is really public domain information and not protected by copyright.
A more likely cause of a computer shutdown tonight is probably some drunk driver ramming a nearby power pole and shutting power to the entire building (and region). I would place that as many times more likely than a Y2K glitch. Dastardly
You are mixing up validation and testing. Validation parts never ship to any customer, or at best ship as engineering samples only. These are the parts that boot operating systems, and get worked over by the engineers to make sure the design or production process was not screwed up. Processor testing for customer ship is a completely different process. It starts by testing the processors while they are still on the wafer because packaging is costly, therefore you want to only package parts that can be sold. Therefore, you want to find out speed too, since a 200MHz part isn't worth packaging these days. This involves a lot more than simulating booting, and only takes a few seconds. Once the processors are packaged they are tested again at temperature. The temperatures are not that much higher than what the high end of the junction temperature spec is. Cushion comes because the test developers have guardbanded the tests in order to guarantee the part works as labeled. Package test is where parts are binned and labeled for sale. Again this only takes a few seconds. Your processor never sees the inside of a computer until the OEM puts it in, or you do it yourself! The rumours I have seen on the net so far say it is a test escape. Basically, a failure mode came up that their testing couldn't catch. So, they have probably fixed the test by now, and are already shipping all good parts again. This sort of thing is can happen to any processor manufacturer since it is impossible to test all possible combinations of data. Just to try every combination for 64 data pins would require be O(64!). Or, longer than the age of the universe per processor.
It is interesting to note what ease-of-use seems to mean every time it comes up.
1) Limit choice in order to limit complexity. 2) If you are used to it, it must be easy to use. (i.e. the UI should be just like Windows, or Mac)
Both of which make things easy to use, but are really the lazy way to accomplish ease of use. Maybe a research project shoudl be done to figure out just what ease of use would really be. It would require a lot of work and vounteers and controls. You would have to take a Unix guru, Unix user, MAC guru, MAC user, Win guru, Win user, some one who just knows how to get into AOL and start Word, and finally some one who has littl eexperience with computers. If enough data points were acquired it shoudn't be to hard to figure out what ease of use really is.
Pretty much the whole measuring delay by clock cycles is pretty ambiguous. The correct way is to use measure the latency in ns, then at least you can compare across clock speeds.
It is interesting to note that if you look sat the first cycle delays in DRAM from FPM to EDO to SDRAM they are pretty consistent. 60ns FPM was 5-3-3-3 at 66MHZ and 70ns EDO is 5-2-2-2 (4-2-2-2 at 60ns) at 66MHZ, at 100MHZ SDRAM is 5-1-1-1-3(2)-1-1-1. Note the middle number is the CAS we hear about not the first one. So, first cycle number has been from 70-50ns from FPM to SDRAM, but that can be attributed to process technology improvement, there really hasn't been an architectural improvement in latency.
You also have to specify which latency you are talking about especially with SDRAM and DRDRAM because the latency depends a lot on the access pattern.
So, before we get inot arguments, let's make sure there is a consistent measurement. Just one point DRDRAM has numberous latency numbers depending on what is being accessed and when.
I don't see any possibel way DRDRAM could be cheaper than (DDR)SDRAM on the same process given similar economies of scale. The extra die area of the DRDRAM alone kills that, then combine it with royalties on the chips, RIMMS, and chipsets, and it is absolutely impossible for DRDRAM to cost less tham SDRAM all other things being equal.
Also, JEDEC is working on memory that will be faster than DDR-SDRAM leveraging on work from the SLDRAM group, and going even beyond that. And, the JEDEC designs don't carry the baggage that DRDRAM carries.
MCA vs ISA - MCA was prprietary IBM technology that was definitely faster than ISA, but for what was out back then ISA was more than good enough.
Beta vs VHS - This was mentioned during the previous article on the Rambus problem, but deserves mentioning again. Beta was better, but how many could actually tell the difference, and how many wanted to pay for a marginal difference.
PC vs MAC - Of course MACOS has weaknesses that are well described, but it was well ahead of DOS.
In the end the succesful technoglogy was the one that had the blend of "just enough performance to do what I want" and low cost. Rambus doesn't do it, PC133 barely does, mostly because the marginal cost of PC133 vs PC100 isn't too bad. PC266 probably has sometime to go just because there isn't a killer app for it.
Rambus only has a chance if the follwing conditions are met:
1) There is a killer app that requires multiple rambus channel type speed. 2) There is not a cheaper alternative that is adequate.
Neither conditon exists today in the mass market PC or workstation. And, even with servers a single Rambus channel really isn't anything special.
I think you may need to look a little harder. I would guess that most of the projects that are easily visible do not push the limits. The only thing I can think of off the top of my head that is pushing the limits is the Extreme Linux work. While other companies and OS are doing large scale distributed computing it is still the bleeding edge.
I also don't believe it will ever be easy to find "innovative" work in Linux. The main reason not being that Linux tends to reimpement existing ideas better, but because Linus and AC are more conservative than people think. Read the kernel archives sometime, and pay attention to what they say. Nothing gets into the main kernel tree without a damn good reason and being fairly stable and tested.
This seems to contradict the idea that Linux develops fast, but it really doesn't. New features get developed tested and become good enough to enter the kernel tree very fast, so being conservative in Linux isn't nearly a slow as it may sound.
Also, I really don't believe Linux shoudl be that innovative. It should be a rock solid platform for others to innovate on top of. And, that is happening, I would consider the uSIMM a good example of people innovating on top of Linux.
Finally, I wish someone would actually define innovative. MS says that the governement is stifling innovation, but they have never defined it. Every talks about innovation and nods there heads like they understand, but I have never heard anyone who can really describe it. Personally, I consider making something better, faster, and cheaper is innovation. New software ideas generally don't come from companies or OS communities. Generally, they just implement and distribute to the masses. The original ideas come from individuals whether inside a company R&D organization, more often academics. And, most of those ideas are probably unfeasible, stupid, or unworkable. The trick is finding the good ones, and taking advantage of them.
First, I agree a school can't necessarily teach you to think. It can make you a better thinker by showing different ways to approach a problem. But, you still have to have the potential to start with.
What you are describing though are bad teachers, not that school can't teach you anything. You also have the data points to show that not all teachers are like that. What you don't have is the background of coming into a school with good teachers, and not being "one of those kids wih genius level IQ." I don't know if I am one of those or not myself. I like to think so, but I had a pretty humbling experience at my school with a lot of people who were a hell of a lot smarter than me.
The other thing is I had a similar department to your CS department, except for me it was the math department. At my school everyone has to take math through Differential Equations (anyone from that school can probably guess it now.) The classic example that matches some of your story was Differential Equations there were two sections and I got the worst one (couldn't take the other due to morning practice). I basically failed the first mid-term because all the professor did was write proofs on the board, and then assign homework that was application. Strike 1: Not giving practice for what you are teaching. Then, the tests would be on the homework type problems. Striek 2: Not testing what your teaching. The only thing that got me through is I got together with people in the good section to work on homework. I went to the mid-term reviews that were taught by the good professor, and would learn and understand about 4 weeks worth of material that was totally incomprehesnible before in about 2 hours. I ended up with a B-. But, the fact that I could learn enough in 2 hours to pass the class from a good teacher, that I couldn't in 3 hours/week of class from a bad teacher taught me a lesson about how much a teacher can influence the learning process.
On the other hand I had a CS professor who taught Algorithms and Theory of Computation. The way I see it these could have been two of the most boring and uninteresting classes ever. But, because of the quality of the teacher and the work he did I learned a lot in those classes, that I continue to apply today. (Thanks Ran).
And, the parallels to Intel/Rambus are hard to ignore.
1) Rambus wants a buttload in license fees.
2) SDRAM, DDR-SDRAM are open standards which anyone can build to, royalty free. (There may be some minor royalties, but these are negotiated or worked around by JEDEC.)
and, one you didn't mention that applies to beta vs vhs and rambus vs ddr.
3) The performance difference is not noticeable by the average consumer.
So, you end up with a marginally better technology, that is more than marginally more expensive. Which is a recipe for disaster in the PC industry. Another point to note is that proprietary hardware has never suceeded in the PC space. Examples:
1) MCA vs ISA - IBM got spanked ont his one even though MCA was the better technology.
2) Apple/Mac vs IBM compatible - Apple and MacOS had superior software and hardware, but got spanked by the open hardware PC.
3) Soundblaster compatible devices - While I am sure there were better alternatives, for some reason everything was soundblaster compatible.
I am sure more coudl be named, but I don't have time.
Actually, AMD pretty much does this with the K6 and Athlon already with 32 bit instructions. Except the translation is done in hardware not software.
Since, no one outside of AMD knows what the internal ops used in the Athlon are we can't really say how hard it would be to make those operations 64-bit. But, if that is easy, and it is easy to decode x86 ops to the internal 64bit ops and then make the output look right. Then, this should pretty much be a slam dunk for AMD.
This doesn't prove a Gaussian curve is "wrong". What it is saying is that the new data is evidence that a gaussian distribution is probably the wrong model to be using under the circumstances described. The theory is that self similar phenomena do not follow a gaussian distribution, but follow this new distribution. It seems to me that the deep mathematical analysis has not really been done, but the experimental evidence suggests the existence of this distribution. There is probably a lot of work ahead in coming up with a mathematical model for the new distibution. What would be real interesting would be if the mathematical model for the distribution reduces to gaussian under specific conditions, kind of like how special relativity reduces to classical mechanics at low speeds.
The biggest implication of the model is int he insurance industry. If it is found that floods, fire, earthquakes, and hurricanes follow the new distribution. It may allow insurers to go back to insuring against earthquakes and hurricanes because tey can actually predict long term income and expenditures more accurately. Maybe they will actually do their job instead of claiming hardship whenever a disaster strikes somewhere.
Insurance Exec: Oh wahhh!!! We can't pay a billion in claims, go to the government.
Translation: We have taken in a net profit of 2 billion dollars over the last two years. But, the billion dollars for this disaster will affect our earnings numbers for the next quarter or two and my stock options will be worthless.
This is a management issue, not a technical one. You are a technician, not a manager.
Actually this bring up a good point that wasn't mentioned. HR alone really does not have the authority to unilaterally and arbitrarily have the network scanned for porn if it hasn't been before. This sort of order should come from above HR, and be OK'd with legal, and all sort of other things before it even gets to the IT person. I get the feeling this hasn't happened, and HR is requesting without any authorization from the higher ups.
Re:BIG RISK:Cassini carries a HUGE load of Plutoni
on
Cassini visits Earth
·
· Score: 1
The bravado expressed on this news group indicates a high degree of scientific arrogance inappropriate for critical and skeptical thinkers, especially when one considers the many rocket disasters of late, the Challenger disaster, the many Apollo failures, and the reliance on Europeans for many critical components on board the craft. Thank you
The thing that pulls the rug out of the arguments is that RTGs have been in craft that have exploded at launch and burned up on reentry. Every RTG in those craft landed intact (although not all were recovered) I believe one was tracked into the ocean where it couldn't be recovered.
Does that eliminate all risk? Of course not, but combining the risk of the RTG actually disintegrating with the probability of the craft actually hitting the earth with the actual damage the dispersed plutonium might do, and the risks are extremely small. Oh and the risks at launch are even smaller, since even if every fuel source on every booster were to explode simultaneously with every molecule of fuel combining with with an oxidizer at the same instant (a complete impossibility) the resulting forces would at worse launch the intact RTGs away from the launch sight, but could not possibily disintegrate them.
Actually, the other poster should have qualified Gravitational theory, with either Newtonian or General Relativity both of which are not completely accurate, and therefore, according to the argument of creationists, should not be taught in school.
Very good points. A recent Scientific American had a review for a book about debunking creationism. The basic point being that there are no arguments for creationism, the evidence for creationism are simply arguments against evolution. Which in no way proves creationism.
Another point is that creationism tends to sit on the fact that because life is so complex there is no way it could have developed without a divine hand. The problem with that argument is that if one of the things that is pointed at as proof of God is found to be completely natural and explainable, it casts doubt on the existence of God at all.
This is probably part of the reason the catholic church embraced evolution. They realized that trying to prove the existence of God via natural evidence was pointless and self defeating. Faith does not require proof, and it cannot be disproven. If proof is attempted it allows for disproof, which like is said before is self defeating for a religion.
Actually, species is pretty well defined. Two creatures are of different species if they cannot interbreed. Although, I am not sure how this is defined for asexual reproducing creatures.
So , while an artificial distinction, it is pretty easy to say whether a new species is seen to evolve and should be possible to observe in the laboratory with some short lived insects by taking two populations and applying a stress to those populations that will cause selection to take place as better adapted individuals survive over less adapted ones. It might take a couple of years, but after a lot of generations we should have two populations cannot interbreed.
That isn't exactly what I was referring to, but it is a good example. A better one is that. Newtonian mechanics has holes in it. So, we have to teach Relativity only. Wait Relativity doesn't explain everything so we have to teach quantum mechanics. Ooops quantum mechanics doesn't apply to times before 1e-42s after the big bang. We can't teach anything.
This is the creationist argument. Evolution doesn't explain everything therefore it can't be taught. Basically the argument is that no science can be taught ever. Which basically makes these people the biggest hypocrites in the world, because they don't want any science taught, but are perfectly happy reaping the benefits of said science.
This isn't ignorance either. This is pure stupidity. Ignorance is not knowing anything. Stupidity is not being able to think. Obviously these people can't even apply there own arguments to themselves which to me says they cannot think, therefore they are stupid. QED.
Slashdot Mirror
Do you think they can sell it for less than 2.99? That is what Linux Mall charges to send you a CD of any of the major distributions.
Dastardly
This is prolly a good book and all but get real people, computers are just tools and the audience this book was intended for knows this.
Really? How many programmer's really know what P and NP mean and imply? Not all are computer scientists.
How many programmer's know that there is no general algorithm to detect infinite loops in a program? (It can be done for a specific program on specific hardware, but takes an enormous amount of memory. Basically you store the entire system state at each step and check for a repeat.)
Dastardly
I wouldn't say a lifetime, but the market is so big that even with Fab 30 AMD can only supply somwhere between 30 and 40% of the PC market. So, it will be quite a while before AMD "beats" Intel, requiring adding 1 or 2 more Fabs. the important thing is that they remain a viable and profitable competitor with the 20-30% marketshare capability as a threat to Intel.
Dastardly
Assume the code being patented is contained in my body. Legally, the patent would make it illegal for me to be alive, because my body would be using the patented gene to make proteins. What if they patented a blood protein. I could now no longer donate blood because the use of the gene is patented by some corporation, and that corporation must be compensated for that donation. So, it seems far fetched, but that is what the letter of the law says.
It could also be argued that because the gene is part of my body, it is my property, therefore granting exclusive rights to use that gene to someone amounts to the illegal seizure of my property. The key point here is that patenting genetic code discovered through the sequencing of human DNA implies the human DNA came from a person or group of people. The only argument that needs to be made is that a person legally owns their own DNA. Once that assertion is made, there is no justification for patents on human DNA sequences.
The only problem is that this argument might work for human DNA, but it won't invalidate other animal or plant DNA patents.
Dastardly
Note: this only applies to copyright law. Patent law on the other hand is a completely different issue although, i believe it should follow the same rules to some extent. I believe patents on genetic codes that occur in nature should be unenforceable since they are simply fact, and regardless of the effort required to decode the fact that my Y chromosome has the sequence hcctgaaggth should not be patentable.
The problem is that under current copyright law facts are not owned by anyone even if you did the work to acquire those facts they are nto owned by you. Using the original poster's example of rainfall data.
If he goe out and collects the data and publishes it to a database, copyright law would say he owns the data base, and protect that ownership. The thing is that he doesn't own the facts in the database. Some one else could use the facts in the database (i.e. each piece of rainfall data) and put it into there own compilation. And, the original owner will have no legal protection. While ethically the original source of the facts must be sited, there original author does not have to be compensated for the use of the data in another work.
This is a good thing. Facts should always be public domain regardless of who did the work to get those facts. In the US this has been established by legal precedent and copyright law. The proposed change to the law would increase the costs of research and development because suddenly every researcher would have to pay for basic facts, and could not legally compile there own database without either paying the original source of the facts, or collecting those same facts all over again.
I don't think the implicaions of this law are understood completely. Current copyright law through various legal precedents grants copyright protection to the format of a collection of data. the classic example is a phone book. It also only applies to the exact organization if that organization is not obvious.
The classic example is a phonebook. A phonebook is a collection of data i.e. names, phone numbers, and addresses. Organized in alphabetical order. As it turns out under current copyright law this has minimal protection. Alphabetical ordering is obvious, and the rest of the directory is information which by law is publlic domain and not protected by copyright.
A law protecting databases and their content could easily extend to a copyright on information. Basically, a database should be covered just like a phone book. Any content in the database would be owned by the creator of that content, but any information would have to continue to be public domain.
Basically, this means that the databases of internet search engines can be extracted and reorganized into a new database, simply because URL and page titles are information and therefore are not and should not be protected.
Dastardly
P.S. Arguably a page title could be considered the property of the creator of the original, but the URL is really public domain information and not protected by copyright.
A more likely cause of a computer shutdown tonight is probably some drunk driver ramming a nearby power pole and shutting power to the entire building (and region). I would place that as many times more likely than a Y2K glitch. Dastardly
This is not informative. It is 90% incorrect. Dastardly
You are mixing up validation and testing. Validation parts never ship to any customer, or at best ship as engineering samples only. These are the parts that boot operating systems, and get worked over by the engineers to make sure the design or production process was not screwed up. Processor testing for customer ship is a completely different process. It starts by testing the processors while they are still on the wafer because packaging is costly, therefore you want to only package parts that can be sold. Therefore, you want to find out speed too, since a 200MHz part isn't worth packaging these days. This involves a lot more than simulating booting, and only takes a few seconds. Once the processors are packaged they are tested again at temperature. The temperatures are not that much higher than what the high end of the junction temperature spec is. Cushion comes because the test developers have guardbanded the tests in order to guarantee the part works as labeled. Package test is where parts are binned and labeled for sale. Again this only takes a few seconds. Your processor never sees the inside of a computer until the OEM puts it in, or you do it yourself! The rumours I have seen on the net so far say it is a test escape. Basically, a failure mode came up that their testing couldn't catch. So, they have probably fixed the test by now, and are already shipping all good parts again. This sort of thing is can happen to any processor manufacturer since it is impossible to test all possible combinations of data. Just to try every combination for 64 data pins would require be O(64!). Or, longer than the age of the universe per processor.
It is interesting to note what ease-of-use seems to mean every time it comes up.
1) Limit choice in order to limit complexity.
2) If you are used to it, it must be easy to use. (i.e. the UI should be just like Windows, or Mac)
Both of which make things easy to use, but are really the lazy way to accomplish ease of use. Maybe a research project shoudl be done to figure out just what ease of use would really be. It would require a lot of work and vounteers and controls. You would have to take a Unix guru, Unix user, MAC guru, MAC user, Win guru, Win user, some one who just knows how to get into AOL and start Word, and finally some one who has littl eexperience with computers. If enough data points were acquired it shoudn't be to hard to figure out what ease of use really is.
Dastardly
Pretty much the whole measuring delay by clock cycles is pretty ambiguous. The correct way is to use measure the latency in ns, then at least you can compare across clock speeds.
It is interesting to note that if you look sat the first cycle delays in DRAM from FPM to EDO to SDRAM they are pretty consistent. 60ns FPM was 5-3-3-3 at 66MHZ and 70ns EDO is 5-2-2-2 (4-2-2-2 at 60ns) at 66MHZ, at 100MHZ SDRAM is 5-1-1-1-3(2)-1-1-1. Note the middle number is the CAS we hear about not the first one. So, first cycle number has been from 70-50ns from FPM to SDRAM, but that can be attributed to process technology improvement, there really hasn't been an architectural improvement in latency.
You also have to specify which latency you are talking about especially with SDRAM and DRDRAM because the latency depends a lot on the access pattern.
So, before we get inot arguments, let's make sure there is a consistent measurement. Just one point DRDRAM has numberous latency numbers depending on what is being accessed and when.
I don't see any possibel way DRDRAM could be cheaper than (DDR)SDRAM on the same process given similar economies of scale. The extra die area of the DRDRAM alone kills that, then combine it with royalties on the chips, RIMMS, and chipsets, and it is absolutely impossible for DRDRAM to cost less tham SDRAM all other things being equal.
Also, JEDEC is working on memory that will be faster than DDR-SDRAM leveraging on work from the SLDRAM group, and going even beyond that. And, the JEDEC designs don't carry the baggage that DRDRAM carries.
Dastardly
Does anyone know who AC got his Athlon system from? It doesn't say in his diary. But, if it were form AMD I would call that support for linux.
MCA vs ISA - MCA was prprietary IBM technology that was definitely faster than ISA, but for what was out back then ISA was more than good enough.
Beta vs VHS - This was mentioned during the previous article on the Rambus problem, but deserves mentioning again. Beta was better, but how many could actually tell the difference, and how many wanted to pay for a marginal difference.
PC vs MAC - Of course MACOS has weaknesses that are well described, but it was well ahead of DOS.
In the end the succesful technoglogy was the one that had the blend of "just enough performance to do what I want" and low cost. Rambus doesn't do it, PC133 barely does, mostly because the marginal cost of PC133 vs PC100 isn't too bad. PC266 probably has sometime to go just because there isn't a killer app for it.
Rambus only has a chance if the follwing conditions are met:
1) There is a killer app that requires multiple rambus channel type speed.
2) There is not a cheaper alternative that is adequate.
Neither conditon exists today in the mass market PC or workstation. And, even with servers a single Rambus channel really isn't anything special.
Dastardly
I think you may need to look a little harder. I would guess that most of the projects that are easily visible do not push the limits. The only thing I can think of off the top of my head that is pushing the limits is the Extreme Linux work. While other companies and OS are doing large scale distributed computing it is still the bleeding edge.
I also don't believe it will ever be easy to find "innovative" work in Linux. The main reason not being that Linux tends to reimpement existing ideas better, but because Linus and AC are more conservative than people think. Read the kernel archives sometime, and pay attention to what they say. Nothing gets into the main kernel tree without a damn good reason and being fairly stable and tested.
This seems to contradict the idea that Linux develops fast, but it really doesn't. New features get developed tested and become good enough to enter the kernel tree very fast, so being conservative in Linux isn't nearly a slow as it may sound.
Also, I really don't believe Linux shoudl be that innovative. It should be a rock solid platform for others to innovate on top of. And, that is happening, I would consider the uSIMM a good example of people innovating on top of Linux.
Finally, I wish someone would actually define innovative. MS says that the governement is stifling innovation, but they have never defined it. Every talks about innovation and nods there heads like they understand, but I have never heard anyone who can really describe it. Personally, I consider making something better, faster, and cheaper is innovation. New software ideas generally don't come from companies or OS communities. Generally, they just implement and distribute to the masses. The original ideas come from individuals whether inside a company R&D organization, more often academics. And, most of those ideas are probably unfeasible, stupid, or unworkable. The trick is finding the good ones, and taking advantage of them.
Dastardly
A couple points...
First, I agree a school can't necessarily teach you to think. It can make you a better thinker by showing different ways to approach a problem. But, you still have to have the potential to start with.
What you are describing though are bad teachers, not that school can't teach you anything. You also have the data points to show that not all teachers are like that. What you don't have is the background of coming into a school with good teachers, and not being "one of those kids wih genius level IQ." I don't know if I am one of those or not myself. I like to think so, but I had a pretty humbling experience at my school with a lot of people who were a hell of a lot smarter than me.
The other thing is I had a similar department to your CS department, except for me it was the math department. At my school everyone has to take math through Differential Equations (anyone from that school can probably guess it now.) The classic example that matches some of your story was Differential Equations there were two sections and I got the worst one (couldn't take the other due to morning practice). I basically failed the first mid-term because all the professor did was write proofs on the board, and then assign homework that was application. Strike 1: Not giving practice for what you are teaching. Then, the tests would be on the homework type problems. Striek 2: Not testing what your teaching. The only thing that got me through is I got together with people in the good section to work on homework. I went to the mid-term reviews that were taught by the good professor, and would learn and understand about 4 weeks worth of material that was totally incomprehesnible before in about 2 hours. I ended up with a B-. But, the fact that I could learn enough in 2 hours to pass the class from a good teacher, that I couldn't in 3 hours/week of class from a bad teacher taught me a lesson about how much a teacher can influence the learning process.
On the other hand I had a CS professor who taught Algorithms and Theory of Computation. The way I see it these could have been two of the most boring and uninteresting classes ever. But, because of the quality of the teacher and the work he did I learned a lot in those classes, that I continue to apply today. (Thanks Ran).
Dastardly
And, the parallels to Intel/Rambus are hard to ignore.
1) Rambus wants a buttload in license fees.
2) SDRAM, DDR-SDRAM are open standards which anyone can build to, royalty free. (There may be some minor royalties, but these are negotiated or worked around by JEDEC.)
and, one you didn't mention that applies to beta vs vhs and rambus vs ddr.
3) The performance difference is not noticeable by the average consumer.
So, you end up with a marginally better technology, that is more than marginally more expensive. Which is a recipe for disaster in the PC industry. Another point to note is that proprietary hardware has never suceeded in the PC space. Examples:
1) MCA vs ISA - IBM got spanked ont his one even though MCA was the better technology.
2) Apple/Mac vs IBM compatible - Apple and MacOS had superior software and hardware, but got spanked by the open hardware PC.
3) Soundblaster compatible devices - While I am sure there were better alternatives, for some reason everything was soundblaster compatible.
I am sure more coudl be named, but I don't have time.
Dastardly
Actually, AMD pretty much does this with the K6 and Athlon already with 32 bit instructions. Except the translation is done in hardware not software.
Since, no one outside of AMD knows what the internal ops used in the Athlon are we can't really say how hard it would be to make those operations 64-bit. But, if that is easy, and it is easy to decode x86 ops to the internal 64bit ops and then make the output look right. Then, this should pretty much be a slam dunk for AMD.
Dastardly
This doesn't prove a Gaussian curve is "wrong". What it is saying is that the new data is evidence that a gaussian distribution is probably the wrong model to be using under the circumstances described. The theory is that self similar phenomena do not follow a gaussian distribution, but follow this new distribution. It seems to me that the deep mathematical analysis has not really been done, but the experimental evidence suggests the existence of this distribution. There is probably a lot of work ahead in coming up with a mathematical model for the new distibution. What would be real interesting would be if the mathematical model for the distribution reduces to gaussian under specific conditions, kind of like how special relativity reduces to classical mechanics at low speeds.
The biggest implication of the model is int he insurance industry. If it is found that floods, fire, earthquakes, and hurricanes follow the new distribution. It may allow insurers to go back to insuring against earthquakes and hurricanes because tey can actually predict long term income and expenditures more accurately. Maybe they will actually do their job instead of claiming hardship whenever a disaster strikes somewhere.
Insurance Exec: Oh wahhh!!! We can't pay a billion in claims, go to the government.
Translation: We have taken in a net profit of 2 billion dollars over the last two years. But, the billion dollars for this disaster will affect our earnings numbers for the next quarter or two and my stock options will be worthless.
Actually this bring up a good point that wasn't mentioned. HR alone really does not have the authority to unilaterally and arbitrarily have the network scanned for porn if it hasn't been before. This sort of order should come from above HR, and be OK'd with legal, and all sort of other things before it even gets to the IT person. I get the feeling this hasn't happened, and HR is requesting without any authorization from the higher ups.
The bravado expressed on this news group indicates a high degree of scientific arrogance inappropriate for critical and skeptical thinkers, especially when one considers the many rocket disasters of late, the Challenger disaster, the many Apollo failures, and the reliance on Europeans for many critical components on board the craft. Thank you
The thing that pulls the rug out of the arguments is that RTGs have been in craft that have exploded at launch and burned up on reentry. Every RTG in those craft landed intact (although not all were recovered) I believe one was tracked into the ocean where it couldn't be recovered.
Does that eliminate all risk? Of course not, but combining the risk of the RTG actually disintegrating with the probability of the craft actually hitting the earth with the actual damage the dispersed plutonium might do, and the risks are extremely small. Oh and the risks at launch are even smaller, since even if every fuel source on every booster were to explode simultaneously with every molecule of fuel combining with with an oxidizer at the same instant (a complete impossibility) the resulting forces would at worse launch the intact RTGs away from the launch sight, but could not possibily disintegrate them.
Actually, the other poster should have qualified Gravitational theory, with either Newtonian or General Relativity both of which are not completely accurate, and therefore, according to the argument of creationists, should not be taught in school.
Very good points. A recent Scientific American had a review for a book about debunking creationism. The basic point being that there are no arguments for creationism, the evidence for creationism are simply arguments against evolution. Which in no way proves creationism.
Another point is that creationism tends to sit on the fact that because life is so complex there is no way it could have developed without a divine hand. The problem with that argument is that if one of the things that is pointed at as proof of God is found to be completely natural and explainable, it casts doubt on the existence of God at all.
This is probably part of the reason the catholic church embraced evolution. They realized that trying to prove the existence of God via natural evidence was pointless and self defeating. Faith does not require proof, and it cannot be disproven. If proof is attempted it allows for disproof, which like is said before is self defeating for a religion.
Dastardly
Actually, species is pretty well defined. Two creatures are of different species if they cannot interbreed. Although, I am not sure how this is defined for asexual reproducing creatures.
So , while an artificial distinction, it is pretty easy to say whether a new species is seen to evolve and should be possible to observe in the laboratory with some short lived insects by taking two populations and applying a stress to those populations that will cause selection to take place as better adapted individuals survive over less adapted ones. It might take a couple of years, but after a lot of generations we should have two populations cannot interbreed.
Dastardly
That isn't exactly what I was referring to, but it is a good example. A better one is that. Newtonian mechanics has holes in it. So, we have to teach Relativity only. Wait Relativity doesn't explain everything so we have to teach quantum mechanics. Ooops quantum mechanics doesn't apply to times before 1e-42s after the big bang. We can't teach anything.
This is the creationist argument. Evolution doesn't explain everything therefore it can't be taught. Basically the argument is that no science can be taught ever. Which basically makes these people the biggest hypocrites in the world, because they don't want any science taught, but are perfectly happy reaping the benefits of said science.
This isn't ignorance either. This is pure stupidity. Ignorance is not knowing anything. Stupidity is not being able to think. Obviously these people can't even apply there own arguments to themselves which to me says they cannot think, therefore they are stupid. QED.
Dastardly