Another Cisco gadget that's cool as a cheap linux box is the NSLU2. For $80, you get a pretty full-featured Linux system. It's the size of a paperback, and draws a negligible amount of power. I use mine as a music server. There's a very lively and helpful user community on IRC. There are various options for modifying or replacing the system it ships with to get a more general-purpose linux box, running off of an external flash drive.
There's nothing particularly special about loop quantum gravity that makes it possible to avoid having a singularity at the big bang. Loop quantum gravity is just one theory of quantum gravity. The best known theory of quantum gravity is string theory. In pretty much any theory of quantum gravity, the classical picture of the big bang singularity is going to get heavily modified. The conditions of the big bang are pretty much the only conditions under which you really need a theory of quantum gravity (unless you're really clever about finding some other situation, like black hole evaporation, where quantum gravitational effects come in). In all theories of quantum gravity, there's a scale called the Planck scale, and when you go beyond that scale (e.g., the universe is hot enough so that the wavelengths of particles are on the order of the Planck length), mysterious stuff happens. Because of this, it's reasonably plausible that the big bang singularity is eliminated in any theory of quantum gravity.
Old attempts to make a theory of a rebounding big bang (with, e.g., a cyclic universe) had various technical problems, which have been solved in recent years. In a rebounding big bang, there are issues to worry about such as what happens to causality, entropy, and the thermodynamic arrow of time. E.g., you could imagine that a universe cycles through a series of big bangs, and that each cycle is a lot like the one before, or you could imagine that the second law of thermodynamics operates across rebounds, so that each cycle has more entropy than the one before. You could imagine that there could be cause and effect relationships extending across rebounds, or that that could be prevented by the laws of physics. Some people believe that there's an unsolved "entropy problem" in the current standard big bang theory. Here is a good FAQ about cyclic models.
Among [IT professionals], I get much more polarized attitudes - They either embrace it, or shun it (with reasons ranging from the "viral" licensing BS, to (yes, seriously) tirades about damned hippies trying to buck the system).
Some embrace it, some shun it... but still others have no clue that it exists. Last week I was in a meeting about the set of standard software to be put on a bunch of the lab machines at the college where I teach. The head of IT thought "GPL" was a piece of software that the faculty wanted installed.
The government is taxing people to provide education that is used by those same people, especially used moreso than the private schools.
You make some good points. But what about the phenomenon of the inner-city single mother who works two jobs so she can send her kids to a Catholic school? She's paying twice: once to support the horrible public schools, and once for the Catholic school. It's not like college admissions policies can magically fix the shortcomings of the public schools. If you go to a bad public school, you get a horrible education, and you're not prepared for a good college.
And they're getting the taxes mostly from people who are in a very good position to use this benefit anyway.
Here's where you really lost me. Say family A makes $40,000/yr, and that amounts to $30,000/yr after taxes. Now say family B makes $130,000/yr, which is $100,000/yr after tax. A has a much harder time affording their taxes than B does, because what's left to A is barely enough to live on. In return, A gets access to public schools that are little more than holding pens. If A really cares about their kids' education, then they'd be better off in a totally privatized system, where they wouldn't have to pay twice to get their kids into a decent school.
They also cut Italian, Latin literature, and French literature.
As a college teacher, I'm uncomfortable with the place that AP exams now occupy in our educational system. When I went to college, it was considered unusual to take AP exams, and nobody had ever heard of a GPA higher than 4.0. Now, with AP classes counting +1 on the GPA, Berkeley is turning away a sizable fraction of all students with 4.0 GPAs. In other words, you essentially can't get into the flagship schools of the UC system unless you have a lot of AP exams to puff up your grades. In one way this is good, because the old system encouraged kids not to take challenging coursework in high school. But a lot of rural and inner-city high schools don't offer AP courses, or don't offer more than one or two, or they offer them, but they're not at a high enough level to prepare you for the exams. There's something horribly wrong with a system of government that taxes working-class people in order to support public education, but effectively excludes their kids from getting the full benefit of the system they're supporting with their taxes.
Looking over the contents of the CS exams, I can't help getting the impression that this is vocational education masquerading as something more academic. It all seems to be focused on the OOP fad, and on being able to code in Java. Stacks and queues are only covered on the AB, not the A level!?!? The hardware part seems pretty lightweight, and there's virtually no theory AFAICT.
I'm not surprised that the grandparent decided to post anonymously. The only thing worse for your slashdot karma than criticizing Apple in a comment on an Apple story is to criticize BSD or Freenet in a comment on a BSD or Freenet story. The grandparent (who has now been modded down to -1, Troll) is factually correct. I tried out freenet several years ago, and poking around in the content that existed, it was extremely heavily weighted toward child pornography. Based on that observation, I made a personal decision that I didn't want to run a freenet node, because having my computer running as a freenet node meant I was contributing to that. Now we could have a reasoned debate about the issues. We could ask whether the individual has a responsibility not to contribute to this, or whether the individual is more like a common carrier. We could ask whether any government restrictions on free speech are morally and philosophically acceptable. We could talk about whether concern about child sexual abuse has turned into hysteria, and has resulted in bad legislation. We could make careful distinctions between government and private action against speech we disapprove of. Yes, we could do all these things, but we won't, because this thread is about Freenet, and therefore it will be heavily modded by people who are fans of Freenet. Ironically enough, Freenet users on Slashdot have shown unlimited willingness to use moderation to silence opposing points of view. How do I know? Because this isn't the first time I've sacrificed karma by trying to make a skeptical post about Freenet in slashdot comments on a Freenet story.
This is not entirely unreasonable. POD operations aimed at self-publishers tend to be flaky and unreliable about issues like quality control, packaging, and promptness in filling orders. Since most self-published books sell only a microscopic number of copies, I suspect Amazon is simply doing this as a way to stay away from business that creates lots of hassles and no significant profit.
TFA refers to PublishAmerica, which is an infamous author mill. I'm not crying any tears for them.
I've self-published some CC-licensed physics textbooks, and I've been reasonably happy with lulu, whose CEO was one of the founders of Red Hat. However, I think most of the people who buy one of lulu's distribution packages probably end up being sorry they did it, because it's just not typically realistic to hope for significant sales of a self-published book through the big retail channels. I just use their free package, where customers order directly from lulu. It's worked great for my needs: noncommercial project, with college bookstores as the customers.
You're conflating POD with self-publishing. Lots of big, established publishers use POD as one of their methods of production. It's not uncommon these days for a publisher to keep a novel in print in paperback by producing 300 units at a time via someone like Lightningsource.
I'll agree with you that self-publishing is full of scams. But: "This will net you a quality book!" Well, when you're talking about "quality" with respect to a novel, the big issues aren't layout and cover design, the real issue is whether the writing is any good. That has nothing to do with methods of production and everything to do with editorial standards.
Self-publishing can be fine, as long as you go into it with realistic expectations -- i.e., you don't expect to make any money. AFAICT, 99% of self-published books don't reach an audience. The other 1% reach an audience, but aren't profitable.
Trying to do OSS development on the Flash platform is kind of a nightmare in terms of licensing.
Re Flex, check out the EULA, e.g., "No Modifications, No Reverse Engineering." The swf spec says "a. You may not use the Specification in any way to create or develop a runtime, client, player, executable or other program that reads or renders SWF files." If you look at the list of codecs that are supported for Flash, or that may be supported in the near future, it's a mixture of totally proprietary codecs and others that are not quite as proprietary, but are not totally free and open either: mp3, a modified version of h.263, AAC audio, H.264 video, Nelly Moser. The EULA for the player says you can't modify it or reverse-engineer it, and can't run it on a portable device. As of a year ago, there were also a lot of compatibility and licensing issues with the Version 2 Components.
If you want to do totally OSS development on the flash platform, you can also do it using mtasc, haxe, and gnash. However, you then have to accept that mtasc supports an old version of actionscript, and haxe isn't the same language. I.e., you can't buy a flash book and expect to get the examples working.
IMO the two biggest issues facing the country right now are the war and the assault on civil liberties. Neither one of those, AFAICT, has been affected in any way by lobbyists, campaign contributions, or earmarks. Individual voters wanted security theater after 9/11, and that's what they got, at the expense of civil liberties. Bush got the war he wanted, not because of lobbyists or PACS, but because Congress is too spineless to ask hard questions. They were spineless about it because the idea of going to war was overwhelmingly popular with individual American voters. The basic problem is that absolute power corrupts absolutely, and our government is too big and powerful these days.
>>please feel free to go to the edge of a cliff and disbelieve in gravity
>Unfortunately, people with these beliefs vote, and often elect people with these beliefs, who set policies and enact laws that affect the scientific community as a whole. Whether or not it concerns cutting off funding for scientific research, or mandating stupid policies at the local school board level, you can't just dismiss these people. They will affect your life, sooner or later.
There's a pretty straightforward libertarian answer to this. If voters who are idiots have power to mess up your life, then it's an indication that government has too much power over your life.
In the public schools, the controversy about whether to teach evolution properly is just one minor example of a much broader issue, which is that many public schools suck in many ways, and the current government monopoly system makes it more difficult for, e.g., inner city parents to send their kids to Catholic schools, since they're getting money sucked out of their incomes to support the lousy public schools. I live in a neighborhood where there's a pretty significant number of Christian fundamentalists. I'm not under the illusion that government is going to convince their kids to believe in evolution. I don't think it's government's role to indoctrinate people into believing certain things. Realistically, the things kids end up growing up believing in are not all that strongly correlated with the views of their parents or their society anyway. I know a guy who was brought up in a fundamentalist family and is now a cancer researcher, and he's extremely antireligious, as a reaction to his background.
When it comes to scientific research, one thing I realized after working as a scientist for a while was that most science is really just pork-barrel politics. Most scientific papers are not even considered important by people in the relevant subfield. Their results are correct, but utterly inconsequential. The reason most papers get published is that people need a long list of publications in order to survive in the world of research. Anyway, the argument about educating voters pretty much fails when it comes to funding research, because the decisions about what kind of research to fund are made in extremely undemocratic ways -- which is a good thing. Funding decisions are mostly made based on peer review. If you want to talk about the total amount of money spent on scientific research, then that is somewhat more democratically decided, but I'm not convinced that it's correlated in any way with voters' beliefs about evolution and the Big Bang. And in any case, so much scientific research is utterly inconsequential that I really don't think the progress of science depends at all on funding levels, within wide margins. If you throw more money at research, you just run up against the limitations in the current state of the art, which determine what experiments are doable at a certain point in history. The main effect is merely to produce a temporary increase in the number of people employed as scientists, and to the extent that merit-based hiring really works (I think it mostly does), the increase in employment is going to be disproportionately among people who are second-rate talents.
I think global warming is the issue where the voting argument for education is the strongest. But realistically the scientific evidence just continues to accumulate, and it's getting to the point where the arguments about whether it's occurring and whether it's caused by humans are becoming as irrelevant as debates about whether evolution occurred. The real problem is that fighting global warming requires making doing things that will change people's lives in ways they won't like.
The group in Germany that did the experimental work specializes in doing measurements of pressures of ~100 GPa. It looks like they use diamond anvils, http://www.mpg.de/bilderBerichteDokumente/dokumentation/pressemitteilungen/2004/pressemitteilung200408022/index.html . So, okay, this would be a really earthshattering development if it led to superconductors that work at room temperature and at ordinary pressures, but it sounds like that may not happen. We already have superconductors that work at liquid nitrogen temperatures, and liquid nitrogen is as cheap as milk.
I use Linux because proprietary apps suck to high heaven, and if you want to run OSS (desktop) apps, Linux is by far the best system.
There's a horribly perverse system of incentives pervading the economics of proprietary apps. A user buying a proprietary GUI app typically has no way of knowing whether it's slow and/or buggy until he's already bought it. Performance is hard to judge until you have it loaded on your own system, and bugginess is hard to judge because the vendor does their best to keep bugs secret, and generally succeeds very well. Because buyers can't make decisions based on performance and quality, they tend to buy based on features. So vendors have a huge economic incentive to bloat their feature list, and push slow, buggy products out the door.
Two experiences that helped to sour me completely on proprietary software:
Bought a copy of Mathematica for my Mac back in the 90's. Upgraded to a new version of MacOS. Mathematica stopped working. Called Wolfram. They told me my only option was to buy a new version of Mathematica.
Bought Adobe PageMaker 6.5 (?) ca. 1997. Wrote a book in it. Found out it was horribly buggy, and was constantly corrupting files. Adobe's tech support came up with lots of excuses to explain why it wasn't their fault.
I teach physics at a community college. Recently I started working on a project to clean up the horribly messed up software situation in our student computer labs. Perfect example of what a mistake it can be to hitch your wagon to proprietary software. We have all these proprietary Windows apps. Every app has a different licensing scheme, and some of them have no explicitly stated licensing scheme at all (e.g., CD-ROMs that came with textbooks). Nobody can find half the original disks and licenses. Some software was bought to run on DOS or Windows 95, and isn't compatible with Windows XP. Some software is abandonware. In one case, faculty are downloading a particular piece of DOS abandonware/shareware from an untrusted third-party site every time they need to teach a particular activity -- can't ask IT to permanently install it, because the vendor is gone, so random people are just posting the.EXE on their web sites, without so much as a checksum. The whole thing is a nightmare.
I've published in PRL, back in the 90's. Basically what happened around then was that physicists were some of the earliest adopters of the internet and the web, and as soon as those tools became available, physicists started making their papers available to their colleagues for free in digital form. They still usually referred to them as "preprints," but in fact they'd still be sending them out after the paper had been accepted by the journal, the copyright transfer had been signed, and the paper had come out in print. Also in that era, arxiv.org was set up to archive preprints systematically. For decades now, arxiv has been a vital, ubiquitous part of the infrastructure of physics research; if arxiv is illegal, then I guess every single working physicist in the world is breaking the law every single working day of their career, because that's how much it gets used. The whole thing was sort of a blindingly obvious application for the internet. As an academic, what you care about is getting your research out there so that people know about it -- that's what builds your career. Nobody ever saw any conflict between the fact that (a) you assigned the copyright to the journal, and (b) you were still giving away copies. You might be able to argue that there was no legal conflict, because fair use applied, but realistically everybody saw it as a nonissue, because it was your own work you were giving out, and the journals were nonprofit entities.
What PRL should really reconsider is its whole policy of demanding copyright transfers. All they really need is a license from the author. This is a case where the legalities have lagged a couple of decades behind real-world practices. PRL is the most prestigious journal to get your work published in, but I think they realize that they're essentially expendable at this point at an institution; the minute a sufficient number of physicists get sufficiently upset with them, print journals can find itself replaced rapidly by open-access journals.
Virtually all submissions to PRL are done in LaTeX format, so there is no cost associated with typesetting. All the referees, and nearly all the editors, are unpaid. The printed format is basically obsolete, and the prices charged to libraries are simply ridiculous. This is a classic case where you just have an ossified institution that refuses to change.
Just go get the cheapest Windows PC you can find [...] and install Linux. It's cheaper than buying a dedicated Linux machine.
Huh? Over the last 5 years or so I've bought five or six low-end desktop machines for $180-250; they all had linux preinstalled. (I used to buy Great Quality machines from Fry's, but now Fry's has stopped carrying those, so the most recent time I did it I bought a Walmart gPC.) I have never, ever seen a desktop PC with Windows preinstalled that cost less than about $350-450.
Regular (not organic) LED lighting is reasonable for some applications already. The efficiency is very high. The light output is fairly low, but it's highly directional, so that's why you're starting to see it a lot in applications like flashlights. You can get LED bulbs that screw into a regular 110 V receptacle. I have an LED reading light next to my bed, and it works great; in the evening when I'm reading in bed, I use a low-wattage CF bulb to light up the room just enough so I don't feel like I'm in a cave, and the LED reading lamp puts enough extra light on the page so that I'm comfortable reading.
In addition, the point of the article is that on-site sales were poor, but on the other hand online sales were successful enough for Wal-Mart to continue selling Linux PCs
It isn't even a real change. The gPC was never really available on the shelf at WalMart. When I decided to buy one for my daughter, I messed around for weeks trying to find one in stores. Walmart has a web page where you can check for availability of items. I live in Southern California, and there are lots and lots of Walmarts within driving distance. None of them ever stocked it on the shelf. Just for grins, I checked other areas, like Chicago, San Francisco,... never did find a store that had it. Finally I realized that I was wasting my time, and just ordered it online, for delivery to the nearest store. Problem solved. So even if what they're saying is "we no longer officially offer it in stores," that's a nonissue, because they never did really offer it in stores.
Gather two anarchists in the same room, and you'll get two different definitions of anarchism. To me, Chomsky's definition sounds identical with democratic socialism, complete with a massive bureaucracy, and that's sure as hell not what most anarchists have in mind. Re "no laws," I think a better formulation would be "no coercion" -- that's probably closer to representing a consensus among people who call themselves anarchists. There may be a law that says you can't run red lights, but the penalty for violating it isn't going to be locking you up in a cage, it's going to be something noncoercive, e.g., people may shun you based on your reputation. I think libertarianism is a much better defined term, and the key distinction between anarchism and libertarianism is that anarchists don't want there to be private property. The messiness comes when you try to explain how your society is going to work when there's no such thing as private property. I think the most common answer is that the use of resources will be controlled by consensus within local communities, but, e.g., Chomsky seems to envision it as being controlled by a centralized state bureaucracy (which somehow manages to stay really, really faithful to the will of the grass roots -- which seems to me about as likely as the tooth fairy).
Yeah, the article is pretty incoherent. Hard to tell whether it's an incoherent summary of a coherent talk, or a correct summary of an incoherent talk.
One problem is that he talks about the internet as if it were a nation-state. The internet is a tool. Calling me a "netizen" is like saying that I'm a citizen of my screwdriver.
If a society is organized along centralized, authoritarian lines, then the problem isn't that that has a bad effect on the internet, the problem is that the whole society is screwed up. I care about whether there's free speech or not; the issue isn't free speech on the internet, it's free speech. I care about "the right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures;" the issue isn't whether TSA employees demand to paw through my laptop's email boxes, the issue is whether the bill of rights is being raped in general in the U.S. as a response to 9/11. If copyrights and patents are out of control, that's an issue for our society as a whole, not just for the internet.
Same goes with homework. If people want to copy each other's homework, who cares - they'll fail the exam anyway.
The problem with this philosophy is that achievement isn't binary, it's a continuum. The professor in this case was basing 10% of the grade on homework. Let's say the other 90% is based on tests, and let's say it's the traditional 90/80/70/60 grading scale. Student A gets 69% on the homework and 69% on the tests, so he gets a D in the course. Student B learns the same amount as student A, and also gets 69% on the tests, the difference being that while B is honestly able to do 69% of the homework, he copies the other 31% of it off of the facebook page without understanding it. Student B is going to pass the course with a C.
Congratulations, you've created a situation where honest students like A are penalized, and dishonest ones like B are rewarded.
I teach physics at a community college. I've tried not counting homework at all, and I've tried counting it for as much as 20% of the students' grades. Over time, I've found what seems to be work best empirically, which is counting it as about 10%. Counting it helps, for several reasons. One is that it provides a record of how the student is doing, and that way both I and the student know what the score is. Another is that many students need a lot of work developing their logical thinking skills, so on a problem that has conceptual aspects, they think any bogus nonsense they write down is a good answer, as long as it sounds like what's in the book. Those students need a homework grade as a wake-up call, so they realize before they get into the exam that they have a problem they need to fix.
In those circumstances, an attendance policy is not necessary. So when a class HAS an attendance policy anyway, then you know that attending class is probably a waste of your time, because if it wasn't, the professor wouldn't need to hold your grade hostage to get you to show up and listen to them drivel 3 hours a week.
Again, you might want to give professors some credit for being willing to learn from experience. I've tried requiring attendance, and I've tried ignoring it completely. My current policy is that I'll drop a student if he isn't attempting the written work (homework papers and 5-minute open-notes reading quizzes given at the beginning of classes for which reading was assigned), but if he wants to leave after the quiz, I don't penalize him. One thing you should consider is that not all professors follow the medieval practice of writing all the material on the board as if books cost ten years' wages and therefore were unavailable to the students. For someone like me who does a lot of activities in class that revolve around student participation rather than passive note-taking, it's death if you have half the students taking the course seriously, and the other half only showing up for every other class. If you let that happen, then you drop below the critical mass that you need if you want active-learning techniques to be successful.
Are you using the driver that that page recommends for that printer? There are often a few of them and one or more of them are often lousy (which is another bit of foolishness).
In the CUPS web interface, there are several of them, and one of them says "(recommended)" after the name. That's the one I'm using. I've tested some of the other ones, and none of them seem to be any better. I don't think the name given on the openprinting site actually correlates exactly with any of the names as listed in the CUPS web interface, but I've tried the ones that have names similar to that. Some of my problems may actually be CUPS problems, or problems with Ubuntu's packaging of cups, rather than problems with the driver per se -- it's hard to tell.
Interesting link, thanks! However, the printer I have, which has horribly low quality linux support, is a Brother HL-1440, which is listed in the "perfectly" category. So I think this doesn't fall in the out of date category, but it does fall in the category where you can't get accurate quality information.
Grandparent says: The card maker wants to be Windows only so don't buy it. Sooner or later hardware vendors will have to come around.
I see zero evidence that the assertion in the second sentence is true.
Parent says: Vendors and users will just stick with Windows if the Linux developers do not make it easy to find compatible items when they are hardware shopping. MS are tyrannical, but the Linux devs are indifferent (or perhaps setting up a site with simple hardware search form is beyond their technical ability?).
Undiplomatic, but basically correct, in my experience. It's way, way, way too hard to find Linux-compatible hardware. Any information you find on the web is almost guaranteed to be out of date, and therefore useless. You walk into Fry's and look around at the wifi cards on the shelf, and typically it's almost impossible to figure out whether a given card (a) has a GPL'd driver for linux, (b) has to be run with ndiswrapper, or (c) won't work on linux at all. The model number won't correlate with anything you can find on the web, because the web-based information is typically at least 6 months old, and the model number is typically less than 6 months old. You also can't get info on how good the linux support is. For instance, I have a Brother laser printer, and yes, it does work on linux, but no it doesn't work well (freezes up and has to be power-cycled, garbles output of mathematical equations, freaks out when used to print over my home network from my wife's mac,...) You're not going to get any of that kind of information on the web. It's not a question of spending more for the hardware that's linux-compatible; I would gladly spend more for linux-compatible hardware, but I can't, because there's no practical way for me to find out how good the compatibility is.
Buying peripherals for a desktop linux box is a tragedy of bad economics in the same way that buying proprietary software is a tragedy of bad economics. When you buy proprietary software, it's extremely difficult to know whether it's any good or not. Maybe you buy the program and it crashes all the time. Oh well, you had no way of knowing that was going to happen, and it's too late to get your money back. Because of that, proprietary software vendors don't compete on quality, they compete on features, and the result is that the quality of nearly all proprietary software is horrible. It's the same way with buying peripherals for a linux box. Users have no good way of knowing how well the peripheral will work on linux, so vendors have zero motivation to compete on linux compatibility. In an ideal world, linux users' leverage with the manufacturers would be small in proportion to linux's small popularity on the desktop. But it's not just small, it's zero, because linux users have no effective way of exerting their power in the marketplace.
Slashdot Mirror
Another Cisco gadget that's cool as a cheap linux box is the NSLU2. For $80, you get a pretty full-featured Linux system. It's the size of a paperback, and draws a negligible amount of power. I use mine as a music server. There's a very lively and helpful user community on IRC. There are various options for modifying or replacing the system it ships with to get a more general-purpose linux box, running off of an external flash drive.
There's nothing particularly special about loop quantum gravity that makes it possible to avoid having a singularity at the big bang. Loop quantum gravity is just one theory of quantum gravity. The best known theory of quantum gravity is string theory. In pretty much any theory of quantum gravity, the classical picture of the big bang singularity is going to get heavily modified. The conditions of the big bang are pretty much the only conditions under which you really need a theory of quantum gravity (unless you're really clever about finding some other situation, like black hole evaporation, where quantum gravitational effects come in). In all theories of quantum gravity, there's a scale called the Planck scale, and when you go beyond that scale (e.g., the universe is hot enough so that the wavelengths of particles are on the order of the Planck length), mysterious stuff happens. Because of this, it's reasonably plausible that the big bang singularity is eliminated in any theory of quantum gravity.
Old attempts to make a theory of a rebounding big bang (with, e.g., a cyclic universe) had various technical problems, which have been solved in recent years. In a rebounding big bang, there are issues to worry about such as what happens to causality, entropy, and the thermodynamic arrow of time. E.g., you could imagine that a universe cycles through a series of big bangs, and that each cycle is a lot like the one before, or you could imagine that the second law of thermodynamics operates across rebounds, so that each cycle has more entropy than the one before. You could imagine that there could be cause and effect relationships extending across rebounds, or that that could be prevented by the laws of physics. Some people believe that there's an unsolved "entropy problem" in the current standard big bang theory. Here is a good FAQ about cyclic models.
Among [IT professionals], I get much more polarized attitudes - They either embrace it, or shun it (with reasons ranging from the "viral" licensing BS, to (yes, seriously) tirades about damned hippies trying to buck the system). ... but still others have no clue that it exists. Last week I was in a meeting about the set of standard software to be put on a bunch of the lab machines at the college where I teach. The head of IT thought "GPL" was a piece of software that the faculty wanted installed.
Some embrace it, some shun it
The government is taxing people to provide education that is used by those same people, especially used moreso than the private schools.
You make some good points. But what about the phenomenon of the inner-city single mother who works two jobs so she can send her kids to a Catholic school? She's paying twice: once to support the horrible public schools, and once for the Catholic school. It's not like college admissions policies can magically fix the shortcomings of the public schools. If you go to a bad public school, you get a horrible education, and you're not prepared for a good college.
And they're getting the taxes mostly from people who are in a very good position to use this benefit anyway.
Here's where you really lost me. Say family A makes $40,000/yr, and that amounts to $30,000/yr after taxes. Now say family B makes $130,000/yr, which is $100,000/yr after tax. A has a much harder time affording their taxes than B does, because what's left to A is barely enough to live on. In return, A gets access to public schools that are little more than holding pens. If A really cares about their kids' education, then they'd be better off in a totally privatized system, where they wouldn't have to pay twice to get their kids into a decent school.
They also cut Italian, Latin literature, and French literature.
As a college teacher, I'm uncomfortable with the place that AP exams now occupy in our educational system. When I went to college, it was considered unusual to take AP exams, and nobody had ever heard of a GPA higher than 4.0. Now, with AP classes counting +1 on the GPA, Berkeley is turning away a sizable fraction of all students with 4.0 GPAs. In other words, you essentially can't get into the flagship schools of the UC system unless you have a lot of AP exams to puff up your grades. In one way this is good, because the old system encouraged kids not to take challenging coursework in high school. But a lot of rural and inner-city high schools don't offer AP courses, or don't offer more than one or two, or they offer them, but they're not at a high enough level to prepare you for the exams. There's something horribly wrong with a system of government that taxes working-class people in order to support public education, but effectively excludes their kids from getting the full benefit of the system they're supporting with their taxes.
Looking over the contents of the CS exams, I can't help getting the impression that this is vocational education masquerading as something more academic. It all seems to be focused on the OOP fad, and on being able to code in Java. Stacks and queues are only covered on the AB, not the A level!?!? The hardware part seems pretty lightweight, and there's virtually no theory AFAICT.
The David Lynch version wasn't just bad because it was compressed, it was simply a bad movie. The SciFi channel's version was actually pretty good.
I have my server set up to block access to IPs that do this sort of thing. In extreme cases, I've seen it bring down the server completely.
I'm not surprised that the grandparent decided to post anonymously. The only thing worse for your slashdot karma than criticizing Apple in a comment on an Apple story is to criticize BSD or Freenet in a comment on a BSD or Freenet story. The grandparent (who has now been modded down to -1, Troll) is factually correct. I tried out freenet several years ago, and poking around in the content that existed, it was extremely heavily weighted toward child pornography. Based on that observation, I made a personal decision that I didn't want to run a freenet node, because having my computer running as a freenet node meant I was contributing to that. Now we could have a reasoned debate about the issues. We could ask whether the individual has a responsibility not to contribute to this, or whether the individual is more like a common carrier. We could ask whether any government restrictions on free speech are morally and philosophically acceptable. We could talk about whether concern about child sexual abuse has turned into hysteria, and has resulted in bad legislation. We could make careful distinctions between government and private action against speech we disapprove of. Yes, we could do all these things, but we won't, because this thread is about Freenet, and therefore it will be heavily modded by people who are fans of Freenet. Ironically enough, Freenet users on Slashdot have shown unlimited willingness to use moderation to silence opposing points of view. How do I know? Because this isn't the first time I've sacrificed karma by trying to make a skeptical post about Freenet in slashdot comments on a Freenet story.
This is not entirely unreasonable. POD operations aimed at self-publishers tend to be flaky and unreliable about issues like quality control, packaging, and promptness in filling orders. Since most self-published books sell only a microscopic number of copies, I suspect Amazon is simply doing this as a way to stay away from business that creates lots of hassles and no significant profit.
TFA refers to PublishAmerica, which is an infamous author mill. I'm not crying any tears for them.
I've self-published some CC-licensed physics textbooks, and I've been reasonably happy with lulu, whose CEO was one of the founders of Red Hat. However, I think most of the people who buy one of lulu's distribution packages probably end up being sorry they did it, because it's just not typically realistic to hope for significant sales of a self-published book through the big retail channels. I just use their free package, where customers order directly from lulu. It's worked great for my needs: noncommercial project, with college bookstores as the customers.
You're conflating POD with self-publishing. Lots of big, established publishers use POD as one of their methods of production. It's not uncommon these days for a publisher to keep a novel in print in paperback by producing 300 units at a time via someone like Lightningsource.
I'll agree with you that self-publishing is full of scams. But: "This will net you a quality book!" Well, when you're talking about "quality" with respect to a novel, the big issues aren't layout and cover design, the real issue is whether the writing is any good. That has nothing to do with methods of production and everything to do with editorial standards.
Self-publishing can be fine, as long as you go into it with realistic expectations -- i.e., you don't expect to make any money. AFAICT, 99% of self-published books don't reach an audience. The other 1% reach an audience, but aren't profitable.
Trying to do OSS development on the Flash platform is kind of a nightmare in terms of licensing.
Re Flex, check out the EULA, e.g., "No Modifications, No Reverse Engineering." The swf spec says "a. You may not use the Specification in any way to create or develop a runtime, client, player, executable or other program that reads or renders SWF files." If you look at the list of codecs that are supported for Flash, or that may be supported in the near future, it's a mixture of totally proprietary codecs and others that are not quite as proprietary, but are not totally free and open either: mp3, a modified version of h.263, AAC audio, H.264 video, Nelly Moser. The EULA for the player says you can't modify it or reverse-engineer it, and can't run it on a portable device. As of a year ago, there were also a lot of compatibility and licensing issues with the Version 2 Components.
If you want to do totally OSS development on the flash platform, you can also do it using mtasc, haxe, and gnash. However, you then have to accept that mtasc supports an old version of actionscript, and haxe isn't the same language. I.e., you can't buy a flash book and expect to get the examples working.
IMO the two biggest issues facing the country right now are the war and the assault on civil liberties. Neither one of those, AFAICT, has been affected in any way by lobbyists, campaign contributions, or earmarks. Individual voters wanted security theater after 9/11, and that's what they got, at the expense of civil liberties. Bush got the war he wanted, not because of lobbyists or PACS, but because Congress is too spineless to ask hard questions. They were spineless about it because the idea of going to war was overwhelmingly popular with individual American voters. The basic problem is that absolute power corrupts absolutely, and our government is too big and powerful these days.
>>please feel free to go to the edge of a cliff and disbelieve in gravity
>Unfortunately, people with these beliefs vote, and often elect people with these beliefs, who set policies and enact laws that affect the scientific community as a whole. Whether or not it concerns cutting off funding for scientific research, or mandating stupid policies at the local school board level, you can't just dismiss these people. They will affect your life, sooner or later.
There's a pretty straightforward libertarian answer to this. If voters who are idiots have power to mess up your life, then it's an indication that government has too much power over your life.
In the public schools, the controversy about whether to teach evolution properly is just one minor example of a much broader issue, which is that many public schools suck in many ways, and the current government monopoly system makes it more difficult for, e.g., inner city parents to send their kids to Catholic schools, since they're getting money sucked out of their incomes to support the lousy public schools. I live in a neighborhood where there's a pretty significant number of Christian fundamentalists. I'm not under the illusion that government is going to convince their kids to believe in evolution. I don't think it's government's role to indoctrinate people into believing certain things. Realistically, the things kids end up growing up believing in are not all that strongly correlated with the views of their parents or their society anyway. I know a guy who was brought up in a fundamentalist family and is now a cancer researcher, and he's extremely antireligious, as a reaction to his background.
When it comes to scientific research, one thing I realized after working as a scientist for a while was that most science is really just pork-barrel politics. Most scientific papers are not even considered important by people in the relevant subfield. Their results are correct, but utterly inconsequential. The reason most papers get published is that people need a long list of publications in order to survive in the world of research. Anyway, the argument about educating voters pretty much fails when it comes to funding research, because the decisions about what kind of research to fund are made in extremely undemocratic ways -- which is a good thing. Funding decisions are mostly made based on peer review. If you want to talk about the total amount of money spent on scientific research, then that is somewhat more democratically decided, but I'm not convinced that it's correlated in any way with voters' beliefs about evolution and the Big Bang. And in any case, so much scientific research is utterly inconsequential that I really don't think the progress of science depends at all on funding levels, within wide margins. If you throw more money at research, you just run up against the limitations in the current state of the art, which determine what experiments are doable at a certain point in history. The main effect is merely to produce a temporary increase in the number of people employed as scientists, and to the extent that merit-based hiring really works (I think it mostly does), the increase in employment is going to be disproportionately among people who are second-rate talents.
I think global warming is the issue where the voting argument for education is the strongest. But realistically the scientific evidence just continues to accumulate, and it's getting to the point where the arguments about whether it's occurring and whether it's caused by humans are becoming as irrelevant as debates about whether evolution occurred. The real problem is that fighting global warming requires making doing things that will change people's lives in ways they won't like.
The group in Germany that did the experimental work specializes in doing measurements of pressures of ~100 GPa. It looks like they use diamond anvils, http://www.mpg.de/bilderBerichteDokumente/dokumentation/pressemitteilungen/2004/pressemitteilung200408022/index.html . So, okay, this would be a really earthshattering development if it led to superconductors that work at room temperature and at ordinary pressures, but it sounds like that may not happen. We already have superconductors that work at liquid nitrogen temperatures, and liquid nitrogen is as cheap as milk.
I use Linux because proprietary apps suck to high heaven, and if you want to run OSS (desktop) apps, Linux is by far the best system.
There's a horribly perverse system of incentives pervading the economics of proprietary apps. A user buying a proprietary GUI app typically has no way of knowing whether it's slow and/or buggy until he's already bought it. Performance is hard to judge until you have it loaded on your own system, and bugginess is hard to judge because the vendor does their best to keep bugs secret, and generally succeeds very well. Because buyers can't make decisions based on performance and quality, they tend to buy based on features. So vendors have a huge economic incentive to bloat their feature list, and push slow, buggy products out the door.
Two experiences that helped to sour me completely on proprietary software:
I teach physics at a community college. Recently I started working on a project to clean up the horribly messed up software situation in our student computer labs. Perfect example of what a mistake it can be to hitch your wagon to proprietary software. We have all these proprietary Windows apps. Every app has a different licensing scheme, and some of them have no explicitly stated licensing scheme at all (e.g., CD-ROMs that came with textbooks). Nobody can find half the original disks and licenses. Some software was bought to run on DOS or Windows 95, and isn't compatible with Windows XP. Some software is abandonware. In one case, faculty are downloading a particular piece of DOS abandonware/shareware from an untrusted third-party site every time they need to teach a particular activity -- can't ask IT to permanently install it, because the vendor is gone, so random people are just posting the .EXE on their web sites, without so much as a checksum. The whole thing is a nightmare.
I've published in PRL, back in the 90's. Basically what happened around then was that physicists were some of the earliest adopters of the internet and the web, and as soon as those tools became available, physicists started making their papers available to their colleagues for free in digital form. They still usually referred to them as "preprints," but in fact they'd still be sending them out after the paper had been accepted by the journal, the copyright transfer had been signed, and the paper had come out in print. Also in that era, arxiv.org was set up to archive preprints systematically. For decades now, arxiv has been a vital, ubiquitous part of the infrastructure of physics research; if arxiv is illegal, then I guess every single working physicist in the world is breaking the law every single working day of their career, because that's how much it gets used. The whole thing was sort of a blindingly obvious application for the internet. As an academic, what you care about is getting your research out there so that people know about it -- that's what builds your career. Nobody ever saw any conflict between the fact that (a) you assigned the copyright to the journal, and (b) you were still giving away copies. You might be able to argue that there was no legal conflict, because fair use applied, but realistically everybody saw it as a nonissue, because it was your own work you were giving out, and the journals were nonprofit entities.
What PRL should really reconsider is its whole policy of demanding copyright transfers. All they really need is a license from the author. This is a case where the legalities have lagged a couple of decades behind real-world practices. PRL is the most prestigious journal to get your work published in, but I think they realize that they're essentially expendable at this point at an institution; the minute a sufficient number of physicists get sufficiently upset with them, print journals can find itself replaced rapidly by open-access journals.
Virtually all submissions to PRL are done in LaTeX format, so there is no cost associated with typesetting. All the referees, and nearly all the editors, are unpaid. The printed format is basically obsolete, and the prices charged to libraries are simply ridiculous. This is a classic case where you just have an ossified institution that refuses to change.
Just go get the cheapest Windows PC you can find [...] and install Linux. It's cheaper than buying a dedicated Linux machine.
Huh? Over the last 5 years or so I've bought five or six low-end desktop machines for $180-250; they all had linux preinstalled. (I used to buy Great Quality machines from Fry's, but now Fry's has stopped carrying those, so the most recent time I did it I bought a Walmart gPC.) I have never, ever seen a desktop PC with Windows preinstalled that cost less than about $350-450.
Regular (not organic) LED lighting is reasonable for some applications already. The efficiency is very high. The light output is fairly low, but it's highly directional, so that's why you're starting to see it a lot in applications like flashlights. You can get LED bulbs that screw into a regular 110 V receptacle. I have an LED reading light next to my bed, and it works great; in the evening when I'm reading in bed, I use a low-wattage CF bulb to light up the room just enough so I don't feel like I'm in a cave, and the LED reading lamp puts enough extra light on the page so that I'm comfortable reading.
In addition, the point of the article is that on-site sales were poor, but on the other hand online sales were successful enough for Wal-Mart to continue selling Linux PCs ... never did find a store that had it. Finally I realized that I was wasting my time, and just ordered it online, for delivery to the nearest store. Problem solved. So even if what they're saying is "we no longer officially offer it in stores," that's a nonissue, because they never did really offer it in stores.
It isn't even a real change. The gPC was never really available on the shelf at WalMart. When I decided to buy one for my daughter, I messed around for weeks trying to find one in stores. Walmart has a web page where you can check for availability of items. I live in Southern California, and there are lots and lots of Walmarts within driving distance. None of them ever stocked it on the shelf. Just for grins, I checked other areas, like Chicago, San Francisco,
Gather two anarchists in the same room, and you'll get two different definitions of anarchism. To me, Chomsky's definition sounds identical with democratic socialism, complete with a massive bureaucracy, and that's sure as hell not what most anarchists have in mind. Re "no laws," I think a better formulation would be "no coercion" -- that's probably closer to representing a consensus among people who call themselves anarchists. There may be a law that says you can't run red lights, but the penalty for violating it isn't going to be locking you up in a cage, it's going to be something noncoercive, e.g., people may shun you based on your reputation. I think libertarianism is a much better defined term, and the key distinction between anarchism and libertarianism is that anarchists don't want there to be private property. The messiness comes when you try to explain how your society is going to work when there's no such thing as private property. I think the most common answer is that the use of resources will be controlled by consensus within local communities, but, e.g., Chomsky seems to envision it as being controlled by a centralized state bureaucracy (which somehow manages to stay really, really faithful to the will of the grass roots -- which seems to me about as likely as the tooth fairy).
Yeah, the article is pretty incoherent. Hard to tell whether it's an incoherent summary of a coherent talk, or a correct summary of an incoherent talk.
One problem is that he talks about the internet as if it were a nation-state. The internet is a tool. Calling me a "netizen" is like saying that I'm a citizen of my screwdriver.
If a society is organized along centralized, authoritarian lines, then the problem isn't that that has a bad effect on the internet, the problem is that the whole society is screwed up. I care about whether there's free speech or not; the issue isn't free speech on the internet, it's free speech. I care about "the right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures;" the issue isn't whether TSA employees demand to paw through my laptop's email boxes, the issue is whether the bill of rights is being raped in general in the U.S. as a response to 9/11. If copyrights and patents are out of control, that's an issue for our society as a whole, not just for the internet.
Same goes with homework. If people want to copy each other's homework, who cares - they'll fail the exam anyway.
The problem with this philosophy is that achievement isn't binary, it's a continuum. The professor in this case was basing 10% of the grade on homework. Let's say the other 90% is based on tests, and let's say it's the traditional 90/80/70/60 grading scale. Student A gets 69% on the homework and 69% on the tests, so he gets a D in the course. Student B learns the same amount as student A, and also gets 69% on the tests, the difference being that while B is honestly able to do 69% of the homework, he copies the other 31% of it off of the facebook page without understanding it. Student B is going to pass the course with a C.
Congratulations, you've created a situation where honest students like A are penalized, and dishonest ones like B are rewarded.
I teach physics at a community college. I've tried not counting homework at all, and I've tried counting it for as much as 20% of the students' grades. Over time, I've found what seems to be work best empirically, which is counting it as about 10%. Counting it helps, for several reasons. One is that it provides a record of how the student is doing, and that way both I and the student know what the score is. Another is that many students need a lot of work developing their logical thinking skills, so on a problem that has conceptual aspects, they think any bogus nonsense they write down is a good answer, as long as it sounds like what's in the book. Those students need a homework grade as a wake-up call, so they realize before they get into the exam that they have a problem they need to fix.
In those circumstances, an attendance policy is not necessary. So when a class HAS an attendance policy anyway, then you know that attending class is probably a waste of your time, because if it wasn't, the professor wouldn't need to hold your grade hostage to get you to show up and listen to them drivel 3 hours a week.
Again, you might want to give professors some credit for being willing to learn from experience. I've tried requiring attendance, and I've tried ignoring it completely. My current policy is that I'll drop a student if he isn't attempting the written work (homework papers and 5-minute open-notes reading quizzes given at the beginning of classes for which reading was assigned), but if he wants to leave after the quiz, I don't penalize him. One thing you should consider is that not all professors follow the medieval practice of writing all the material on the board as if books cost ten years' wages and therefore were unavailable to the students. For someone like me who does a lot of activities in class that revolve around student participation rather than passive note-taking, it's death if you have half the students taking the course seriously, and the other half only showing up for every other class. If you let that happen, then you drop below the critical mass that you need if you want active-learning techniques to be successful.
Are you using the driver that that page recommends for that printer? There are often a few of them and one or more of them are often lousy (which is another bit of foolishness).
In the CUPS web interface, there are several of them, and one of them says "(recommended)" after the name. That's the one I'm using. I've tested some of the other ones, and none of them seem to be any better. I don't think the name given on the openprinting site actually correlates exactly with any of the names as listed in the CUPS web interface, but I've tried the ones that have names similar to that. Some of my problems may actually be CUPS problems, or problems with Ubuntu's packaging of cups, rather than problems with the driver per se -- it's hard to tell.
Interesting link, thanks! However, the printer I have, which has horribly low quality linux support, is a Brother HL-1440, which is listed in the "perfectly" category. So I think this doesn't fall in the out of date category, but it does fall in the category where you can't get accurate quality information.
Grandparent says: The card maker wants to be Windows only so don't buy it. Sooner or later hardware vendors will have to come around.
I see zero evidence that the assertion in the second sentence is true.
Parent says: Vendors and users will just stick with Windows if the Linux developers do not make it easy to find compatible items when they are hardware shopping. MS are tyrannical, but the Linux devs are indifferent (or perhaps setting up a site with simple hardware search form is beyond their technical ability?). ...) You're not going to get any of that kind of information on the web. It's not a question of spending more for the hardware that's linux-compatible; I would gladly spend more for linux-compatible hardware, but I can't, because there's no practical way for me to find out how good the compatibility is.
Undiplomatic, but basically correct, in my experience. It's way, way, way too hard to find Linux-compatible hardware. Any information you find on the web is almost guaranteed to be out of date, and therefore useless. You walk into Fry's and look around at the wifi cards on the shelf, and typically it's almost impossible to figure out whether a given card (a) has a GPL'd driver for linux, (b) has to be run with ndiswrapper, or (c) won't work on linux at all. The model number won't correlate with anything you can find on the web, because the web-based information is typically at least 6 months old, and the model number is typically less than 6 months old. You also can't get info on how good the linux support is. For instance, I have a Brother laser printer, and yes, it does work on linux, but no it doesn't work well (freezes up and has to be power-cycled, garbles output of mathematical equations, freaks out when used to print over my home network from my wife's mac,
Buying peripherals for a desktop linux box is a tragedy of bad economics in the same way that buying proprietary software is a tragedy of bad economics. When you buy proprietary software, it's extremely difficult to know whether it's any good or not. Maybe you buy the program and it crashes all the time. Oh well, you had no way of knowing that was going to happen, and it's too late to get your money back. Because of that, proprietary software vendors don't compete on quality, they compete on features, and the result is that the quality of nearly all proprietary software is horrible. It's the same way with buying peripherals for a linux box. Users have no good way of knowing how well the peripheral will work on linux, so vendors have zero motivation to compete on linux compatibility. In an ideal world, linux users' leverage with the manufacturers would be small in proportion to linux's small popularity on the desktop. But it's not just small, it's zero, because linux users have no effective way of exerting their power in the marketplace.