So first of all this seems really neat. In OS X Spotlight is the killer feature for me. When Spotlight works (finds my stuff while I type) I'm very happy, and when its laggy, it feels like my whole computer is freezing shut, so Spotlight is very important to my workflow. Now Ubuntu is taking it a step further than Spotlight, and is not just indexing applications and filenames, but web browser bookmarks, sub-commands in programs, menu items, and more. I think this is really great, and I would love to see this come back to Spotlight in OS X. This is exactly why I use Linux, to try out new stuff while the others are sitting on their hands. A lot of Mac users don't find Spotlight as useful as I do, but I liken that problem to what Steve Jobs said about people who can't type: the solution is to wait for those people to all die off:) Linux and her various distributions shouldn't wait to innovate, that's why I have always loved Linux, it doesn't wait.
When I was at T.J. Watson (as an intern), the other guys I worked with who were researching a new CPU architecture were roughly divided into two camps: people who were doing the same thing as at least 10-15 other people, and people who were doing something either completely unique, or working with just 2 or 3 others. The guys that worked with 10-15 people worked with IT to get the source control system (Jazz) and basic dev environment standardized, and working with relative 'ease'. The people who were on their own were working on advance projects, or very unfinished parts of the system, and consequently were liable for their _own_ IT. Basically there was a pyramid, broad support and common need at the bottom, and no support and total control at the top. There wasn't this Us vs. Them attitude: IT helped us by figuring out how to support popular parts of the project that had broad use, and didn't interfere with more complex and unfinished bits unless called upon (e.g., to help configure a personal MySQL server for some reason). I have had similar experiences at other good software/hardware shops, and have come to believe this is really how it ought to be.
Computer Science and software engineering are rife with themes. Many so-called inventions and 'new' ideas are applying these tried and true themes to a new permutation of some old problem. For instance, folding two loops together to reap stale items in a hash-table while simultaneously doing a query by iterating across a bucket list (a previous but recent slashdot patent posting). You can tell someone (a Judge) what JIT is, that it effectively combines caching of already-compiled code with partial compilation, but he can't appreciate that software engineering and computer science are pervaded by the concepts of caching, and right-sizing work. He can't possibly appreciate how obvious some of these 'inventions' are, and rank them fairly on a scale of truly inventive (LZW in my opinion) to 'someone-skilled-in-the-art-could-do-that' (twiddling bits in FAT to support short file-names). I think this is in general the primary source of frustration for engineers and scientists: that judges and patent clerks who really have no good sense of taste or knowlege on the matter make such important decisions. Redhat pointed out once that in the _vast_ majority of patent suits, the person being sued is never accused of actually _reading_ the patent, but infringing accidentally. People don't read software patents, so their claimed benefit of being able to publish great ideas by protecting them for the inventor is just bunk: society eats the bar while the inventor is anomolously protected for really no reason. They are basically landmines that only rich or organized people can buy, and most of the community knows it. Giving judges crash courses in Java is a promising start, but its also a depressing reminder of how far we have to go.
I think the author's assumption that people would search for "When are the Patriots playing next year?" rather than "patriots game schedule" is flat out wrong. People know they are using computers, and not talking to a person, and they compensate accordingly. Google therefore, also compensates accordingly, by finding every page on the internet with "patriots", "game", and "schedule" in some close proximity. They may (and probably do) do more, but Google's approach has always been index everything you possibly can, and NLP has always taken a back seat. The Bing folks on the other hand have explicitly tried to optimize for NLP cases. However which engine is better isn't a matter of can you ask it questions in English, but can someone find what they are looking for. Given that most people know that "Googling" is not the same as asking a question, it is not fair to only test NLP queries.
Many government functions that in real-life require authentication are fully online.
This is a good reason to have an ID system the government can use to identify citizens on-line. However the notion of enabling private parties to sue each other over comments made on websites on the grounds of slander is not a good idea. As you pointed out, the vast majority of these "slanderers" are children or inconsequential people, so one wonders why anyone would care? This seems reasonable given that in most other countries people generally disregard or ignore what is said about them in mere comments, and only sue actual concerns with a physical address (e.g., a newspaper) when they have to. Furthermore, it seems that the majority of such suits would be frivolous and used primarily as a means of censuring other people by intimidation. If the state were to use such a power (the party who you suggest should be suing your parent poster) it would literally be some form of an Anti-Sedition law, which is generally regarded as at best a war-time necessity, but at worst a gross violation of free speech. The next step would be the state having the ability to literally track and follow its citizens and observe what they say and post on the internet. If companies have to maintain this information, at best the state would only need a subpoena, and at worst, they would soon pass a law giving them a computer protocol to access it at will. Such an event is bound to occur when you have a country where companies are used to maintaining pools of people's real identities. You make this all sound as if we should think this is reasonable, but it just sounds alarming, and your defense of this type of law, even in its early form, is in many ways even more alarming. Are you sure you even want to _look_ down this road?
A lot of companies (i.e., Apple, Google, Oracle) use licensing that is "open source" in that the code can be used by anyone, however they are careful to always make sure that private modifications need not be published. Such companies avoid GNU like the plague, and only use it when forced (e.g., gcc). These companies then go further to make your stronger licensing ineffective by using DRM (e.g., Droid and TiVO) to make it even more unpleasant to hack their source base by depriving you of real control over the hardware. These guys use open source as a way to cheaply disseminate a platform they can advertise on, not as a movement or a service to the community.
With a license like that used in Google's user-land environment in Android, fixing patches only helps Verizon, Motorola, and Google, but the little guys won't see anything cool until Verizon and Google finds it unprofitable to maintain a separate fork any longer (which can be either short or long depending on the value). Even once said patches are published, good luck finding a cheap platform you can run it on that isn't locked down by your service provider. So there are huge disincentives for an unaffiliated hacker that go beyond mere access to the code. Rather than contributing to a movement, said hacker would just feel like a patsy that works for Google/Verizon for free.
Slashdot posts many stories related to Wikipedia, and in each of those stories related to its trustworthiness, it is endlessly repeated that you should follow the footnotes to where the source is cited, and do two things: (1) fact check the statement you are about to cite, and (2) fix it if its wrong. Using this simple strategy, you never have to invite the wrath of your teachers because you would never cite Wikipedia (but instead cite what it cites), and additionally you'd be doing mankind a service by keeping Wikipedia accurate. Yes there are vandals out there, but the operating principle of Wikipedia is that people interested in sharing truthful information far outweigh those that seek to vandalize and misrepresent it. If that is true then Wikipedia will continue to be right far more often that its not. This is a good thing, because at the rate the population is exploding, Wikipedia seems to be the only strategy of cataloging our cultural story that will scale. I don't see how 120 philosophers who probably have grants to write and classes to teach will find the time to do it.
I realize you are a teacher, and so please don't misunderstand, but when I was younger, learning to program was an _escape_ from the regular cirriculum that was forced upon me. I looked forward to sitting down and learning how to program, including scooping up any algorithms or math that would help me along the way. I wanted to write the first AI, and experiment with video games. Please please PLEASE don't teach students about bandwidth, how files are stored, or pretty much _anything_ that involvres paper, pencil, falling asleep in class, and tests. Even if its not supposed to be a programming class, just teach them how to program, the IT concepts will come automatically, and the students will be much happier.
You could go with Flash, python, or if you are daring, C/++.
I am a layman (not a mathematician) however there are several large points of suspicion that I can identify with this proof. First of all, its 102 pages long. Second of all, its a proof by contradiction, namely that certain known statistical behaviors of a formula are contradicted for the author's constructions if P=NP. So in reality, a proof like this requires not only examination of the particular proof in question, but of all other theorems and inferences that are relied upon to construct the contradiction as well. Given the already enormous length of the proof (102 pages), in addition to all related theorems and inferences (thousands of pages?) that must _also_ be correct, it will take a long time to 'verify'.
This is a good reason for computer checked proofs:
I agree with you, however its not just corporate and government press releases, but press releases from everyone, each individual. Twitter and blogging have made this possible. This is actually a good thing, not only is there much more information being freely offered up by individuals closest to the story, but it is in an electronic format that can be rapidly processed and aggregated, mashed up, or simply ignored.
The real issue is that news houses are turning into sweat-shop filters, finding contradictions and doing natural lang to squeeze relevant facts together into a (usually) coherent sentence. The next big mover in this industry is going to be someone who figures out how to collect data from Twitter and blogs, and then automatically find factual contradictions and put relevant facts together into an article. Such a news house could more easily afford then to send the humans off to ask questions, and conduct novel investigation into important matters.
If developers don't want to use the phone, the platform's potential will be limited to the imagination and the business model of the original vendor, which is usually very limited. Android phones like this one will be selected against. Users will want to 'unlock' their phone's power by clicking the install button in the windows program they download from that.org site everyone they know goes to, and phones that brick when they do this will eventually not be bought. Really its a stupid move for Motorola.
"CONAN: And again, we'd like to think of our brain as something that's been trained in, you know, Cartesian logic, when in fact, our brain is sort of hard-wired to leap to conclusions very quickly.
Mr. NYHAN: That's right. And what's interesting is in some of these cases, it's the people who are most sophisticated who are best able to defend their beliefs and keep coming up with more elaborate reasons"
I remember taking a neuroscience course in college once with a professor who had done experimentation that he thought suggested that what separated humans from other mammals (the cortex) was primarily a mechanism to _slow_ learning. In fact in studies I've read child apes are able to more quickly learn how to use tools than child humans. Humans are slow learners in the same way that a feedback control loop needs a dampener: it allows us to stabilize and converge on techniques and facts that serve us well without too easily 'forgetting' them.
WARNING: anecdotal evidence Walking and talking with people, the more 'reasonable' of us tend to simply be those that think about the issues (whatever they may be) more than others, and so misinformation in their minds will more quickly be 'flushed out'. However you don't _want_ people to just believe 'facts' without great trepidation, that is a good thing, its called skepticism, and it should be hard to overcome. Facts printed in news stories or articles (as mentioned in the article) are often wrong, like the countless stories that mis-reported the Toyota accelerator problem without doing their fact-checking first (one of the biggest proponents was a repeat insurance defrauder).
"Without a computer you have to learn how to think."
This was depressing to read. Apparently we have some amazing programmers, because they've figured out how to write programs that can think for us. If using a computer to do an internet search rather than tediously going to the library, or to perform a calculation rather than doing it out on paper is 'thinking for us', then we must not think of very interesting things at all.
Another post:
"similar to the classic homework before TV law"
Comparisons between computers and TV sets are non-starters. Computers are only interesting to the public at large because they either (1) let them play games, or (2) let them more easily interact with others (i.e., read news, look stuff up, social network, chat). In both cases computers seem to be infinitely healthier for the average kid than a TV set.
Kids naturally adapt to their environment, and today their environment is one where ubiquitous computing is on the rise, and the internet is real and really important. What course in public school grades you on computer/internet literacy? Was it included in the study? We are only seeing one side of the story here.
Many black-market 'recyling' out-fits ship their stuff to China. China is one of the worst offenders for re-selling and re-using recycled equipment. Towns in rural China are plagued with toxic chemicals seeping into their ground water. What's worse is the local governments, which are far from national oversight or are benefactors of nation indifference, abuse their power to take a cut of the profits and forbid foreign correspondents from filming the whole mess.
It is best to ratify a treaty only once you are prepared to enforce that treaty's obligations, rather than ratify and flagrantly ignore a treaty, otherwise the concept of having a treaty is worth nothing.
"because it's not true of scientists as a class, and therefore not true of Science. Science, by definition, is the removal of these weaknesses from the pursuit of knowledge."
I'm sorry, but as a practicing scientist that has to compete for funding like the rest I can tell you authoritatively that that is a load of bull. Just put out a new mission objective for NSF funding, and watch as all the academics out there find new ways to declare why their lab should receive funding so they can find the answer to X, Y, and Z. For instance NSF recently went ahead with a big push to find out how energy use in computing can be cut down. My adviser, who originally could care less about the environment, is now more gung-ho than ever to find waste in any aspect of computing, and now he has tons of NSF funding to go along with his enthusiasm. Will the hammer find the nails that it is looking for? Probably, publish or perish as they say.
"It's pretty easy for people to believe this, because we recognise that there's some of this in all of us."...because there is. A lot of people here align themselves with the global warming folks. They forget that folks need to really believe global warming before they are willing to make others lose their job over it. Therefore it is the responsibility of the scientist to make the _undeniable_ and _simple_ case to the public that global warming is (1) real, (2) can be averted or treated via this list of remedies, and (3) is not up for debate because of this list of compelling reasons. The ideas on the table (e.g., cap-n-trade in the US) for curbing global warming are drastic, draconian, and very scary. Any sort of forward motion on this agenda should be accompanied with reams of evidence that can't be so easily torn apart by a nest of skeptical lobbying vipers. If anything, that is the lowest standard that should be met before engaging on a major economic experiment.
Sadly however, some of the most important documents in support of global warming still reflect an immature method of scientific investigation, for instance, if even 5% of the citations in the (in?)famous IPCC report on climate change are wrong, misleading, or inconclusive, it raises serious doubts about the document's conclusions:
FAST 2009 has a paper on semantic data management using a file system built on top of an object store powered by MySQL. Performance isn't great, but it uses a distributed file system solution to solve the synchronization issue in a very nice way (e.g., synchronize all albums with my iPod, all photos with my laptop and computer, etc...). You can specify rules and I liked it when I heard about it. However performance is actually important, despite their claim:).
Perspective: Semantic Data Management for the Home
Brandon Salmon, Carnegie Mellon University; Steven W. Schlosser, Intel Research Pittsburgh; Lorrie Faith Cranor and Gregory R. Ganger, Carnegie Mellon University
HTML Paper
http://www.usenix.org/events/fast09/tech/full_papers/salmon/salmon_html/index.html
PDF Paper
http://www.usenix.org/events/fast09/tech/full_papers/salmon/salmon.pdf
Slides
http://www.usenix.org/events/fast09/tech/slides/salmon.pdf
Notice that almost all of the posts highlighted the "best decision" as hiring the right people, or finding co-founders that stuck with them. From what I've seen in the past, picking the right people is vital, its one of those things that is expected of you, and so we almost take it for granted, but if you don't do it right, you will mess the whole thing up.
Yes it is slow for a lot of things, and for a lot of other things it isn't. I have gotten very fast with it (and yes I develop too, but that doesn't have much to do with knowing about the UI). I surf, play games (not too much, but some times), and use apps. I generally load all the apps I plan on using right away (I do this on all systems), then use them, then close the computer and move on, simple. OSX lets me do it relatively fast, the uncustomizability bugs me, but it is still relatively fast.
Once the programs are up, I can work fast, but not being able to alt-tab between windows and alt-# between workspaces (and not having workspaces to begin with) bugs the hell out of me. THAT IS WHY I RUN ROOTLESS X. Rootless X is slow, but I use window maker, turn off all the options, and speed is unotacible (I'm a utilitarian guy), now I have workspaces, X, OSX, and anything else I want including Cocoa, Java, and OpenGL. I went back from 10.2 to 10.1.5, waiting for a better release (looks like since 10.2.2 just came out that time has come). In another year I see myself using linux, but while my computer is still relatively fast for its age, I'll continue to run the latest hot shit OSX system. Ultimately OSX works pretty much like any other graphics oriented bloat-ware system I have used (XP,2000,Magical Mandrake) speed wise, but runs much more stabily then any of those.
OSX crashes when I run OpenGL apps, and they freeze/lock or crash, otherwise it *NEVER* crashes on me. The trick with OSX is when you do media oriented things (then it shines), I notice a lot of speed with media and the sort, additionally it runs games very nicely (maybe not as fast as 95 or 98 of course, but it actually doesn't run all over your saved games with errors...). The trick with OSX (as with many a GUI) is to load all the programs you want right off the bat, put them in the backround, get done what you want done, close all the programs, and log off. If you load them right away, they are only an "alt-tab" (or two) away, and you never have to sit and wait for load, of course having a lot of memory helps this process, I run 1GB on my powerbook (not too expensive, life time warrenty, check out price watch).
The main advantage of an OSX system is you get it all. I can run anything from OpenGL games (like q3 or ut2k3 (if I want)) to office apps (MSOffice X, Open Office (almost)) to Xlib apps, to Cocoa apps, to fast simple free development (project builder, emacs, vim, make, whatever) and can do it almost as fast if not exactly as fast as windows, and still get the stability of unix! Like all things, it is a dynamic, there are comprimises, OSX represents one of those comprimises, one that fits me rather well for now, but it is not and can never be the alpha and the omega of computing, that you must find for yourself. NO reccomondation or advice or opinion will help you in that quest, only time tested experience, and practical observation.
Peace mac-heads.
I think GNU is a good thing if you want your code to be free and open to others, and if you don't mind that any modifications to that code is also free and open to others. This is not a good thing for the government. Tax payers also include evil horrible private corporations that provide the economic backbone of our country and millions upon millions of jobs. These people have a right to modify that code, expending THEIR resources, and to KEEP their modifications to themself and SECRET, they paid for it right? You and I also have the right to modify that code and KEEP THOSE MODIFICATIONS SECRET AND TO OURSELVES, we worked for it right? Code that is "public" by the government should be licensed under terms that say the ORIGINAL CODE that we funded with TAX PAYER money is availiable, and will ALWAYS be availiable to the public, and that is where it should end. No tax payer, or free-sourcer has the damn right to snoop my modifications that I paid for, not him, that isn't public, that is theft.
In a perfect world everybody would pay for everything, and book keeping would be free, you would still be paying licensing to Daimler Benz for the diezel engine, but it would only cost you less then a cent, no body would steal, business men would be honest, and the free market system would hum along perfectly well, but we do not live in a perfect world. This is why we have expirations on patents, copyrights, and other titles, and this is why we have public funding for some things people won't pay for. Unfortunately government funding is usually partial, so how do you handle cases where only a dollar of funding comes from the government? In a perfect world you would highlight the segments of code they paid for, and license it how the government wants, and license the rest how you want...yeah right.
The only reasonable solution I can imagine is if code that is paid for by more then 50% by the government can NOT be GNU'd. GNU is an invasive license that "sucks" in modifications made to the code. National parks don't suck up your tent into some "tree vortex" when you leave. We come in, pitch our tent, have fun with it, and when we are bored we leave, it is usually just that simple. GNU is much trickier.
Ultimately a GNU license would be a mistake for government, it "sucks" in modifications, it steals from tax payers, and it undermines the spirit of the government. As long as code modified that is GNU, ALSO remains GNU, the government should NOT allow this kind of licensing. The only kind of license the government should ever endorse is one where the ORIGINAL code, and JUST the original code, because this is all that was paid for, and all that should be provided, be availiable to the public (including businesses) for now and always.
I thought genius' were also supposed to be able to spell, but then someone like you probably doesn't concern "contraining" himself to the English language.
As a researcher in protein folding I can tell you from experience that neural nets have a lot in common with the way humans think (sometimes) and one property they both seem to share is the more related the input data is with other things, the faster the learning rate. Many know this as associativity. These researchers are trying to associate different pitches of bugs with different structual elements, and I think that is great, because now debugging will be faster because we can relate things to pitch as WELL as text, and think in terms of bugs much faster. Now we just need a visual language where the shapes of the objects are dependant upon those objects' methods and members...
My mac has NO floppydrive, and I am quite happy for it, but sad at the same because I miss the cheap and simple communication between two computers. I don't think PC manufacturers should phase out the floppy UNTIL they phase in a cheap wireless data protocol, like BLUETOOTH maybe, or making IR PORTS standard on PC TOWERS as well as laptops. Thats's my 3 and a half cents.
Ultimately, although the business model may or may not be flawed, it is still the responsibility of every DECENT man to resist the URGE to STEAL music. STEALING is taking something without permission from the OWNER, but then the question is, who owns it? The BlackSuits will always fight for the imprisonment of Free internet until the question of who owns the music at different stages is resolved. Until legislation is passed reforming internet ownership, we must check ourselves and stay in accordance with the will of the American People, and respect the CURRENT business model by not stealing property from their owners.
It would be possible except that each layer's transparency would have to be near 100% accurate. If there is any light diffusion, then the "backward" pixels in the z axis would become fuzzy and unclear after shining through so many layers. I assume you were implying a cube of plastic, maybe 1024x1024 with 1024 layers? The most backward pixels would have to go through 1024 layers unscathed! And still be clear enough to not overlap any of the other pixels just fractions of a millimeter away! Good luck either way I suppose.
Slashdot Mirror
So first of all this seems really neat. In OS X Spotlight is the killer feature for me. When Spotlight works (finds my stuff while I type) I'm very happy, and when its laggy, it feels like my whole computer is freezing shut, so Spotlight is very important to my workflow. Now Ubuntu is taking it a step further than Spotlight, and is not just indexing applications and filenames, but web browser bookmarks, sub-commands in programs, menu items, and more. I think this is really great, and I would love to see this come back to Spotlight in OS X. This is exactly why I use Linux, to try out new stuff while the others are sitting on their hands. A lot of Mac users don't find Spotlight as useful as I do, but I liken that problem to what Steve Jobs said about people who can't type: the solution is to wait for those people to all die off :) Linux and her various distributions shouldn't wait to innovate, that's why I have always loved Linux, it doesn't wait.
When I was at T.J. Watson (as an intern), the other guys I worked with who were researching a new CPU architecture were roughly divided into two camps: people who were doing the same thing as at least 10-15 other people, and people who were doing something either completely unique, or working with just 2 or 3 others. The guys that worked with 10-15 people worked with IT to get the source control system (Jazz) and basic dev environment standardized, and working with relative 'ease'. The people who were on their own were working on advance projects, or very unfinished parts of the system, and consequently were liable for their _own_ IT. Basically there was a pyramid, broad support and common need at the bottom, and no support and total control at the top. There wasn't this Us vs. Them attitude: IT helped us by figuring out how to support popular parts of the project that had broad use, and didn't interfere with more complex and unfinished bits unless called upon (e.g., to help configure a personal MySQL server for some reason). I have had similar experiences at other good software/hardware shops, and have come to believe this is really how it ought to be.
Computer Science and software engineering are rife with themes. Many so-called inventions and 'new' ideas are applying these tried and true themes to a new permutation of some old problem. For instance, folding two loops together to reap stale items in a hash-table while simultaneously doing a query by iterating across a bucket list (a previous but recent slashdot patent posting). You can tell someone (a Judge) what JIT is, that it effectively combines caching of already-compiled code with partial compilation, but he can't appreciate that software engineering and computer science are pervaded by the concepts of caching, and right-sizing work. He can't possibly appreciate how obvious some of these 'inventions' are, and rank them fairly on a scale of truly inventive (LZW in my opinion) to 'someone-skilled-in-the-art-could-do-that' (twiddling bits in FAT to support short file-names). I think this is in general the primary source of frustration for engineers and scientists: that judges and patent clerks who really have no good sense of taste or knowlege on the matter make such important decisions. Redhat pointed out once that in the _vast_ majority of patent suits, the person being sued is never accused of actually _reading_ the patent, but infringing accidentally. People don't read software patents, so their claimed benefit of being able to publish great ideas by protecting them for the inventor is just bunk: society eats the bar while the inventor is anomolously protected for really no reason. They are basically landmines that only rich or organized people can buy, and most of the community knows it. Giving judges crash courses in Java is a promising start, but its also a depressing reminder of how far we have to go.
I think the author's assumption that people would search for "When are the Patriots playing next year?" rather than "patriots game schedule" is flat out wrong. People know they are using computers, and not talking to a person, and they compensate accordingly. Google therefore, also compensates accordingly, by finding every page on the internet with "patriots", "game", and "schedule" in some close proximity. They may (and probably do) do more, but Google's approach has always been index everything you possibly can, and NLP has always taken a back seat. The Bing folks on the other hand have explicitly tried to optimize for NLP cases. However which engine is better isn't a matter of can you ask it questions in English, but can someone find what they are looking for. Given that most people know that "Googling" is not the same as asking a question, it is not fair to only test NLP queries.
Many government functions that in real-life require authentication are fully online.
This is a good reason to have an ID system the government can use to identify citizens on-line. However the notion of enabling private parties to sue each other over comments made on websites on the grounds of slander is not a good idea. As you pointed out, the vast majority of these "slanderers" are children or inconsequential people, so one wonders why anyone would care? This seems reasonable given that in most other countries people generally disregard or ignore what is said about them in mere comments, and only sue actual concerns with a physical address (e.g., a newspaper) when they have to. Furthermore, it seems that the majority of such suits would be frivolous and used primarily as a means of censuring other people by intimidation. If the state were to use such a power (the party who you suggest should be suing your parent poster) it would literally be some form of an Anti-Sedition law, which is generally regarded as at best a war-time necessity, but at worst a gross violation of free speech. The next step would be the state having the ability to literally track and follow its citizens and observe what they say and post on the internet. If companies have to maintain this information, at best the state would only need a subpoena, and at worst, they would soon pass a law giving them a computer protocol to access it at will. Such an event is bound to occur when you have a country where companies are used to maintaining pools of people's real identities. You make this all sound as if we should think this is reasonable, but it just sounds alarming, and your defense of this type of law, even in its early form, is in many ways even more alarming. Are you sure you even want to _look_ down this road?
A lot of companies (i.e., Apple, Google, Oracle) use licensing that is "open source" in that the code can be used by anyone, however they are careful to always make sure that private modifications need not be published. Such companies avoid GNU like the plague, and only use it when forced (e.g., gcc). These companies then go further to make your stronger licensing ineffective by using DRM (e.g., Droid and TiVO) to make it even more unpleasant to hack their source base by depriving you of real control over the hardware. These guys use open source as a way to cheaply disseminate a platform they can advertise on, not as a movement or a service to the community.
With a license like that used in Google's user-land environment in Android, fixing patches only helps Verizon, Motorola, and Google, but the little guys won't see anything cool until Verizon and Google finds it unprofitable to maintain a separate fork any longer (which can be either short or long depending on the value). Even once said patches are published, good luck finding a cheap platform you can run it on that isn't locked down by your service provider. So there are huge disincentives for an unaffiliated hacker that go beyond mere access to the code. Rather than contributing to a movement, said hacker would just feel like a patsy that works for Google/Verizon for free.
Slashdot posts many stories related to Wikipedia, and in each of those stories related to its trustworthiness, it is endlessly repeated that you should follow the footnotes to where the source is cited, and do two things: (1) fact check the statement you are about to cite, and (2) fix it if its wrong. Using this simple strategy, you never have to invite the wrath of your teachers because you would never cite Wikipedia (but instead cite what it cites), and additionally you'd be doing mankind a service by keeping Wikipedia accurate. Yes there are vandals out there, but the operating principle of Wikipedia is that people interested in sharing truthful information far outweigh those that seek to vandalize and misrepresent it. If that is true then Wikipedia will continue to be right far more often that its not. This is a good thing, because at the rate the population is exploding, Wikipedia seems to be the only strategy of cataloging our cultural story that will scale. I don't see how 120 philosophers who probably have grants to write and classes to teach will find the time to do it.
I realize you are a teacher, and so please don't misunderstand, but when I was younger, learning to program was an _escape_ from the regular cirriculum that was forced upon me. I looked forward to sitting down and learning how to program, including scooping up any algorithms or math that would help me along the way. I wanted to write the first AI, and experiment with video games. Please please PLEASE don't teach students about bandwidth, how files are stored, or pretty much _anything_ that involvres paper, pencil, falling asleep in class, and tests. Even if its not supposed to be a programming class, just teach them how to program, the IT concepts will come automatically, and the students will be much happier.
You could go with Flash, python, or if you are daring, C/++.
I am a layman (not a mathematician) however there are several large points of suspicion that I can identify with this proof. First of all, its 102 pages long. Second of all, its a proof by contradiction, namely that certain known statistical behaviors of a formula are contradicted for the author's constructions if P=NP. So in reality, a proof like this requires not only examination of the particular proof in question, but of all other theorems and inferences that are relied upon to construct the contradiction as well. Given the already enormous length of the proof (102 pages), in addition to all related theorems and inferences (thousands of pages?) that must _also_ be correct, it will take a long time to 'verify'.
This is a good reason for computer checked proofs:
http://www.zdnet.com/blog/emergingtech/computers-checking-mathematical-proofs/1087
I agree with you, however its not just corporate and government press releases, but press releases from everyone, each individual. Twitter and blogging have made this possible. This is actually a good thing, not only is there much more information being freely offered up by individuals closest to the story, but it is in an electronic format that can be rapidly processed and aggregated, mashed up, or simply ignored.
The real issue is that news houses are turning into sweat-shop filters, finding contradictions and doing natural lang to squeeze relevant facts together into a (usually) coherent sentence. The next big mover in this industry is going to be someone who figures out how to collect data from Twitter and blogs, and then automatically find factual contradictions and put relevant facts together into an article. Such a news house could more easily afford then to send the humans off to ask questions, and conduct novel investigation into important matters.
If developers don't want to use the phone, the platform's potential will be limited to the imagination and the business model of the original vendor, which is usually very limited. Android phones like this one will be selected against. Users will want to 'unlock' their phone's power by clicking the install button in the windows program they download from that .org site everyone they know goes to, and phones that brick when they do this will eventually not be bought. Really its a stupid move for Motorola.
Excerpt from the article:
"CONAN: And again, we'd like to think of our brain as something that's been trained in, you know, Cartesian logic, when in fact, our brain is sort of hard-wired to leap to conclusions very quickly.
Mr. NYHAN: That's right. And what's interesting is in some of these cases, it's the people who are most sophisticated who are best able to defend their beliefs and keep coming up with more elaborate reasons"
I remember taking a neuroscience course in college once with a professor who had done experimentation that he thought suggested that what separated humans from other mammals (the cortex) was primarily a mechanism to _slow_ learning. In fact in studies I've read child apes are able to more quickly learn how to use tools than child humans. Humans are slow learners in the same way that a feedback control loop needs a dampener: it allows us to stabilize and converge on techniques and facts that serve us well without too easily 'forgetting' them.
WARNING: anecdotal evidence
Walking and talking with people, the more 'reasonable' of us tend to simply be those that think about the issues (whatever they may be) more than others, and so misinformation in their minds will more quickly be 'flushed out'. However you don't _want_ people to just believe 'facts' without great trepidation, that is a good thing, its called skepticism, and it should be hard to overcome. Facts printed in news stories or articles (as mentioned in the article) are often wrong, like the countless stories that mis-reported the Toyota accelerator problem without doing their fact-checking first (one of the biggest proponents was a repeat insurance defrauder).
end anecdotal evidence
The first comment in the linked previous post:
"Without a computer you have to learn how to think."
This was depressing to read. Apparently we have some amazing programmers, because they've figured out how to write programs that can think for us. If using a computer to do an internet search rather than tediously going to the library, or to perform a calculation rather than doing it out on paper is 'thinking for us', then we must not think of very interesting things at all.
Another post:
"similar to the classic homework before TV law"
Comparisons between computers and TV sets are non-starters. Computers are only interesting to the public at large because they either (1) let them play games, or (2) let them more easily interact with others (i.e., read news, look stuff up, social network, chat). In both cases computers seem to be infinitely healthier for the average kid than a TV set.
Kids naturally adapt to their environment, and today their environment is one where ubiquitous computing is on the rise, and the internet is real and really important. What course in public school grades you on computer/internet literacy? Was it included in the study? We are only seeing one side of the story here.
The conclusion of the study is that kids
Many black-market 'recyling' out-fits ship their stuff to China. China is one of the worst offenders for re-selling and re-using recycled equipment. Towns in rural China are plagued with toxic chemicals seeping into their ground water. What's worse is the local governments, which are far from national oversight or are benefactors of nation indifference, abuse their power to take a cut of the profits and forbid foreign correspondents from filming the whole mess.
http://www.engadget.com/2008/11/10/video-chinas-toxic-wastelands-of-consumer-electronics-revealed/
It is best to ratify a treaty only once you are prepared to enforce that treaty's obligations, rather than ratify and flagrantly ignore a treaty, otherwise the concept of having a treaty is worth nothing.
"because it's not true of scientists as a class, and therefore not true of Science. Science, by definition, is the removal of these weaknesses from the pursuit of knowledge."
I'm sorry, but as a practicing scientist that has to compete for funding like the rest I can tell you authoritatively that that is a load of bull. Just put out a new mission objective for NSF funding, and watch as all the academics out there find new ways to declare why their lab should receive funding so they can find the answer to X, Y, and Z. For instance NSF recently went ahead with a big push to find out how energy use in computing can be cut down. My adviser, who originally could care less about the environment, is now more gung-ho than ever to find waste in any aspect of computing, and now he has tons of NSF funding to go along with his enthusiasm. Will the hammer find the nails that it is looking for? Probably, publish or perish as they say.
"It's pretty easy for people to believe this, because we recognise that there's some of this in all of us." ...because there is. A lot of people here align themselves with the global warming folks. They forget that folks need to really believe global warming before they are willing to make others lose their job over it. Therefore it is the responsibility of the scientist to make the _undeniable_ and _simple_ case to the public that global warming is (1) real, (2) can be averted or treated via this list of remedies, and (3) is not up for debate because of this list of compelling reasons. The ideas on the table (e.g., cap-n-trade in the US) for curbing global warming are drastic, draconian, and very scary. Any sort of forward motion on this agenda should be accompanied with reams of evidence that can't be so easily torn apart by a nest of skeptical lobbying vipers. If anything, that is the lowest standard that should be met before engaging on a major economic experiment.
Sadly however, some of the most important documents in support of global warming still reflect an immature method of scientific investigation, for instance, if even 5% of the citations in the (in?)famous IPCC report on climate change are wrong, misleading, or inconclusive, it raises serious doubts about the document's conclusions:
http://climatequotes.com/scientists/the-ipccs-questionable-citations/
FAST 2009 has a paper on semantic data management using a file system built on top of an object store powered by MySQL. Performance isn't great, but it uses a distributed file system solution to solve the synchronization issue in a very nice way (e.g., synchronize all albums with my iPod, all photos with my laptop and computer, etc...). You can specify rules and I liked it when I heard about it. However performance is actually important, despite their claim :).
Perspective: Semantic Data Management for the Home
Brandon Salmon, Carnegie Mellon University; Steven W. Schlosser, Intel Research Pittsburgh; Lorrie Faith Cranor and Gregory R. Ganger, Carnegie Mellon University
HTML Paper
http://www.usenix.org/events/fast09/tech/full_papers/salmon/salmon_html/index.html
PDF Paper
http://www.usenix.org/events/fast09/tech/full_papers/salmon/salmon.pdf
Slides
http://www.usenix.org/events/fast09/tech/slides/salmon.pdf
Notice that almost all of the posts highlighted the "best decision" as hiring the right people, or finding co-founders that stuck with them. From what I've seen in the past, picking the right people is vital, its one of those things that is expected of you, and so we almost take it for granted, but if you don't do it right, you will mess the whole thing up.
Once the programs are up, I can work fast, but not being able to alt-tab between windows and alt-# between workspaces (and not having workspaces to begin with) bugs the hell out of me. THAT IS WHY I RUN ROOTLESS X. Rootless X is slow, but I use window maker, turn off all the options, and speed is unotacible (I'm a utilitarian guy), now I have workspaces, X, OSX, and anything else I want including Cocoa, Java, and OpenGL. I went back from 10.2 to 10.1.5, waiting for a better release (looks like since 10.2.2 just came out that time has come). In another year I see myself using linux, but while my computer is still relatively fast for its age, I'll continue to run the latest hot shit OSX system. Ultimately OSX works pretty much like any other graphics oriented bloat-ware system I have used (XP,2000,Magical Mandrake) speed wise, but runs much more stabily then any of those.
OSX crashes when I run OpenGL apps, and they freeze/lock or crash, otherwise it *NEVER* crashes on me. The trick with OSX is when you do media oriented things (then it shines), I notice a lot of speed with media and the sort, additionally it runs games very nicely (maybe not as fast as 95 or 98 of course, but it actually doesn't run all over your saved games with errors...). The trick with OSX (as with many a GUI) is to load all the programs you want right off the bat, put them in the backround, get done what you want done, close all the programs, and log off. If you load them right away, they are only an "alt-tab" (or two) away, and you never have to sit and wait for load, of course having a lot of memory helps this process, I run 1GB on my powerbook (not too expensive, life time warrenty, check out price watch).
The main advantage of an OSX system is you get it all. I can run anything from OpenGL games (like q3 or ut2k3 (if I want)) to office apps (MSOffice X, Open Office (almost)) to Xlib apps, to Cocoa apps, to fast simple free development (project builder, emacs, vim, make, whatever) and can do it almost as fast if not exactly as fast as windows, and still get the stability of unix! Like all things, it is a dynamic, there are comprimises, OSX represents one of those comprimises, one that fits me rather well for now, but it is not and can never be the alpha and the omega of computing, that you must find for yourself. NO reccomondation or advice or opinion will help you in that quest, only time tested experience, and practical observation. Peace mac-heads.
In a perfect world everybody would pay for everything, and book keeping would be free, you would still be paying licensing to Daimler Benz for the diezel engine, but it would only cost you less then a cent, no body would steal, business men would be honest, and the free market system would hum along perfectly well, but we do not live in a perfect world. This is why we have expirations on patents, copyrights, and other titles, and this is why we have public funding for some things people won't pay for. Unfortunately government funding is usually partial, so how do you handle cases where only a dollar of funding comes from the government? In a perfect world you would highlight the segments of code they paid for, and license it how the government wants, and license the rest how you want...yeah right.
The only reasonable solution I can imagine is if code that is paid for by more then 50% by the government can NOT be GNU'd. GNU is an invasive license that "sucks" in modifications made to the code. National parks don't suck up your tent into some "tree vortex" when you leave. We come in, pitch our tent, have fun with it, and when we are bored we leave, it is usually just that simple. GNU is much trickier.
Ultimately a GNU license would be a mistake for government, it "sucks" in modifications, it steals from tax payers, and it undermines the spirit of the government. As long as code modified that is GNU, ALSO remains GNU, the government should NOT allow this kind of licensing. The only kind of license the government should ever endorse is one where the ORIGINAL code, and JUST the original code, because this is all that was paid for, and all that should be provided, be availiable to the public (including businesses) for now and always.
I thought genius' were also supposed to be able to spell, but then someone like you probably doesn't concern "contraining" himself to the English language.
Why not just do your principles and your peers a favor and stop trying to cheat the system?
As a researcher in protein folding I can tell you from experience that neural nets have a lot in common with the way humans think (sometimes) and one property they both seem to share is the more related the input data is with other things, the faster the learning rate. Many know this as associativity. These researchers are trying to associate different pitches of bugs with different structual elements, and I think that is great, because now debugging will be faster because we can relate things to pitch as WELL as text, and think in terms of bugs much faster. Now we just need a visual language where the shapes of the objects are dependant upon those objects' methods and members...
My mac has NO floppydrive, and I am quite happy for it, but sad at the same because I miss the cheap and simple communication between two computers. I don't think PC manufacturers should phase out the floppy UNTIL they phase in a cheap wireless data protocol, like BLUETOOTH maybe, or making IR PORTS standard on PC TOWERS as well as laptops. Thats's my 3 and a half cents.
Ultimately, although the business model may or may not be flawed, it is still the responsibility of every DECENT man to resist the URGE to STEAL music. STEALING is taking something without permission from the OWNER, but then the question is, who owns it? The BlackSuits will always fight for the imprisonment of Free internet until the question of who owns the music at different stages is resolved. Until legislation is passed reforming internet ownership, we must check ourselves and stay in accordance with the will of the American People, and respect the CURRENT business model by not stealing property from their owners.
It would be possible except that each layer's transparency would have to be near 100% accurate. If there is any light diffusion, then the "backward" pixels in the z axis would become fuzzy and unclear after shining through so many layers. I assume you were implying a cube of plastic, maybe 1024x1024 with 1024 layers? The most backward pixels would have to go through 1024 layers unscathed! And still be clear enough to not overlap any of the other pixels just fractions of a millimeter away! Good luck either way I suppose.