That is an online transcription of Claude Shannon's thoughts on the matter. Google pointed to this link from Sharif, but you can see a printed version in the second volume of his collected works, I believe. I ran across this during my blissful graduate student days.
I love how this talk ends with him asking his audience to come and look at this machine that he built.
When I teach my courses in EECS at Berkeley, I always use the board. In fact, most of my colleagues whose offices are near mine all use the board most of the time. We know that students don't like powerpoint and it isn't good for them.
Many students need the (eye+ear)-to-brain-to-hand-to-eye-to-brain loop to internalize the material. Watching the Prof. write on the board engages some students taking notes better than the flash of instantly appearing powerpoint. Especially when diagrams and equations are involved.
Yes, it's a pain to walk out of lecture covered in chalk dust, but it's what is best for the students. After all, my anecdotal experience has been that those few students who get more out of seeing powerpoint slides than the blackboard also tend to get a lot out of just reading the textbook on their own and asking questions in office hours. While those that resonate the most with the blackboard tend to get less out of just reading the book on their own.
Since this is actually my research area, I thought it would be good to give some input here. Part of the controversy is simply due to the language used.
1) "Whitespace" is used in two subtly different senses by people that causes some confusion.
A) From the perspective of the potential new user of the spectrum, a "whitespace" is where the band is clean and so it could be used to deliver relatively high data-rate without having to put out too much transmit power relative to the desired range.
B) From the perspective of the existing user of that spectrum, the above perspective is troubling since it seems to ignore the externalities imposed by interference to others. The existing users' perspective is better captured by the idea of a "spectrum hole" that reflects where a new user could safely transmit without significantly bothering too many existing users. However, spectrum holes are also called "whitespaces" and this causes confusion.
The apparent weasel words "significantly" and "too many" above reflect a real set of engineering-tradeoffs underneath that must be navigated at least partially at the political level.
2) "Interference" is used by people in two different senses and this also causes confusion.
A) Interference is a purely technical concept that describes how performance degrades for a receiver with the introduction of additional signals into the environment. Here also there is some ambiguity because of a distinction between what would necessarily degrade performance even for an ideal or well-engineered receiver and what is feared to degrade the performance of possibly poorly designed or shoddily built receivers.
B) Interference is also an English word that encompasses uses like "you're interfering with my business model by offering a competing service."
Keep this in mind as you read any general articles about this subject. There are real tradeoffs involved in this topic, but sometimes the language used obscures or obfuscates them rather than making them clearer.
O.k. It is not often that I will post using my real name, but this seems to be a suitable opportunity since Cognitive Radio is actually one of my main research areas. (On a day I have mod points no less. Ah well.)
Here's the real deal:
The problem of ignoring rules and how to deal with this is a real and very important problem. A public policy researcher named Faulhaber coined a nice phrase for this: "Hit and run radios." This is *not* something that only effects "cognitive radios," this is an issue that must be confronted whenever we are dealing with radio transmitters that are capable of being frequency-agile. Even if there were a real-time market for spectrum, how do you ensure that radios pay for it? (Imagine a toll-road without a toll-booth)
Our group at Berkeley is one of those taking a serious look at the problem, but it is far from licked. A lot more needs to be done to make this workable, but from what we can see so far, it does not seem to be impossible.
I'll come out of the closet here. I have assigned this to my students in advanced courses as well. But I always make it optional. Students have a choice: write up lecture notes for one lecture to share with their fellow students in class or find an article related to the course material in Wikipedia and improve it substantially.
My experience has been that those that do this have made very nice contributions for the community. I check up on it to make sure that it is not confused. Of course, I have only tried this in the relatively small classes that we have here at Berkeley.
The academic world is about the developing and sharing of knowledge with our fellow human beings. Wikipedia seems like one of the right ways to do this for well established results with immediate benefits and very little pain.
Arghh... Stupid thinkpad trackpoint slipped from preview to submit. What I meant to say was:
For printed documentation, it is clear that there is a tradeoff between the configurability and documentability of a program. But computer-based documentation need not in principle suffer from this problem, especially if it is being served (or more ambitiously, displayed) by the same computer that holds the program.
Ideally, such online documentation should automatically update when the keystrokes, menu positions, or mouse-clicks governing a program change. EMACS help does a reasonable job of doing this when you ask it how to do something. What we need is a way for the documentation to symbolicly link to the functionality instead of just having a key combination explicitly written in. Then, the help browser could query the programs' preferences to insert the currently appropriate text. For Free documentation, we can imagine doing a recompile for printed documentation that would result in docs already customized for the particular site (or distribution) altered defaults.
We are in the free software community. Something like this should be easier for us to do than it is for proprietary developers who might fear exposing their internal function to an external help browser.
For printed documentation, it is clear that there is a tradeoff between the configurability and documentability of a program. But computer-based documentation need not in principle
We are in the free software community. It is clear that there is a need for a more sophisticated documentation system that will allow developers to
Let me start out by saying that I really like my Prius and have enjoyed driving it for over a year and a half now. But there are some shortcomings that I have noticed that nobody else has mentioned yet.
1) The milage on very short trips is substantially worse than the numbers quoted here. I live about 1.5 miles from where I work and sometimes drive my Prius instead of walking. On such trips the milage is in the low 40's or high 30's. It gets even worse for micro-trips like those taken by the parking attendants who double park my car when the lot is full and move it around when someone needs to get behind it.
This is a far cry from the 50+ it can get if it is warmed up and going on a long city drive. But long city drives are presumably quite rare as compared to short trips to the neighborhood grocery store, etc. So the city milage numbers they advertise are a bit misleading. The technical problem is that there is no waying of hinting to the car that this is only going to be a very short trip so don't bother trying to warm the car up. My milage average now hovers around 43 or so as compared to the around 48 I used to average when I commuted 30mi/day on the highway.
2) Using the air conditioner activates the engine even if the car is standing still. From my experience, it appears that it can not drive the AC from the battery.
I wish that the software for the engine was Free Software and that the car had an usb port or something so that enterprising folks could add hinting to the control of the car.
Go back and study the kind of opposing arguments that people used to use against democracy (we can't let the people have a say in the making of laws. better leave that to experts.), jury trials (we can't let the people decide what the facts are for a case. better leave that to experts.), the freedom of information acts (we can't let the people have access to government records. better leave them to the experts.), the abolition of slavery (taking away ownership rights from the plantation lords would remove the economic incentives to produce.), vulgate translations of scripture (we can't let the people read and interpret religious texts. better leave that to the experts.) and a host of other prior controvercies.
Then see how you can morph those arguments into arguments against Free Software. You'll find it an enlightening exercise.
This is admittedly a US-centric question since I realize that other countries have different attitudes towards patents. But here in the USA, our constitution says that only limited time exclusivity priveledges may be granted and that too only to "promote the progress of science and the useful arts" by encouraging the publication of what would otherwise be kept trade secrets. This strikes me as a reasonable attitude.
In the context of the World Wide Web, it seems that the standardization process itself enforces publication to ensure interoperability. And moreover, the network effects and gain in acceptance made possible by an impartial third party standardization process give a strong incentive to technology developers to participate in the standardization process, and thereby publish specifications at least at the interface level.
So any incentive that the patent monopoly might have provided for publication of the interface is much less necessary. Why does the committee think that requiring patent holders to grant a royalty free license to anyone implementing the standard interface is unreasonable in the context of the web? If they have any claims that broad, they clearly do not serve the public policy intentions of the patent process in the context of the WWW.
Developers are free to keep any innovative technical details of how they implement a product that meets that interface secret. They're also free to get a patent on their particular innovations (if they are truly innovative) in these implementations. In fact, it seems to make sense to require patent-holders to grant a royalty-free license to at least one particular unencumbered reference implementation of the complete standard. After all, if no such implementation exists, then the patent is again effectively on the standard itself.
And if no such implementation exists at the time of the standard's adoption, why does the committee feel it is unreasonable to require the patent holders to grant royalty free licenses to all free software developers who so generously offer to create such a reference implementation? (e.g. Mozilla)
The quote from Samuel Adams is just great, but I think that the lines elided in the quote are also worth thinking about. Here is the quote without any chunks taken out:
"Contemplate the mangled bodies of your countrymen, and then say, What should be the reward of such sacrifices? Bid us and our posterity bow the knee, supplicate the friendship, and plough, and sow, and reap, to glut the avarice of the men who have let loose on us the dogs of war to riot in our blood and hunt us from the face of the earth? If ye love wealth better than liberty, the tranquillity of servitude than the animating contest of freedom, - go from us in peace. We ask not your counsels or arms. Crouch down and lick the hands which feed you. May your chains sit lightly upon you, and may posterity forget that ye were our countrymen!"
The sections in bold above were elided in the quote from the article. I understand why the author dropped them, because he wanted to emphasize that accepting chains upon yourself is an act of fear and cowardice. The quote in its original form is a statement against tyranny and external oppression.
Now, if only we can be clear about identifying whose avarice let loose on us the dogs of war!
But now let's suppose that, after years of lobbying, messrs Kuhn and Stallman get a law passed that makes proprietary licenses illegal. We are now in the world of the FSF's premise.
As a user, my flerbage doesn't change. I never wanted to issue software under a proprietary license to begin with, so the new license doesn't touch me.
But as a developer, things are very different now. If I walk up to someone and offer them the same proprietary license that I did before the law was passed, police may come to my house to drag me off to jail, or kill me if I resist arrest. My flerbage has seriously decreased.
This rhetoric misleads people. I am not a lawyer, but as far as I know, an illegal contract is simply unenforceable. That means that if someone does not satisfy some illegal requirement in a contract, all that the government will do is ignore you when you come to them trying to enforce it.
Think about it this way. Suppose I agreed to a contract with someone that said that in exchange for him giving me a pen that I would pay him 10 dollars and convert to his religion. Suppose then that he gave me the pen, and I gave him the 10 dollars. But two weeks later, he sees me walking into a temple, rather than his own exclusivist church. He may be mad, but he has no legal recourse. If he tried to enforce the illegal term in the contract (the part about converting to his religion), the court would probably just say that an illegal term is not enforceable. We have the freedom to practice any religion we choose, even if we claim to have agreed to another one.
Would cops come to my door to force me to go to his church? No. Would cops go to his door and arrest him? Not unless he tried to threaten or mislead people into compliance with his illegal contract. They'd probably just laugh at the guy.
So in Eric's example where the unalianable right to distribute copies is explicitly recognized by statute, the developer who offered the proprietary license (presumably something which included a non-distribution clause) would just be laughed at if he tried to enforce this contract in court. Nobody would drag him off to jail.
There is a well established method for determining what is a truly novel and worthwhile contribution: peer review. The academic journals and conferences do it, and the basic principle is that the best people to evaluate a particular submission are those who actively do work in that area. It operates under the assumption that it is the professional responsibility of people to be willing to review and comment on the work of others. The right to do this could be guaranteed under law just like jury duty is protected from employers retaliation.
If we must have a patent system, then it seems necessary that the patent applications be peer reviewed by real practicioners and academics in their fields and not just by the underpaid patent clerks at the PTO. They're not all Albert Einstein!:-) By having people who are truly skilled in the specific art evaluating the patent applications, it would be much less likely for truly obvious things to get blocked by trivial patents. The knowledge that the patent applications are going to be subjected to such strict scrutiny would probably be enough to discourage some of the most egregious attempts to patent obvious things. This would make the volume of applications much more manageable as well.
Any additional costs imposed by this system could be recovered in the form of higher patent fees. At the current time, the US patent fees are a small fraction of the attorney fees that most of us have to pay to prepare the patent.
I'm a PhD myself, and so I can't really speak from personal experience on this topic. But a few of my friends are MBAs, and the best reason I've heard is from one of my female friends. This may not be very politically correct, but it is important to a lot of people.
She said that she got her MBA from a top school to have flexibility in her life going forward. Management isn't yet a science or even a vibrant craft that advances rapidly. The training that people receive in an MBA program isn't something that goes out of date very easily. MBAs are also in demand in different sectors of the economy and in all geographic regions. Engineers and scientists do not have that luxury. Our knowledge and skills get out of date quickly without continuous learning and practice.
Her reasoning was that when she got married and had kids, she'd like to be able to take a few years/decade of her life and be able to dedicate them to the hard work of raising her children. But after that, she'd like to be able to serve society in some sort of capacity by taking a paid position in a nearby company without having to uproot her family and disrupt her husband's career. She might not be making the big bucks, but at least she wouldn't be starting at rung 0 of the corporate ladder if she had the durable skills that an MBA certifies.
I guess that this reason is a big part of why a seemingly disproportionate number of female engineers decide to get MBAs, rather than "staying in engineering." It may not be relevant to the general Slashdot crowd, but we should keep in mind that other people have different priorities.
All persons born or naturalized in the United States and subject to the jurisdiction thereof, are citizens of the United States and of the State wherein they reside. No State shall make or enforce any law which shall abridge the privileges or immunities of citizens of the United States; nor shall any State deprive any person of life, liberty, or property,
without due process of law; nor deny to any person within its jurisdiction the equal protection of the laws.
An odd world indeed where you deny some of the citizens (the ones who can't afford to pay a private organization that presumably can price its "product" however it wishes!) the right to view, share, and circulate the law among themselves. At the very least, non-commercial distribution with any changes clearly marked ought to be allowed if we want the citizenry to have "equal protection of the laws."
Another player in the business of satellite launches is good for everyone. A lot of services are enabled by satellites, the failures of Iridium notwithstanding. I think that the current waiting list for satellite launches is quite long, and colored with political favoritism. More capacity would make it easier for new and innovative ideas to get their chance to shoot for the stars.
I only hope that India will not let the environment take a beating on the way up.
Privacy is a "negative" good; it represents the absence of something undesirable --- namely unseemly exposure. That it is subjective in what constitutes "unseemly" is mostly irrelevant.
If you look closely at the underpinnings of economic theory and the logic of market-based solutions to distribution and allocation problems, you will see why "negative" goods don't quite fit in. (Hint: look at marginal cost of production curves)
It is easier to understand this by considering a paradigmatic example of a "negative" good: A burly man holding a brick offers to sell you a contract guaranteeing that your window won't be broken. You value your window not being broken. So you pay the burly businessman what it is worth to you. A few blocks down, a slightly less affluent citizen decides that he would rather eat and be able to buy medicines. His window is broken.
Market based solutions are quite effective for a substantial range of "positive" goods. We also know how to implement them. But it would be silly mistake ease of implementation with appropriateness. Sorry folks. If you want to address the real situation, we're stuck with other more messy ways of distributed decision-making and allocation.
http://dmoz.org/Arts/ Lit erature/Authors/R/Rushdie,_Salman/ has got to be purged from the directory. After all, a lot of countries (including some so called democracies like India) have banned "The Satanic Verses" as being libelous to Islam.
Re-De-centralized Usenet
on
Deja For Sale
·
· Score: 2
Personally, I think that the Usenet archive could be very attractive to an ambitious "P2P" company that wants to show that they can re-de-centralize USENET archives while maintaining searchability. Volunteers might be brought in to help host the millions of posts themselves.
Distributed full-text searching, possibly with some sort of centralized assistance, but truly distributed access would make for a pretty mighty technology demo. I wonder if they're up to the task.
There is an SULEV rated Honda Accord EX available at dealers in California now, while the Toyota Prius is not yet available (though some dealers are taking orders). The Accord isn't a hybrid and so the milage alone is nothing to write home about, but the emissions (per mile! not per gallon...) are very competitive for everything but CO2.
...They both require someone or something to act upon the instructions before the task is complete. By outlawing all talk of assassination, including those within works of fiction [emphasis mine], we achieve the same level of protection that outlawing dissemination of DeCSS accomplishes.
So, what if someone wrote a short story which featured DeCSS code as an integral part of the story? As long as the original copyright holder (the person who figured out a way to decipher the CSS algorithm) granted rights to reproduce freely it would not be a copyright violation. And the short story could be written to clearly use the code in an artistic way to advance the plot line. (It could be about how a corrupt corporate conglomerate tries to stifle dissent).
Meanwhile, someone else could write a filter which would extract the compilable code from the short story itself. This could be written in a general way so as to apply to lots of "literate programs" and hence have nothing to do with DeCSS per se.
If a dedicated bunch of artistic people went around creating stories incorporating snippets of source code (Doesn't Stephenson's Cryptonomicon do that?), then I can't imagine how any sane legal mind could hold an entire genre of literature to be illegal to distribute.
If users in the privacy of their own homes wish to extract, recombine, and compile the source code, that is like someone making a private collage for their own house. If they want to execute the result and break some other law, then the MPAA can have fun suing the actual offenders instead of the artists. After all, if I use a collage incorporating a sharp point from someone else's sculpture to injure a third party, the offense is mine and mine alone.
Probalistic Analysis of Blacklists
on
Gnutella Vs. SPAM
·
· Score: 3
Blacklists are a very good initial approach, but can be thwarted by the use of probabalistic responses. Basically, a SPAMMER would only reply to any given query with some small probability $p$. Then, using the approach of "random query, then real query" will let the SPAMMER through with probability $(1-p)p$. By setting $p=0.5$, the SPAMMER can still get through with probability $0.25$.
Of course, by keeping a cumulative blacklist based on $n$ prior "random queries," the probability of success for a SPAMMER will go down as $p(1-p)^n$. By using $p=1/(n+1)$, the SPAMMER can keep his probability of success above $1/(n+1)^2$. This is bad since practical $n$ is limited by the rate at which the SPAMMER changes his server's identity.
The upshot is that any blacklist-based system is guaranteed to let a lot of SPAM through if the SPAMMERS are adaptive (change identities) and many in number (more than $(n+1)^2$ where $n$ is the effective time constant of adaptation). Collaborative blacklists can increase $n$, but introduce significant trust issues since adversaries can start blacklisting legitimate sites.
While I'm partial to compiling from source myself, there are some security concerns I could imagine with configuration and compilation scripts that are "overly intelligent." It would be hard to verify at a glance that they actually do what they are supposed to and don't in fact end up setting up a major back-door into your system. (It is a fundamental fact that very powerful systems always have the potential to generate inscrutable behaviour...)
A simple "put files X,Y,Z in locations A,B,C" packaging system is easier to verify as secure at a glance.
Perhaps this signals the need for an intermediate system. One which only has read-only access at first and proceeds to intelligently build a package of the simple type. Users can then verify that it doesn't do anything funny, and then let it place the files in their final locations. Does this sound a lot like a SRC RPM to you?;-) It should. But a little more care probably needs to be built into the system to prevent any source configuring scripts to write to the system at large.
I know, users should be smart enough to run this phase as a non-priveleged user, but I'm suggesting that this should be the default behaviour of the package management software and it should require explicit arguments to run configuration scripts as root.
There is so much for the future of Russian economy to be gained from the space program, its not even funny
Practically all the benefits (spinoffs and otherwise) you are talking about from the "Space Program" can come from unmanned exploration work as well. Much of the added cost and complexity of manned projects like the "Space Station" comes from the practical need to get failure probability down to near zero. None of us wants one of our comrades to lose his life up there. Yet this added cost and complexity *has few if any practical spinoffs* on Earth.
Space is probably about the only thing the Russians are good at, as an industrialized nation, right now
From a scientific perspective, this is truly sad. Russia has produced absolutely brilliant contributions to humanity in lots of fields. In my own area of research (probability and information theory) Russians like Kolmogorov, Dobrusin, and Pinsker have done seminal work which has advanced the frontiers of human knowledge. And you know what, this sort of work doesn't need Billions of dollars worth of equipment to do. Yet, in today's Russia, top notch scientists and mathemeticians are having a very hard time making a decent living for their families.
I would much rather that my country (the USA) support Russian science in more cost effective ways than this Space Station monstrosity. A Billion dollars buys a lot of science if spent right.
I'm in favor of space exploration as much as the next guy, but I can't believe that we're wasting so many resources on this useless thing. Robots are much more cost effective. I'd rather have systematic (unmanned) infrastructure in space for exploration.
It's one thing for the U.S.A. with its booming economy and basically working society, but Russia is a basket-case! Why are we wasting thousands of Russian minds on this inane task with absolutely no benefits to the Russian people or even to the state of human knowlege?
Slashdot Mirror
http://alum.sharif.edu/~mynaderi/Claude%20Shannon.html
That is an online transcription of Claude Shannon's thoughts on the matter. Google pointed to this link from Sharif, but you can see a printed version in the second volume of his collected works, I believe. I ran across this during my blissful graduate student days.
I love how this talk ends with him asking his audience to come and look at this machine that he built.
When I teach my courses in EECS at Berkeley, I always use the board. In fact, most of my colleagues whose offices are near mine all use the board most of the time. We know that students don't like powerpoint and it isn't good for them.
Many students need the (eye+ear)-to-brain-to-hand-to-eye-to-brain loop to internalize the material. Watching the Prof. write on the board engages some students taking notes better than the flash of instantly appearing powerpoint. Especially when diagrams and equations are involved.
Yes, it's a pain to walk out of lecture covered in chalk dust, but it's what is best for the students. After all, my anecdotal experience has been that those few students who get more out of seeing powerpoint slides than the blackboard also tend to get a lot out of just reading the textbook on their own and asking questions in office hours. While those that resonate the most with the blackboard tend to get less out of just reading the book on their own.
Since this is actually my research area, I thought it would be good to give some input here. Part of the controversy is simply due to the language used.
1) "Whitespace" is used in two subtly different senses by people that causes some confusion.
A) From the perspective of the potential new user of the spectrum, a "whitespace" is where the band is clean and so it could be used to deliver relatively high data-rate without having to put out too much transmit power relative to the desired range.
B) From the perspective of the existing user of that spectrum, the above perspective is troubling since it seems to ignore the externalities imposed by interference to others. The existing users' perspective is better captured by the idea of a "spectrum hole" that reflects where a new user could safely transmit without significantly bothering too many existing users. However, spectrum holes are also called "whitespaces" and this causes confusion.
The apparent weasel words "significantly" and "too many" above reflect a real set of engineering-tradeoffs underneath that must be navigated at least partially at the political level.
2) "Interference" is used by people in two different senses and this also causes confusion.
A) Interference is a purely technical concept that describes how performance degrades for a receiver with the introduction of additional signals into the environment. Here also there is some ambiguity because of a distinction between what would necessarily degrade performance even for an ideal or well-engineered receiver and what is feared to degrade the performance of possibly poorly designed or shoddily built receivers.
B) Interference is also an English word that encompasses uses like "you're interfering with my business model by offering a competing service."
Keep this in mind as you read any general articles about this subject. There are real tradeoffs involved in this topic, but sometimes the language used obscures or obfuscates them rather than making them clearer.
O.k. It is not often that I will post using my real name, but this seems to be a suitable opportunity since Cognitive Radio is actually one of my main research areas. (On a day I have mod points no less. Ah well.)
Here's the real deal:
The problem of ignoring rules and how to deal with this is a real and very important problem. A public policy researcher named Faulhaber coined a nice phrase for this: "Hit and run radios." This is *not* something that only effects "cognitive radios," this is an issue that must be confronted whenever we are dealing with radio transmitters that are capable of being frequency-agile. Even if there were a real-time market for spectrum, how do you ensure that radios pay for it? (Imagine a toll-road without a toll-booth)
Our group at Berkeley is one of those taking a serious look at the problem, but it is far from licked. A lot more needs to be done to make this workable, but from what we can see so far, it does not seem to be impossible.
I'll come out of the closet here. I have assigned this to my students in advanced courses as well. But I always make it optional. Students have a choice: write up lecture notes for one lecture to share with their fellow students in class or find an article related to the course material in Wikipedia and improve it substantially.
My experience has been that those that do this have made very nice contributions for the community. I check up on it to make sure that it is not confused. Of course, I have only tried this in the relatively small classes that we have here at Berkeley.
The academic world is about the developing and sharing of knowledge with our fellow human beings. Wikipedia seems like one of the right ways to do this for well established results with immediate benefits and very little pain.
Arghh... Stupid thinkpad trackpoint slipped from preview to submit. What I meant to say was:
For printed documentation, it is clear that there is a tradeoff between the configurability and documentability of a program. But computer-based documentation need not in principle suffer from this problem, especially if it is being served (or more ambitiously, displayed) by the same computer that holds the program.
Ideally, such online documentation should automatically update when the keystrokes, menu positions, or mouse-clicks governing a program change. EMACS help does a reasonable job of doing this when you ask it how to do something. What we need is a way for the documentation to symbolicly link to the functionality instead of just having a key combination explicitly written in. Then, the help browser could query the programs' preferences to insert the currently appropriate text. For Free documentation, we can imagine doing a recompile for printed documentation that would result in docs already customized for the particular site (or distribution) altered defaults.
We are in the free software community. Something like this should be easier for us to do than it is for proprietary developers who might fear exposing their internal function to an external help browser.
For printed documentation, it is clear that there is a tradeoff between the configurability and documentability of a program. But computer-based documentation need not in principle
We are in the free software community. It is clear that there is a need for a more sophisticated documentation system that will allow developers to
Let me start out by saying that I really like my Prius and have enjoyed driving it for over a year and a half now. But there are some shortcomings that I have noticed that nobody else has mentioned yet.
1) The milage on very short trips is substantially worse than the numbers quoted here. I live about 1.5 miles from where I work and sometimes drive my Prius instead of walking. On such trips the milage is in the low 40's or high 30's. It gets even worse for micro-trips like those taken by the parking attendants who double park my car when the lot is full and move it around when someone needs to get behind it.
This is a far cry from the 50+ it can get if it is warmed up and going on a long city drive. But long city drives are presumably quite rare as compared to short trips to the neighborhood grocery store, etc. So the city milage numbers they advertise are a bit misleading. The technical problem is that there is no waying of hinting to the car that this is only going to be a very short trip so don't bother trying to warm the car up. My milage average now hovers around 43 or so as compared to the around 48 I used to average when I commuted 30mi/day on the highway.
2) Using the air conditioner activates the engine even if the car is standing still. From my experience, it appears that it can not drive the AC from the battery.
I wish that the software for the engine was Free Software and that the car had an usb port or something so that enterprising folks could add hinting to the control of the car.
Go back and study the kind of opposing arguments that people used to use against democracy (we can't let the people have a say in the making of laws. better leave that to experts.), jury trials (we can't let the people decide what the facts are for a case. better leave that to experts.), the freedom of information acts (we can't let the people have access to government records. better leave them to the experts.), the abolition of slavery (taking away ownership rights from the plantation lords would remove the economic incentives to produce.), vulgate translations of scripture (we can't let the people read and interpret religious texts. better leave that to the experts.) and a host of other prior controvercies.
Then see how you can morph those arguments into arguments against Free Software. You'll find it an enlightening exercise.
This is admittedly a US-centric question since I realize that other countries have different attitudes towards patents. But here in the USA, our constitution says that only limited time exclusivity priveledges may be granted and that too only to "promote the progress of science and the useful arts" by encouraging the publication of what would otherwise be kept trade secrets. This strikes me as a reasonable attitude.
In the context of the World Wide Web, it seems that the standardization process itself enforces publication to ensure interoperability. And moreover, the network effects and gain in acceptance made possible by an impartial third party standardization process give a strong incentive to technology developers to participate in the standardization process, and thereby publish specifications at least at the interface level.
So any incentive that the patent monopoly might have provided for publication of the interface is much less necessary. Why does the committee think that requiring patent holders to grant a royalty free license to anyone implementing the standard interface is unreasonable in the context of the web? If they have any claims that broad, they clearly do not serve the public policy intentions of the patent process in the context of the WWW.
Developers are free to keep any innovative technical details of how they implement a product that meets that interface secret. They're also free to get a patent on their particular innovations (if they are truly innovative) in these implementations. In fact, it seems to make sense to require patent-holders to grant a royalty-free license to at least one particular unencumbered reference implementation of the complete standard. After all, if no such implementation exists, then the patent is again effectively on the standard itself.
And if no such implementation exists at the time of the standard's adoption, why does the committee feel it is unreasonable to require the patent holders to grant royalty free licenses to all free software developers who so generously offer to create such a reference implementation? (e.g. Mozilla)
The quote from Samuel Adams is just great, but I think that the lines elided in the quote are also worth thinking about. Here is the quote without any chunks taken out:
"Contemplate the mangled bodies of your countrymen, and then say, What should be the reward of such sacrifices? Bid us and our posterity bow the knee, supplicate the friendship, and plough, and sow, and reap, to glut the avarice of the men who have let loose on us the dogs of war to riot in our blood and hunt us from the face of the earth? If ye love wealth better than liberty, the tranquillity of servitude than the animating contest of freedom, - go from us in peace. We ask not your counsels or arms. Crouch down and lick the hands which feed you. May your chains sit lightly upon you, and may posterity forget that ye were our countrymen!"
The sections in bold above were elided in the quote from the article. I understand why the author dropped them, because he wanted to emphasize that accepting chains upon yourself is an act of fear and cowardice. The quote in its original form is a statement against tyranny and external oppression.
Now, if only we can be clear about identifying whose avarice let loose on us the dogs of war!
Eric writes:
But now let's suppose that, after years of lobbying, messrs Kuhn and Stallman get a law passed that makes proprietary licenses illegal. We are now in the world of the FSF's premise.
As a user, my flerbage doesn't change. I never wanted to issue software under a proprietary license to begin with, so the new license doesn't touch me.
But as a developer, things are very different now. If I walk up to someone and offer them the same proprietary license that I did before the law was passed, police may come to my house to drag me off to jail, or kill me if I resist arrest. My flerbage has seriously decreased.
This rhetoric misleads people. I am not a lawyer, but as far as I know, an illegal contract is simply unenforceable. That means that if someone does not satisfy some illegal requirement in a contract, all that the government will do is ignore you when you come to them trying to enforce it.
Think about it this way. Suppose I agreed to a contract with someone that said that in exchange for him giving me a pen that I would pay him 10 dollars and convert to his religion. Suppose then that he gave me the pen, and I gave him the 10 dollars. But two weeks later, he sees me walking into a temple, rather than his own exclusivist church. He may be mad, but he has no legal recourse. If he tried to enforce the illegal term in the contract (the part about converting to his religion), the court would probably just say that an illegal term is not enforceable. We have the freedom to practice any religion we choose, even if we claim to have agreed to another one.
Would cops come to my door to force me to go to his church? No. Would cops go to his door and arrest him? Not unless he tried to threaten or mislead people into compliance with his illegal contract. They'd probably just laugh at the guy.
So in Eric's example where the unalianable right to distribute copies is explicitly recognized by statute, the developer who offered the proprietary license (presumably something which included a non-distribution clause) would just be laughed at if he tried to enforce this contract in court. Nobody would drag him off to jail.
No loss of flerbage here.
There is a well established method for determining what is a truly novel and worthwhile contribution: peer review. The academic journals and conferences do it, and the basic principle is that the best people to evaluate a particular submission are those who actively do work in that area. It operates under the assumption that it is the professional responsibility of people to be willing to review and comment on the work of others. The right to do this could be guaranteed under law just like jury duty is protected from employers retaliation.
:-) By having people who are truly skilled in the specific art evaluating the patent applications, it would be much less likely for truly obvious things to get blocked by trivial patents. The knowledge that the patent applications are going to be subjected to such strict scrutiny would probably be enough to discourage some of the most egregious attempts to patent obvious things. This would make the volume of applications much more manageable as well.
If we must have a patent system, then it seems necessary that the patent applications be peer reviewed by real practicioners and academics in their fields and not just by the underpaid patent clerks at the PTO. They're not all Albert Einstein!
Any additional costs imposed by this system could be recovered in the form of higher patent fees. At the current time, the US patent fees are a small fraction of the attorney fees that most of us have to pay to prepare the patent.
I'm a PhD myself, and so I can't really speak from personal experience on this topic. But a few of my friends are MBAs, and the best reason I've heard is from one of my female friends. This may not be very politically correct, but it is important to a lot of people.
She said that she got her MBA from a top school to have flexibility in her life going forward. Management isn't yet a science or even a vibrant craft that advances rapidly. The training that people receive in an MBA program isn't something that goes out of date very easily. MBAs are also in demand in different sectors of the economy and in all geographic regions. Engineers and scientists do not have that luxury. Our knowledge and skills get out of date quickly without continuous learning and practice.
Her reasoning was that when she got married and had kids, she'd like to be able to take a few years/decade of her life and be able to dedicate them to the hard work of raising her children. But after that, she'd like to be able to serve society in some sort of capacity by taking a paid position in a nearby company without having to uproot her family and disrupt her husband's career. She might not be making the big bucks, but at least she wouldn't be starting at rung 0 of the corporate ladder if she had the durable skills that an MBA certifies.
I guess that this reason is a big part of why a seemingly disproportionate number of female engineers decide to get MBAs, rather than "staying in engineering." It may not be relevant to the general Slashdot crowd, but we should keep in mind that other people have different priorities.
An odd world indeed where you deny some of the citizens (the ones who can't afford to pay a private organization that presumably can price its "product" however it wishes!) the right to view, share, and circulate the law among themselves. At the very least, non-commercial distribution with any changes clearly marked ought to be allowed if we want the citizenry to have "equal protection of the laws."
Another player in the business of satellite launches is good for everyone. A lot of services are enabled by satellites, the failures of Iridium notwithstanding. I think that the current waiting list for satellite launches is quite long, and colored with political favoritism. More capacity would make it easier for new and innovative ideas to get their chance to shoot for the stars.
I only hope that India will not let the environment take a beating on the way up.
Privacy is a "negative" good; it represents the absence of something undesirable --- namely unseemly exposure. That it is subjective in what constitutes "unseemly" is mostly irrelevant.
If you look closely at the underpinnings of economic theory and the logic of market-based solutions to distribution and allocation problems, you will see why "negative" goods don't quite fit in. (Hint: look at marginal cost of production curves)
It is easier to understand this by considering a paradigmatic example of a "negative" good: A burly man holding a brick offers to sell you a contract guaranteeing that your window won't be broken. You value your window not being broken. So you pay the burly businessman what it is worth to you. A few blocks down, a slightly less affluent citizen decides that he would rather eat and be able to buy medicines. His window is broken.
Market based solutions are quite effective for a substantial range of "positive" goods. We also know how to implement them. But it would be silly mistake ease of implementation with appropriateness. Sorry folks. If you want to address the real situation, we're stuck with other more messy ways of distributed decision-making and allocation.
http://dmoz.org/Arts/ Lit erature/Authors/R/Rushdie,_Salman/ has got to be purged from the directory. After all, a lot of countries (including some so called democracies like India) have banned "The Satanic Verses" as being libelous to Islam.
Personally, I think that the Usenet archive could be very attractive to an ambitious "P2P" company that wants to show that they can re-de-centralize USENET archives while maintaining searchability. Volunteers might be brought in to help host the millions of posts themselves.
Distributed full-text searching, possibly with some sort of centralized assistance, but truly distributed access would make for a pretty mighty technology demo. I wonder if they're up to the task.
There is an SULEV rated Honda Accord EX available at dealers in California now, while the Toyota Prius is not yet available (though some dealers are taking orders). The Accord isn't a hybrid and so the milage alone is nothing to write home about, but the emissions (per mile! not per gallon...) are very competitive for everything but CO2.
Some information about Honda's environmental impact can be found here at Honda's official website. A PDF file describing the SULEV spec is available from the transportation website of the Union of Concerned Scientists.
The only question is why all the other states are lagging behind California! The technology is out there, and cleaner air benefits everyone.
So, what if someone wrote a short story which featured DeCSS code as an integral part of the story? As long as the original copyright holder (the person who figured out a way to decipher the CSS algorithm) granted rights to reproduce freely it would not be a copyright violation. And the short story could be written to clearly use the code in an artistic way to advance the plot line. (It could be about how a corrupt corporate conglomerate tries to stifle dissent).
Meanwhile, someone else could write a filter which would extract the compilable code from the short story itself. This could be written in a general way so as to apply to lots of "literate programs" and hence have nothing to do with DeCSS per se.
If a dedicated bunch of artistic people went around creating stories incorporating snippets of source code (Doesn't Stephenson's Cryptonomicon do that?), then I can't imagine how any sane legal mind could hold an entire genre of literature to be illegal to distribute.
If users in the privacy of their own homes wish to extract, recombine, and compile the source code, that is like someone making a private collage for their own house. If they want to execute the result and break some other law, then the MPAA can have fun suing the actual offenders instead of the artists. After all, if I use a collage incorporating a sharp point from someone else's sculpture to injure a third party, the offense is mine and mine alone.
Blacklists are a very good initial approach, but can be thwarted by the use of probabalistic responses. Basically, a SPAMMER would only reply to any given query with some small probability $p$. Then, using the approach of "random query, then real query" will let the SPAMMER through with probability $(1-p)p$. By setting $p=0.5$, the SPAMMER can still get through with probability $0.25$.
Of course, by keeping a cumulative blacklist based on $n$ prior "random queries," the probability of success for a SPAMMER will go down as $p(1-p)^n$. By using $p=1/(n+1)$, the SPAMMER can keep his probability of success above $1/(n+1)^2$. This is bad since practical $n$ is limited by the rate at which the SPAMMER changes his server's identity.
The upshot is that any blacklist-based system is guaranteed to let a lot of SPAM through if the SPAMMERS are adaptive (change identities) and many in number (more than $(n+1)^2$ where $n$ is the effective time constant of adaptation). Collaborative blacklists can increase $n$, but introduce significant trust issues since adversaries can start blacklisting legitimate sites.
While I'm partial to compiling from source myself, there are some security concerns I could imagine with configuration and compilation scripts that are "overly intelligent." It would be hard to verify at a glance that they actually do what they are supposed to and don't in fact end up setting up a major back-door into your system. (It is a fundamental fact that very powerful systems always have the potential to generate inscrutable behaviour...)
;-) It should. But a little more care probably needs to be built into the system to prevent any source configuring scripts to write to the system at large.
A simple "put files X,Y,Z in locations A,B,C" packaging system is easier to verify as secure at a glance.
Perhaps this signals the need for an intermediate system. One which only has read-only access at first and proceeds to intelligently build a package of the simple type. Users can then verify that it doesn't do anything funny, and then let it place the files in their final locations. Does this sound a lot like a SRC RPM to you?
I know, users should be smart enough to run this phase as a non-priveleged user, but I'm suggesting that this should be the default behaviour of the package management software and it should require explicit arguments to run configuration scripts as root.
There is so much for the future of Russian economy to be gained from the space program, its not even funny
Practically all the benefits (spinoffs and otherwise) you are talking about from the "Space Program" can come from unmanned exploration work as well. Much of the added cost and complexity of manned projects like the "Space Station" comes from the practical need to get failure probability down to near zero. None of us wants one of our comrades to lose his life up there. Yet this added cost and complexity *has few if any practical spinoffs* on Earth.
Space is probably about the only thing the Russians are good at, as an industrialized nation, right now
From a scientific perspective, this is truly sad. Russia has produced absolutely brilliant contributions to humanity in lots of fields. In my own area of research (probability and information theory) Russians like Kolmogorov, Dobrusin, and Pinsker have done seminal work which has advanced the frontiers of human knowledge. And you know what, this sort of work doesn't need Billions of dollars worth of equipment to do. Yet, in today's Russia, top notch scientists and mathemeticians are having a very hard time making a decent living for their families.
I would much rather that my country (the USA) support Russian science in more cost effective ways than this Space Station monstrosity. A Billion dollars buys a lot of science if spent right.
I'm in favor of space exploration as
much as the next guy, but I can't believe
that we're wasting so many resources on
this useless thing. Robots are much more
cost effective. I'd rather have systematic
(unmanned) infrastructure in space for
exploration.
It's one thing for the U.S.A. with its
booming economy and basically working
society, but Russia is a basket-case!
Why are we wasting thousands of Russian
minds on this inane task with absolutely
no benefits to the Russian people or even
to the state of human knowlege?