There will be a problem with any license, even near duplicates of the GPL, if the applicable state law includes UCITA. Right now, this means Virginia and Maryland, with Iowa close behind them. Why? Because UCITA permits unilateral and unlimited modification of the license terms. Just because it is GPL today does not mean it is GPL tomorrow. An email notification attempt followed by a brief wait is all it takes to change a GPL'ed program into a proprietary program.
Things get far more complex as the author count increases and the variety of potentially applicable state laws increase. But UCITA rules on the original license will probably be held to flow through to all the derivative works.
So a change of management can take the best of organizations with perfect GPL intentions and convert it into proprietary software.
Right. Regulations are not working because regulators and politicians are nowhere near as skilled as accountants and engineers. Under energy deregulation the commercial and industrial sector have more than tripled the amount of goods and services provided per BTU of fossil fuel used (1980 vs 1995 figures).
How? Instead of trendy thinking and publicity driven decisions, they do plain old accounting and engineering calculations. These almost always lead to the answer: "Energy efficiency". Solar makes sense for remote sites (where bringing in power or fuel is very expensive) and mobile equipment. Some places wind makes sense. But more than 95% of the time you get the most for your money with energy efficiency.
Why don't you hear more about this? The commercial sector does it to save money (by saving energy). Spending money on publicity is for politicians and bureaucrats. Commercial customers want to save money.
(And for you PC users. About half the energy used by PCs and workstations is spent powering the monitor so that it runs some pretty picture screensaver. The simple step of switching to a DPMS screensaver cuts power consumption in half. You can save more with further changes, but pretty picture screensavers remain commonplace. Just a reminder.)
Software is by no means the only meritocracy around. Consider sports. We have all experienced the sports meritocracy. You think it is easy to be a standout football player? You think all those jocks never work hard? They face a meritocracy just as strict in its own field. They get a daily win/lose report. Everyone can see who is good at sports and who is bad at sports.
Does this lead to maturity? What words do you think of when discussing the jocks in school? Don't words like arrogant, insensitive, intolerant, testosterone poisoned, and selfish come to mind? Given your experiences with jocks, how do you feel about a world where the jocks have all the power? You've probably had a taste of that at some time.
The outside perspective on programmers is similar. Being a meritocracy does not lead to maturity, wisdom, leadership or organizational skills.
If the FTC had agreed to the original Disney offer a significantly worse precedent would have been set. Even though Disney's intentions were to maintain the privacy, the data would have been sold without restriction. The FTC would have been agreeing to the concept that collected data was saleable.
Under this agreement they are setting the much tighter terms that data may be sold if and only if the purchaser maintains the same agreement that was in place when the information was gathered. It is a reasonable legal tradeoff. If you make the rules on a sale any harder, the database owners might win the argument on contract violation. The FTC would have been arguing that contract terms get changed in the customer's favor during a bankruptcy.
This agreement is consistent with how other contracts are handled during a bankruptcy. The ideal is to maintain all contracts unchanged. The ideal is unreachable (since one party is bankrupt) so the bankruptcy court is authorized to make such modifications to contracts as are needed to minimize the deviations. Bankruptcy law establishes which contracts take precedence, who loses first, how parties negotiate over relative importance and value, etc.
Since the law was written before privacy agreements existed or mattered it was not established that they should be considered like contracts. This sets that precedent.
The "why" question is well answered by the elk pages. Scheme's functionality is a marvelous match to the needs of extension languages. - The compiler/interpreter is of modest size. - The performance is very good (for an interpreter) - The language is known, well defined, and complete. But is has the problems of - having lots of irritating silly parentheses. - being considered weird.
I know that when I needed a programmable control language for a network tester we looked at the alternatives of TCL, Perl, and Scheme. (This was some years ago, when Perl integration was very hard and Python nearly an unknown.) The implementation cost of integrating Scheme was about 20% that of the other two, and with Scheme we got the immediate support for the kind of complex nested data structures that we needed. With the other two, the test script implementers would have had to write all that.
But, the learning curve for LISP is much too high. What the testers really want (and now have) is a pretty GUI interface.
And how many of the registered voters are real? There have been a number of voting frauds in the US where instead of the hassles of finding and reusing dead people, they just registered imaginary people. The present "Motor Voter Registration" makes it especially easy to register imaginary people.
It also makes it really hard to remove imaginary people. When I moved most recently I checked for registered voters at my new address. It went back several occupants. I found that I was still validly registered at my old address in my old precinct. And you needed a formal challenge hearing to remove these people. So for several years my address had less than 50% turnout.
This is where the present employment contracts (or lack thereof) fail to reflect reality. Unionized locations like factory lines deal with these issues explicitly. There are requirements for payphone access, break times, etc built into the contract. You can indeed be fired for using a company phone or violating break time rules.
Most people have a more informal employmet contract that permits infrequent minor use of telephones but with highly varied privacy protections. My recent large corporate employers had very clear rules: - telephone usage was monitored, but with privacy protections. The monitors got anonymized data. Managers got only aggregated data on total usages. The corporate rules then made it clear that if the anonymized data indicated abuse, the anonymity would be lifted and sanctions up to criminal charges might follow. - Similar monitoring of email and web browsing with privacy protections spelled out. Plus, every login included a banner reminding users of business use only.
These companies took these contracts seriously. They did monitor. They did anonymize. They enforced privacy protections. And they responded rapidly to abuses. The legal departments enforced both sides of this contract strictly because it was their belief that the privacy protections were needed to maintain employee acceptance, the privacy protections were strictly enforced to protect against countersuits when abusers were sanctioned, and the abuse monitoring response was rapid both to avoid establishing a precedent for tolerating abuse and to catch abuse at a stage where a minor reprimand was appropriate and likely to end the abuse.
The present legal environment is still one that the business's equipment (phones, computers, supplies, etc.) are to be used only for purposes approved and authorized by the business. This does make sense. It belongs to them. They pay for it. As part of your compensation they allow limited personal use with the understanding that you sacrifice privacy.
There is one (and to my knowledge only one) exception. The law explicitly permits limited non-interference use of corporate facilities for union and labor related purposes. This includes such things as discussion of working conditions, wages, etc. There are strict limits on when and how business facilities can be used, but when within these limits there are also very strong privacy protections.
For everything else, use a cell phone (on your break time), use your own personal equipment, do it from home.
If the URL matches something in the blocklist, then Junkbuster does not even attempt to load it. Instead, it sends an internally generated 404 that is also sometimes informative. This is better than playing games with DNS entries because most browsers handle 404 errors more gracefully. The DNS entry games don't always work as well.
So once you have "doubleclick.com" in your blocklist, absolutely anything on any doubleclick URL gets a 404 with no network traffic generated.
If you really feel the need, junkbuster has the option of sending wafers. These are cookies with some entertaining legalese that are designed to fit within the rules for a cookie. Or you can invent your own wafer.
But the issue with 1x1 web-bugs is not cookies. These web-bugs are already encoded with the tracking information so that the mere attempt to load the image provides the tracking information to the perpetrators.
There is also the information theoretical limit. The bandwidth used and the interfering noise can be combined to find the theoretical limit. People used to approximate Shannon's law by saying the limit was 1 bit/sec per Hz. For typical S/N ratios over things like analog lines using old modems this was not too far off. But as you go to better signalling systems the S/N ratios start allowing you to do better than that. Hence the ability to transfer 50+ Kb/sec over a nominal 4 KHz analog channel.
The S/N term is logarithmic and better telemetry systems and coding systems eventually run up against quantum walls, thermal noise limits, etc. that impose some fundamental limits on available S/N. So you still cannot escape Shannon's law.
The present bandwidth allocations are a sometimes crude attempt to manage the co-existance of different imperfect hardware, signalling, and coding systems. To a certain degree some signalling and coding systems can co-exist, e.g. spread spectrum signals hiding underneath the noise threshold for SSB signals. This kind of co-existance is permitted in very few bands. Most bands are allocated to a single telemetry and coding system so that the necessary guard bands, emission quality, and other constraints can be described easily and enforced easily. Real world hardware is imperfect, so you do need some saftey margins as well as guard bands due to signalling technology.
Up in the millimeter bands the attenuation due to molecular absorbtion has interesting effects. You get line of sight transmission that is relatively unaffected by rain and fog, blocked completely by most structures, and attenuates at 20-100db/km. This leads to some inherent excellent re-usability when used for short distance point to point links. It originally (and still) was used for other interesting things like covert radars and covert links that deliberatly operated near the peaks of molecular absorbtion bands.
There is a real danger in letting your desire for perfection interfere with taking reasonable safety precautions. Drug safety in the US is imperfect, but a whole lot better than in the "good old days" when patent medicines were severely toxic, drug toxicity reactions not publicized, etc.
Similar efforts to provide guidelines for working with Recombinant DNA, with hazardous materials, etc. have made significant improvements. The vast majority of the players want to behave responsibly and will follow reasonable guidelines. The fact that there are a few percent who will ignore the safety procedures does not make safety procedures irrelevant. They reduce the level of irresponsible behavior, and we can hope that the remaining hostile behavior is infrequent enough that we can deal with the resulting problems.
And you will find that most of the government players will also try to follow good safe practices, though there is always reason to doubt the wisdom of the TLAs.
There is already quite a lot of energy saved this way. CFL sales have been substantial for years and are still growing. The latest trends are in the natural spectrum (better light quality) phosphors and the low harmonics CFLs. (Harmonics cause "hidden" power loss in the grid and make reactance compensation more difficult.) The trick is to get those to be as efficient as the high harmonics and "blue" flourescents. Give them a few years and they will be there.
Also vaguely related is the growth in the use of LEDs for lighting. The red LED is more efficient at generating red than a flourescent, and much more efficient than an incandescent. This is why you are seeing more and more LED based stop lights. Even for white light, the combination LEDs are more efficient than incandescents. The present barriers to their use are: 1. Production costs are high (think 10-50 times more than incandescent). 2. Large surface areas are needed for bright light. 3. Proper color balance is very hard. 4. DC power is used, meaning special power supplies and their attendent efficiency loss. (Electronic equipment already need DC power, so they don't suffer this problem.)
So for a while LEDs will be mostly used in situations where color balance is unimportant, like red lights.
There is an exception of sorts, but one that probably does not apply. There are rules concerning employer interference with union activities, both existing unions and actions to establish a union presence. These do impose some restrictions on employer monitoring. They can still monitor, but not in a way that interferes with legitimate union activity. This can sometimes apply to monitoring of email.
This probably does not apply to the situation described, but it does sometimes apply.
For a similar US Army site, see http://carlisle-www.army.mil/usassi/ssipubs/stdypr og.htm. It has a collection of monographs on topics of more current interest. For example, an excellent discussion of relative capabilities and possible intentions of PRC vs. Taiwan.
A short examination of the law should ease your fears. The US Courts have generally held that there is no protection for speech that is intended to cause harm. You can point to both the Constitution, Supreme Court rulings, and many laws that illustrate this. Key examples are:
- Fraud. False speech intended to cause harm can be criminal. This extends into things like false advertising, etc.
- Libel and Slander, where again false speech intended to cause harm can be cause for civil actions.
The legal grey zones tend to be those where intent or harm is unclear, and those where intent may include political purposes. But even with political speech, the free speech protections are quite limited when harm is involved. The grey zones are in things like signs and sound trucks, where the harm is visual and aural pollution. When the harm reaches the level of property destruction or personal injury, the free speech protections are not exemptions from civil actions and criminal prosecution.
Dis-allowing root usage is an excellent idea for many applications. The fact that NFS by default makes root usage aggravating has acted as an anti-root effect in many networked environments. If popular user programs throw up nag screens saying "You should not be root" it has a significant effect. Programs like gimp and other GUI end-user programs should all include a little "am I root?" query followed by a nag message indicating that the user is probably running a mis-configured system.
The instruction definitions for MMX and 3DNow instructions are already there in gas. You can make use of them by means of the asm() feature in gcc and egcs. I've written several MMX-enhanced programs using egcs. But, these compilers will not themselves issue the egcs instructions. Also, you need to be careful to manually maintain the proper relationship with other FPU usage.
Problems with this approach are: - gdb does not understand the FPU registers. Debugging MMX code is a real chore. You need to store things into memory before gdb can see them. - It is up to you to decide when and how FPU registers need initialization. - You are working in assembler and need to understand how to properly use asm(), __volatile__, and the like.
But it definitely works. I got reasonable speedups. MMX, 3DNow, etc. noticably inferior to AltiVec as an instruction set, but that has nothing to do with gcc and egcs. The asm() integration with the rest of the egcs compiler does make short MMX sequences quite reasonable. For longer code sequences it is better to write separate modules in assembler.
Both the comments here and in the article misconstrue one of the major complaints. This is due to a misinterpretation of the meaning of "error". For many of the older texts there is no single correct text. An important part of the text is the annotation that reveals the various different texts that have been found.
Shakespeare and the Bible are two of the most widely studied texts. Neither has a "correct" text. Every Shakespeare play has multiple originals that differ. The Bible as no originals left, just multiple copies. These copies differ in details. Some of the differences are copying errors, some are deliberate modifications, some are the translations, and some are the result of language evolution.
There are tools, like SGML markups, that allow the text to be annotated to reflect what is known about the history and evolution of a text. Project Guttenburg does not capture this information.
For some texts this is not important. For other texts this is important. Some readers will not care. Other readers will care. The ones that care are complaining about the loss.
IMPORTANT NOTE: This license is only for use within the educational system. That means: - no personal use - no commercial use - debatable for use in home schooling or distance learning. These would at least need official sponsorship from an educational system.
Still, an astonishingly good license considering that it came from Microsoft.
The extent to which a police state reduces crime has been demonstrated in Eastern Europe and the Soviet Union. Removal of the constant surveillance, wiretaps, copier controls, computer controls, etc. has been tried. The crime rate skyrocketed. It was more than a small increase. Crime went from extremely rare up to Western European levels and (in Russia) above. We also have the genuine public reaction to all this. In Eastern Europe the return to Communism forces are real, but a minority. Given the choice of police state or widespread crime, the public prefers widespread crime. Only in Russia, where the government has joined up with the criminals, is a return to the police state viewed favorably by the general public.
The "hoax" was the result of combined ignorance, fear, cynicism, and venality.
Ignorance - The reporters cannot distinguish between an annoying computer failure and a catastrophic computer failure.
Fear - Most fear computers. They are a genuine job threat to many in the news business and to many people that they know. Most of their friends and colleagues are in the older, dying industries that are threatened by computers.
Cynicism - They know that they cannot trust the government or corporate spokesmen to tell the truth. So they did not believe the initial warnings. Then once they learned enough to understand that there was a real problem, they did not believe all the reports as the real problems were fixed. Early Y2K tests found genuine problems, this was reported in the technical channels, and the later tests found the problems to be fixed and this was reported in the technical channels. The media did not understand and did not trust the government and corporate reports.
Venality - The news media, especially the electronic media, sell a product. Emotion sells much more easily than reason. Whether on CNN or slashdot, the emotional response is the easy response and the universal response. The easy Y2K emotion is fear. Fear sells. Fear maintains ratings. Fear is universal. Reason requires lots of boring detail and lacks punch. So the reports all emphasized fear.
The best example of what really happened is from school. We all know people who don't start studying until the night before finals. They cram and pass. They don't start the term paper until the day before it is due. They pull an all-nighter and pass. This may even be the most efficient use of their resources (given their priorities). And a lot of people from the top to the bottom of corporations and government have not changed. They let Y2K slide until the last minute, then they crammed, pulled all-nighters, and passed.
Look harder and accept that the project will be more obscure. The various related number theory projects are open source. Look at the ECMNET or Mersene page for ideas. Finding another factor to a Cunningham Number may take 6-12 CPU months, but it does happen.
You will not find servers or fame, in part because these attract script kiddies and vandals. But it does use those wasted CPU cycles.
The heirs of Dr King have sued for their copyright fees. They just have not won yet. The speech is copyrighted, the performance was copyrighted.
The issue with professor's notes is also covered by prior lawsuits. The question will be the extent to which the notes replicate the presentation. Capturing the information, but not the presentation, lets you avoid copyright issues. But if you capture the presentation also, there are copyright issues. None of the news reports provide enough information to judge whether these notes would violate copyright restrictions.
To be fair, I can and do use Linux as my primary desktop. And a mix of Linux and Solaris on production servers. But my "business desktop" remains NT. Because:
1.) Corporate file format requirements. I must be able to read, write, and modify documents that are shared activities. These documents are in MSWord format, MS Excell, MS Powerpoint, Visio, and PDF formats. I know and sometimes use Star Office. But it cannot modify MS documents without loss of formatting. There is no Visio. There is only PDF reader, not a full function PDF creator.
2.) Corporate communications requirements. I am required to have dial in and LAN access to Lotus Notes. No client support yet.
3.) Hardware variation support. Linux cannot support some of the highly integrated devices found in laptops and low-cost PC's. In my particular case it is a laptop. I don't get to pick the model. I have to take what the corporation provides. It is good quality, but has Linux problems. In general Linux support trails hardware availability by 6-12 months.
You can point in each case to a truthful "we are working on it". But working on it is not the same as available and robust today. These are reasons why I anticipate that within a few years Linux (and probably also *BSD) will be viable on the desktop. But viable in a few years is not the same as viable now.
Other people will have other particular problems, but the general categories of mandated file formats, mandated corporate communications, and hardware variations will keep coming up.
Slashdot Mirror
There will be a problem with any license, even near duplicates of the GPL, if the applicable state law includes UCITA. Right now, this means Virginia and Maryland, with Iowa close behind them. Why? Because UCITA permits unilateral and unlimited modification of the license terms. Just because it is GPL today does not mean it is GPL tomorrow. An email notification attempt followed by a brief wait is all it takes to change a GPL'ed program into a proprietary program.
Things get far more complex as the author count increases and the variety of potentially applicable state laws increase. But UCITA rules on the original license will probably be held to flow through to all the derivative works.
So a change of management can take the best of organizations with perfect GPL intentions and convert it into proprietary software.
Right. Regulations are not working because regulators and politicians are nowhere near as skilled as accountants and engineers. Under energy deregulation the commercial and industrial sector have more than tripled the amount of goods and services provided per BTU of fossil fuel used (1980 vs 1995 figures).
How? Instead of trendy thinking and publicity driven decisions, they do plain old accounting and engineering calculations. These almost always lead to the answer: "Energy efficiency". Solar makes sense for remote sites (where bringing in power or fuel is very expensive) and mobile equipment. Some places wind makes sense. But more than 95% of the time you get the most for your money with energy efficiency.
Why don't you hear more about this? The commercial sector does it to save money (by saving energy). Spending money on publicity is for politicians and bureaucrats. Commercial customers want to save money.
(And for you PC users. About half the energy used by PCs and workstations is spent powering the monitor so that it runs some pretty picture screensaver. The simple step of switching to a DPMS screensaver cuts power consumption in half. You can save more with further changes, but pretty picture screensavers remain commonplace. Just a reminder.)
Software is by no means the only meritocracy around. Consider sports. We have all experienced the sports meritocracy. You think it is easy to be a standout football player? You think all those jocks never work hard? They face a meritocracy just as strict in its own field. They get a daily win/lose report. Everyone can see who is good at sports and who is bad at sports.
Does this lead to maturity? What words do you think of when discussing the jocks in school? Don't words like arrogant, insensitive, intolerant, testosterone poisoned, and selfish come to mind? Given your experiences with jocks, how do you feel about a world where the jocks have all the power? You've probably had a taste of that at some time.
The outside perspective on programmers is similar. Being a meritocracy does not lead to maturity, wisdom, leadership or organizational skills.
If the FTC had agreed to the original Disney offer a significantly worse precedent would have been set. Even though Disney's intentions were to maintain the privacy, the data would have been sold without restriction. The FTC would have been agreeing to the concept that collected data was saleable.
Under this agreement they are setting the much tighter terms that data may be sold if and only if the purchaser maintains the same agreement that was in place when the information was gathered. It is a reasonable legal tradeoff. If you make the rules on a sale any harder, the database owners might win the argument on contract violation. The FTC would have been arguing that contract terms get changed in the customer's favor during a bankruptcy.
This agreement is consistent with how other contracts are handled during a bankruptcy. The ideal is to maintain all contracts unchanged. The ideal is unreachable (since one party is bankrupt) so the bankruptcy court is authorized to make such modifications to contracts as are needed to minimize the deviations. Bankruptcy law establishes which contracts take precedence, who loses first, how parties negotiate over relative importance and value, etc.
Since the law was written before privacy agreements existed or mattered it was not established that they should be considered like contracts. This sets that precedent.
The "why" question is well answered by the elk pages. Scheme's functionality is a marvelous match to the needs of extension languages.
- The compiler/interpreter is of modest size.
- The performance is very good (for an interpreter)
- The language is known, well defined, and complete.
But is has the problems of
- having lots of irritating silly parentheses.
- being considered weird.
I know that when I needed a programmable control language for a network tester we looked at the alternatives of TCL, Perl, and Scheme. (This was some years ago, when Perl integration was very hard and Python nearly an unknown.) The implementation cost of integrating Scheme was about 20% that of the other two, and with Scheme we got the immediate support for the kind of complex nested data structures that we needed. With the other two, the test script implementers would have had to write all that.
But, the learning curve for LISP is much too high. What the testers really want (and now have) is a pretty GUI interface.
And how many of the registered voters are real? There have been a number of voting frauds in the US where instead of the hassles of finding and reusing dead people, they just registered imaginary people. The present "Motor Voter Registration" makes it especially easy to register imaginary people.
It also makes it really hard to remove imaginary people. When I moved most recently I checked for registered voters at my new address. It went back several occupants. I found that I was still validly registered at my old address in my old precinct. And you needed a formal challenge hearing to remove these people. So for several years my address had less than 50% turnout.
This is where the present employment contracts (or lack thereof) fail to reflect reality. Unionized locations like factory lines deal with these issues explicitly. There are requirements for payphone access, break times, etc built into the contract. You can indeed be fired for using a company phone or violating break time rules.
Most people have a more informal employmet contract that permits infrequent minor use of telephones but with highly varied privacy protections. My recent large corporate employers had very clear rules:
- telephone usage was monitored, but with privacy protections. The monitors got anonymized data. Managers got only aggregated data on total usages. The corporate rules then made it clear that if the anonymized data indicated abuse, the anonymity would be lifted and sanctions up to criminal charges might follow.
- Similar monitoring of email and web browsing with privacy protections spelled out. Plus, every login included a banner reminding users of business use only.
These companies took these contracts seriously. They did monitor. They did anonymize. They enforced privacy protections. And they responded rapidly to abuses. The legal departments enforced both sides of this contract strictly because it was their belief that the privacy protections were needed to maintain employee acceptance, the privacy protections were strictly enforced to protect against countersuits when abusers were sanctioned, and the abuse monitoring response was rapid both to avoid establishing a precedent for tolerating abuse and to catch abuse at a stage where a minor reprimand was appropriate and likely to end the abuse.
The present legal environment is still one that the business's equipment (phones, computers, supplies, etc.) are to be used only for purposes approved and authorized by the business. This does make sense. It belongs to them. They pay for it. As part of your compensation they allow limited personal use with the understanding that you sacrifice privacy.
There is one (and to my knowledge only one) exception. The law explicitly permits limited non-interference use of corporate facilities for union and labor related purposes. This includes such things as discussion of working conditions, wages, etc. There are strict limits on when and how business facilities can be used, but when within these limits there are also very strong privacy protections.
For everything else, use a cell phone (on your break time), use your own personal equipment, do it from home.
If the URL matches something in the blocklist, then Junkbuster does not even attempt to load it. Instead, it sends an internally generated 404 that is also sometimes informative. This is better than playing games with DNS entries because most browsers handle 404 errors more gracefully. The DNS entry games don't always work as well.
So once you have "doubleclick.com" in your blocklist, absolutely anything on any doubleclick URL gets a 404 with no network traffic generated.
I can't answer for guidescope.
If you really feel the need, junkbuster has the option of sending wafers. These are cookies with some entertaining legalese that are designed to fit within the rules for a cookie. Or you can invent your own wafer.
But the issue with 1x1 web-bugs is not cookies. These web-bugs are already encoded with the tracking information so that the mere attempt to load the image provides the tracking information to the perpetrators.
There is also the information theoretical limit. The bandwidth used and the interfering noise can be combined to find the theoretical limit. People used to approximate Shannon's law by saying the limit was 1 bit/sec per Hz. For typical S/N ratios over things like analog lines using old modems this was not too far off. But as you go to better signalling systems the S/N ratios start allowing you to do better than that. Hence the ability to transfer 50+ Kb/sec over a nominal 4 KHz analog channel.
The S/N term is logarithmic and better telemetry systems and coding systems eventually run up against quantum walls, thermal noise limits, etc. that impose some fundamental limits on available S/N. So you still cannot escape Shannon's law.
The present bandwidth allocations are a sometimes crude attempt to manage the co-existance of different imperfect hardware, signalling, and coding systems. To a certain degree some signalling and coding systems can co-exist, e.g. spread spectrum signals hiding underneath the noise threshold for SSB signals. This kind of co-existance is permitted in very few bands. Most bands are allocated to a single telemetry and coding system so that the necessary guard bands, emission quality, and other constraints can be described easily and enforced easily. Real world hardware is imperfect, so you do need some saftey margins as well as guard bands due to signalling technology.
Up in the millimeter bands the attenuation due to molecular absorbtion has interesting effects. You get line of sight transmission that is relatively unaffected by rain and fog, blocked completely by most structures, and attenuates at 20-100db/km. This leads to some inherent excellent re-usability when used for short distance point to point links. It originally (and still) was used for other interesting things like covert radars and covert links that deliberatly operated near the peaks of molecular absorbtion bands.
There is a real danger in letting your desire for perfection interfere with taking reasonable safety precautions. Drug safety in the US is imperfect, but a whole lot better than in the "good old days" when patent medicines were severely toxic, drug toxicity reactions not publicized, etc.
Similar efforts to provide guidelines for working with Recombinant DNA, with hazardous materials, etc. have made significant improvements. The vast majority of the players want to behave responsibly and will follow reasonable guidelines. The fact that there are a few percent who will ignore the safety procedures does not make safety procedures irrelevant. They reduce the level of irresponsible behavior, and we can hope that the remaining hostile behavior is infrequent enough that we can deal with the resulting problems.
And you will find that most of the government players will also try to follow good safe practices, though there is always reason to doubt the wisdom of the TLAs.
There is already quite a lot of energy saved this way. CFL sales have been substantial for years and are still growing. The latest trends are in the natural spectrum (better light quality) phosphors and the low harmonics CFLs. (Harmonics cause "hidden" power loss in the grid and make reactance compensation more difficult.) The trick is to get those to be as efficient as the high harmonics and "blue" flourescents. Give them a few years and they will be there.
Also vaguely related is the growth in the use of LEDs for lighting. The red LED is more efficient at generating red than a flourescent, and much more efficient than an incandescent. This is why you are seeing more and more LED based stop lights. Even for white light, the combination LEDs are more efficient than incandescents. The present barriers to their use are:
1. Production costs are high (think 10-50 times more than incandescent).
2. Large surface areas are needed for bright light.
3. Proper color balance is very hard.
4. DC power is used, meaning special power supplies and their attendent efficiency loss. (Electronic equipment already need DC power, so they don't suffer this problem.)
So for a while LEDs will be mostly used in situations where color balance is unimportant, like red lights.
There is an exception of sorts, but one that probably does not apply. There are rules concerning employer interference with union activities, both existing unions and actions to establish a union presence. These do impose some restrictions on employer monitoring. They can still monitor, but not in a way that interferes with legitimate union activity. This can sometimes apply to monitoring of email.
This probably does not apply to the situation described, but it does sometimes apply.
For a similar US Army site, see http://carlisle-www.army.mil/usassi/ssipubs/stdypr og.htm. It has a collection of monographs on topics of more current interest. For example, an excellent discussion of relative capabilities and possible intentions of PRC vs. Taiwan.
A short examination of the law should ease your fears. The US Courts have generally held that there is no protection for speech that is intended to cause harm. You can point to both the Constitution, Supreme Court rulings, and many laws that illustrate this. Key examples are:
- Fraud. False speech intended to cause harm can be criminal. This extends into things like false advertising, etc.
- Libel and Slander, where again false speech intended to cause harm can be cause for civil actions.
The legal grey zones tend to be those where intent or harm is unclear, and those where intent may include political purposes. But even with political speech, the free speech protections are quite limited when harm is involved. The grey zones are in things like signs and sound trucks, where the harm is visual and aural pollution. When the harm reaches the level of property destruction or personal injury, the free speech protections are not exemptions from civil actions and criminal prosecution.
Dis-allowing root usage is an excellent idea for many applications. The fact that NFS by default makes root usage aggravating has acted as an anti-root effect in many networked environments. If popular user programs throw up nag screens saying "You should not be root" it has a significant effect. Programs like gimp and other GUI end-user programs should all include a little "am I root?" query followed by a nag message indicating that the user is probably running a mis-configured system.
The instruction definitions for MMX and 3DNow instructions are already there in gas. You can make use of them by means of the asm() feature in gcc and egcs. I've written several MMX-enhanced programs using egcs. But, these compilers will not themselves issue the egcs instructions. Also, you need to be careful to manually maintain the proper relationship with other FPU usage.
Problems with this approach are:
- gdb does not understand the FPU registers. Debugging MMX code is a real chore. You need to store things into memory before gdb can see them.
- It is up to you to decide when and how FPU registers need initialization.
- You are working in assembler and need to understand how to properly use asm(), __volatile__, and the like.
But it definitely works. I got reasonable speedups. MMX, 3DNow, etc. noticably inferior to AltiVec as an instruction set, but that has nothing to do with gcc and egcs. The asm() integration with the rest of the egcs compiler does make short MMX sequences quite reasonable. For longer code sequences it is better to write separate modules in assembler.
Both the comments here and in the article misconstrue one of the major complaints. This is due to a misinterpretation of the meaning of "error". For many of the older texts there is no single correct text. An important part of the text is the annotation that reveals the various different texts that have been found.
Shakespeare and the Bible are two of the most widely studied texts. Neither has a "correct" text. Every Shakespeare play has multiple originals that differ. The Bible as no originals left, just multiple copies. These copies differ in details. Some of the differences are copying errors, some are deliberate modifications, some are the translations, and some are the result of language evolution.
There are tools, like SGML markups, that allow the text to be annotated to reflect what is known about the history and evolution of a text. Project Guttenburg does not capture this information.
For some texts this is not important. For other texts this is important. Some readers will not care. Other readers will care. The ones that care are complaining about the loss.
IMPORTANT NOTE: This license is only for use within the educational system. That means:
- no personal use
- no commercial use
- debatable for use in home schooling or distance learning. These would at least need official sponsorship from an educational system.
Still, an astonishingly good license considering that it came from Microsoft.
The extent to which a police state reduces crime has been demonstrated in Eastern Europe and the Soviet Union. Removal of the constant surveillance, wiretaps, copier controls, computer controls, etc. has been tried. The crime rate skyrocketed. It was more than a small increase. Crime went from extremely rare up to Western European levels and (in Russia) above. We also have the genuine public reaction to all this. In Eastern Europe the return to Communism forces are real, but a minority. Given the choice of police state or widespread crime, the public prefers widespread crime. Only in Russia, where the government has joined up with the criminals, is a return to the police state viewed favorably by the general public.
The "hoax" was the result of combined ignorance, fear, cynicism, and venality.
Ignorance - The reporters cannot distinguish between an annoying computer failure and a catastrophic computer failure.
Fear - Most fear computers. They are a genuine job threat to many in the news business and to many people that they know. Most of their friends and colleagues are in the older, dying industries that are threatened by computers.
Cynicism - They know that they cannot trust the government or corporate spokesmen to tell the truth. So they did not believe the initial warnings. Then once they learned enough to understand that there was a real problem, they did not believe all the reports as the real problems were fixed. Early Y2K tests found genuine problems, this was reported in the technical channels, and the later tests found the problems to be fixed and this was reported in the technical channels. The media did not understand and did not trust the government and corporate reports.
Venality - The news media, especially the electronic media, sell a product. Emotion sells much more easily than reason. Whether on CNN or slashdot, the emotional response is the easy response and the universal response. The easy Y2K emotion is fear. Fear sells. Fear maintains ratings. Fear is universal. Reason requires lots of boring detail and lacks punch. So the reports all emphasized fear.
The best example of what really happened is from school. We all know people who don't start studying until the night before finals. They cram and pass. They don't start the term paper until the day before it is due. They pull an all-nighter and pass. This may even be the most efficient use of their resources (given their priorities). And a lot of people from the top to the bottom of corporations and government have not changed. They let Y2K slide until the last minute, then they crammed, pulled all-nighters, and passed.
You will not find servers or fame, in part because these attract script kiddies and vandals. But it does use those wasted CPU cycles.
The heirs of Dr King have sued for their copyright fees. They just have not won yet. The speech is copyrighted, the performance was copyrighted.
The issue with professor's notes is also covered by prior lawsuits. The question will be the extent to which the notes replicate the presentation. Capturing the information, but not the presentation, lets you avoid copyright issues. But if you capture the presentation also, there are copyright issues. None of the news reports provide enough information to judge whether these notes would violate copyright restrictions.
To be fair, I can and do use Linux as my primary desktop. And a mix of Linux and Solaris on production servers. But my "business desktop" remains NT. Because:
1.) Corporate file format requirements. I must be able to read, write, and modify documents that are shared activities. These documents are in MSWord format, MS Excell, MS Powerpoint, Visio, and PDF formats. I know and sometimes use Star Office. But it cannot modify MS documents without loss of formatting. There is no Visio. There is only PDF reader, not a full function PDF creator.
2.) Corporate communications requirements. I am required to have dial in and LAN access to Lotus Notes. No client support yet.
3.) Hardware variation support. Linux cannot support some of the highly integrated devices found in laptops and low-cost PC's. In my particular case it is a laptop. I don't get to pick the model. I have to take what the corporation provides. It is good quality, but has Linux problems. In general Linux support trails hardware availability by 6-12 months.
You can point in each case to a truthful "we are working on it". But working on it is not the same as available and robust today. These are reasons why I anticipate that within a few years Linux (and probably also *BSD) will be viable on the desktop. But viable in a few years is not the same as viable now.
Other people will have other particular problems, but the general categories of mandated file formats, mandated corporate communications, and hardware variations will keep coming up.