I'm not so such that the particular DRM scheme that the GPL prohibits is one that kernel developers would want to support. It's prohibiting selling devices which have GPL firmware but require binaries signed with a private key that isn't included in the source, so that people can't actually install modified versions. The current draft seems to prohibit any systems which use signing and have a public key whose matching private key isn't included, but they've said that they want to fix this issue in later drafts. (So that, for example, GNU TLS could give RMS's public key as an example without giving his private key as well) Most likely, the result will be that a program which includes a public key without the corresponding private key must be modifiable to replace the public key with a different one.
The other anti-DRM measure is that it includes a denial of the magic statement in the DMCA, such that, in case anybody thought that you could stop somebody from defeating your GPLv3-licensed copy protection by suing them under the DMCA, they're wrong. Of course, a GPL-licensed copy protection scheme is going to be easy to defeat, anyway, since all versions of the license require that the user be able to modify the code to remove it, so it couldn't really work as a practical matter. Of course, some level of DRM is fine: the user of a program should be able to prevent other people from getting the data; sending encrypted documents and maintaining privacy is a fine use for GPL software. The point is that it is the person who runs the program who can choose whether or not to include each check, not some vendor or other entity.
So as a practical matter, the only situation in which DRM and the GPL could be used together was when a single system had a GPL portion and an immutable, vendor-chosen portion, and the vendor-chosen portion has the ability to inspect the GPL portion for changes. This isn't something that anyone who releases code under the GPL is likely to want to encourage, although there's a slim chance that people would choose the GPL over the BSD license out of curiousity, hoping to see the source to modified and distributed versions they can't actually run.
It wouldn't have to change the user's settings. It would only have to make the information available to people's phones. People could set their phone to have a special ring for when the user is at a show, much like special rings based on caller ID. And phones could come pre-programmed with a silent ring for this condition regardless of caller. So you don't change the phone settings; a single configuration would cause the phone to behave differently in response to where you happened to be.
That is, however, an optional part of the license, so anyone intending to release somethign as Free Software can simply choose not to use it. I believe I've seen that debian legal finally decided that it's okay for the GFDL to support invariant sections, so long as documents debian ships have no such sections. (But other issues with the GFDL are more serious and apply to all documents.) I'd guess that no project team that wants to keep from being kicked out would accept the addition of an invariant section to their project, given that it would require debian to fork their project.
Of course, if you consider the users of a web service program to be the people running the web service rather than the people using the service provided by those people, then it is fundamentally non-free to put any restriction on what the people running the service can do. (Like writing a compiler that can't be used to compile certain programs would be, or writing a word processor which required the produced documents to be put in the public domain.)
I'm a bit mystified as to why they didn't have the option be: "If this optional section is included, then the use of the software to provide a service to others is only permitted if the terms of this license for copying, distribution, and modification, as given above, are followed." It seems foolish to protect the mechanism for sending the source code, rather than the right of the service users to get the source code through whatever mechanism is appropriate.
If you can dual-boot (which shouldn't be any harder than getting Windows to boot at all), you need one fewer computer. Or you might want $1000 of thermal engineering and design. How many vendors are selling Intel hardware in consumer-level packaging that puts the entire computer inside a flat panel display? Chances are that the iMac is also the quietest off-the-shelf Intel machine available now.
But even without open content, why doesn't the London Times license summaries from Brittanica to use as the background information in their articles? Or, at least, have a well-researched in-house archive of background info that they use whenever it's relevant?
Wikipedia is, in fact, licensed such that it would be completely fine on Wikipedia's end if a newspaper used a paragraph or two, provided they mentioned this fact.
It is a bit of a mystery why newspapers rarely cite sources for background information, and why this is supposed to be a good thing. The background information in an article is, by convention, a brief quotation from an expert who probably wasn't thinking about the topic before the reporter asked, or a reasonable reference work that's been paraphrased by a reporter who doesn't know anything about the topic, or whatever the reporter happens to think is right. Why wouldn't the reader prefer a summary directly from a reference work, cited in the byline? Certainly, for the portion of the article which is news, we want the reporter's words, because the reporter will presumably tell us something that no previously-existing text says, but I don't see a benefit to having the old information rewritten for the article.
If a gas station in your state doesn't accept passports or any form of ID your state provides, I bet they won't sell much booze, cigarettes, or porn. It's possible that gas stations in states that do the Real ID thing will stop accepting out-of-state ID (assuming they accept it now), but it's a bit unlikely that a business would cut off almost all of their customers and stay around.
When is the last time spyware was able to execute using Firefox with NoScript (whitelisted javascript), reading the EULA's [aka don't download 5,500 'free' games] and not installing kazaa [aka reem my computer please]?
Actually, writing open source software is one of the least expensive forms of entertainment for someone who needs to have a computer for other reasons already and has the necessary skills. Commercializing a piece of software requires a certain amount of money to be invested, and involves a certain amount of additional work. Open source doesn't have the potential monetary payoff, but it doesn't have the risk of monetary loss, either, and it's a better value proposition for someone with some code and no business or marketting experience. There's no way to reliably come out ahead if you have some random code, and your choices for breaking even are to just keep it or release it, and releasing it has some chance of generating interest from potential employers.
For in-house software, it is generally economically better, also. Consider Apache: it was wriiten by a set of sysadmins who were each paid by their employers to have a web server running with some extra features. None of them could have individually have made a product which would have been able to compete with the original NCSA reference version, so the wasn't a market for their work. As a group, they were able to make something worthwhile for a lot of people, but the copyright owners were too widely spread to form a company to make money on it. Open source allowed each participant to give away something worthless to them (the distribution rights to little features) in return for something valuable (the distribution rights to little features they'd otherwise have to implement themselves). Obviously, companies probably won't make any money by releasing open-source software, but they don't make any money on in-house software anyway (they make money with it, instead), they probably won't enable others to outcompete them with it, and they have a chance of eliciting useful improvements from others.
If you actually go to the real article instead of the slashdot-style blurb that slashdot linked to for some reason, you'll find that the study had people attempt to avoid a stopped car and drive "as fast as they were comfortable" through some cones. Surely drivers ought to be able to avoid traffic (especially if they're driving in California on the highway, where this seems to happen every day), and they should be able to find some safe speed to drive around cones. They did perfectly fine when driving their own cars. And chances are that they would have done fine with the 15-year old car 15 years ago (not as good results, but not messing up badly like they did now).
One minor issue with the first test is that they didn't know the road was wet, which means that they were misled about road conditions. Generally, being clueless like this indicates that the driver has already made a mistake, because you're supposed to watch out for that sort of thing, but it's a bit hard if somebody's sprayed down the road on a sunny day.
The real issue with the study is that, in order to pass, you'd have to drive in ways that are bad for driving the cars they ordinarily drive. If you've got ABS, you're supposed to not pump the brakes, because that just gives you less stopping power. If you've got traction control, you should turn tighter if you have to swerve. The point of these features is not that you can do stupid things and keep control of the vehicle; it's that you can do necessary things in more extreme situations and keep control of the vehicle. Nobody will be impressed if you crash into another car instead of swerving because you would have skidded in a 1990 BMW if you'd tried to avoid it.
When I drive a Subaru on snow, I have a tendancy to understeer on turns, because I expect to fishtail to complete the turn. This is kind of a bad habit, because I don't really drive anything else on snow these days, and it means I turn less gracefully than I could. (On the other hand, I could drive fine one day when there was thin snow over black ice. I only went out because I knew I could handle fishtailing, and the car actually skidded the way I expected; people who didn't learn to drive locally before traction control stayed home that day.)
Drivers haven't gotten worse; they've gotten accustomed to being able to drive in a way that is safer in modern cars.
Major worms tend to happen after full disclosure, because full disclosure allows people with no particular interest in exploiting a flaw to exploit it, and major worms are general of little or no value to their creators. This is not to say that any particular unreported flaw is being exploited, but if nefarious types know about a flaw, chances are that the result will be more effective targetted phishing, data theft, etc., not a major worm, and nobody will ever realize that the flaw had been exploited, because professionals are more likely to find a flaw (simply by time spent looking), and they're likely to use it for financial gain (either by reporting it, if they're working for a security company, or by exploiting it).
Chances are that they wouldn't want to put back doors in published source, particularly with all of the tracking of origins of patches in, at least, the Linux kernel this days.
It's not like the government will be the only people looking at the code, and the government generally doesn't want to publish clear documentation of domestic spying.
For that matter, the NSA is already a contributor to the Linux kernel, employs a maintainer (Stephen Smalley), and hosts a mailing list and web site on their module. But you can bet that a number of people review any changes they make.
I'll buy that commercial education-specific software is rarely, if ever, useful for education. (For that matter, any software which is not actually evaluated on its success at performing its stated function is likely to be bad, and educational software generally has this disconnect.)
On the other hand, there is software and there are sites which are actually very useful for learning. For example, I was recently curious as to whether a US president had ever been impeached when his party controlled Congress. Without a computer, I would have had a hard time determining whether the Johnson impeachment, which is generally reported as a separation-of-powers issue, also occurred with an opposing party holding a majority in the Senate (and, if so, whether it was a two-thirds majority). A quick check on Wikipedia reveals that the House was controlled by the Republicans (as of the 1866 election) and Johnson was a Democrat (although Lincoln was a Republican, and Johnson was his Vice-President; thus there is a party-affiliation link to the fact that Congress wanted to prevent Johnson from replacing Lincoln's cabinet appointees). A couple of links further gets the Senate's chronological list of Senators, which would tell me the exact membership of the Senate at the relevant times if it weren't 3 in the morning and I was willing to look through the couple of pages and add up parties. For that matter, it's worth noting that Congress had recently passed a law such that the successor to the presidency would be the (Republican, Lincoln-appointed) Secretary of State.
But the point is that computers can store and index a lot more information than can be conveniently managed in books, which means that students can do primary research for themselves, and investigate cause and effect for themselves, rather than having it reported to them by textbooks. Furthermore, they can essentially skip "knowledge", because the computer can answer for them more questions than any person could hope to know the answer to, and they can build the rest of Bloom's taxonomy on (easy) research skills instead of laboriously gained knowledge.
There's no replacement for a book in presenting a detailed argument on a particular topic, but a computer is far better for researching a topic than any single book or short list of books.
The weird thing isn't that they have those positions. The weird thing is that they have those titles. Why is the ITD Director also called the CIO, when the Commissioner of Revenue isn't called the CFO, and the Governor isn't called the CEO? The only thing I can think of is that they're trying to attract applicants from the business IT world who would like a CIO title. But this still seems like an odd desire, since the organizations are likely to have different sorts of demands on technology.
Actually, there's been a bit of work on a couple of Linux system calls, "splice" and "tee", which would allow programs to connect devices together without copying the data through a task's address space. With the appropriate driver support, you could literally splice a network connection to a video device, and the network card would start DMAing the data straight to the video card. Of course, if you need to do anything to the data in software in between (such as using a video codec that you don't have hardware support for), it needs to go to a program in between, and UNIX has been "streaming" in this sense for decades, using pipes (which also got some optimization in Linux at the same time).
Of course, to really do streaming video with the OS out of the way, you'd want to have microcontroller/FPGA/DSP coprocessors on your video and sound cards that could be loaded with the codec of the week, so you could actually get the OS out of the loop in real life situations.
In particular, it uses the OpenGL engine for everything, instead of only using it for 3D. And, actually, I don't think OS X or Vista is using OpenGL as the common layer; I thought they had these things rendering the windowing system with drivers for the particular video card, rather than having per-card drivers that implement an independant 3D standard, and rendering the windowing system with that standard. What isn't new is having the video drivers use the 3D functionality of the cards; what is new is using this functionality only through a standard interface. It doesn't matter for the quality of the eye-candy, but it does mean that there's less stuff happening in the proprietary drivers, and those drivers are providing a standardized API.
What I would like to see would be closed-source nVidia OpenGL drivers that run in userspace, using a generic API for access to the device. (For that matter, it would be great if they'd specify their card's bus behavior, so that the kernel could present nVidia cards to userspace as a character device that nobody knows how to use, but which will at least not screw up memory or lock the bus).
The discontinuity is only in the assignment of seconds of linear time (i.e., the number of seconds that have elapsed since some reference) to days and time of day. For purposes like GPS, where you don't care at all about making the time relevant to humans, the linear time is perfectly sufficient, and there aren't leap seconds or leap year days because those concepts are only relevant to the alignment of units larger than seconds. That site is a bit misleading, actually. GPS time doesn't use an hour:minute:second form. And the conversion of the GPS time values to human-readable clock values can include leap seconds, because the GPS signal includes a field for the number of accumulated leap seconds since GPS time started. So if you'd been watching the display on your GPS at midnight this new years, you probably would have seen it behave oddly as the leap second count field changed, most likely in the form of showing 12:59:59 for two seconds, because knowing to show 12:59:60 would require a bit of extra state tracking (you need to notice that the leap second count is different between two consecutive messages) and practically nobody is going to be paying attention and care.
If the car only steered you slightly in response to markings, it wouldn't be any worse than when they leave the bumpy reflectors in the old lane markings. The subjective experience would be like if there were slight speed bumps along the edges of lanes, so that you'd need to push against them slightly. It would just be a bit more force feedback.
Stanley would probably do fine at avoiding obstacles, but it wouldn't have any clue how other drivers may be expected to behave. Also, they'd need to extend its visual range; the cameras only see 80 meters, which gives it 7 seconds to react at 25 mph, but only 3 seconds at 55, which isn't really enough time to deal with some potential hazards, even if it reacts instantly. And, of course, surface roads will be difficult in other ways, because the car has to deal with a variety of regulations for driver behavior beyond not crashing into things, like stopping for pedestrians who aren't actually in your path and rules for right-of-way between vehicles.
But if it's just a talking head, it's only worth the same 1000 words, 900 times a minute. Somehow, I don't think I want to watch something with only 0.02% new content after the first 70 milliseconds.
I looked at AWT slightly, and Swing even less, and I think I'm going to wait until it is actually cool and the version is official before I look more closely. (I've mainly being using servlets, or, rather, a very little bit of the servlet API and a lot more of my own library, because the servlet API is mostly wrong, but it's barely good enough that you can write something good over it.)
The improvements in Java 5 are mostly old JSRs from long before C#; it's possible, however, that C# was responsible for getting Sun to stop endless debating improvements and actually include them.
It's actually just a method that takes as its last argument an array of Object, but it's more convenient to call now, because you can leave out the "new Object[] {... }" part. The only variability allowed is the length of the array. This enables features like a type-safe printf that's convenient to use.
The language changes in Java 5 are sufficiently significant that they eliminate most of the ways Java was awkward up to that point. Actually using the earlier versions involves a whole lot of annoying kludges which make it unnecessarily hard to learn and use. I think that Java is a better design overall, but they're similar enough that you may as well learn whichever has a more expressive version being taught at your school.
What I actually object to is the personification of whatever process is responsible for genetic change. Calling this God, with all of the semantic associations that word has, makes certain suggestions which are not clearly laid out.
For example, implies that religious people wishing to do the Will of God ought to use antibiotics to speed the spread of drug-resistant bacteria, because bacteria gaining drug resistance is clearly a genetic process. There is the assumption that all acts of creation must align with the religious concept of goodness, even though our understanding of disease is that some genetic changes are made to facilitate the slaughter of innocents.
Furthermore, there is the idea that the process running genetic variation has an understanding of human thought. There is, so far as I can tell, nothing to suggest that humanity is any more comprehensible to whatever creates life than the other way around.
Now, if you remove from ID the suggestions that a study of religion would reveal something about genetic change, it is probably a fine idea: clearly massive trial and error is a possible design process, and generate-and-test and genetic algorithms are studied as artificial intelligence. If ID advocates were willing to accept the possibility that the intelligent designer is an idiot savant with a special aptitude for protein folding and long chains of dependencies, but lacking the ability to use language, recognize emotion, or form unified plans, then ID could be a well-supported theory. But you shouldn't assume that whatever is responsible for the structure of the human eye could play a competant game of go.
Slashdot Mirror
I'm not so such that the particular DRM scheme that the GPL prohibits is one that kernel developers would want to support. It's prohibiting selling devices which have GPL firmware but require binaries signed with a private key that isn't included in the source, so that people can't actually install modified versions. The current draft seems to prohibit any systems which use signing and have a public key whose matching private key isn't included, but they've said that they want to fix this issue in later drafts. (So that, for example, GNU TLS could give RMS's public key as an example without giving his private key as well) Most likely, the result will be that a program which includes a public key without the corresponding private key must be modifiable to replace the public key with a different one.
The other anti-DRM measure is that it includes a denial of the magic statement in the DMCA, such that, in case anybody thought that you could stop somebody from defeating your GPLv3-licensed copy protection by suing them under the DMCA, they're wrong. Of course, a GPL-licensed copy protection scheme is going to be easy to defeat, anyway, since all versions of the license require that the user be able to modify the code to remove it, so it couldn't really work as a practical matter. Of course, some level of DRM is fine: the user of a program should be able to prevent other people from getting the data; sending encrypted documents and maintaining privacy is a fine use for GPL software. The point is that it is the person who runs the program who can choose whether or not to include each check, not some vendor or other entity.
So as a practical matter, the only situation in which DRM and the GPL could be used together was when a single system had a GPL portion and an immutable, vendor-chosen portion, and the vendor-chosen portion has the ability to inspect the GPL portion for changes. This isn't something that anyone who releases code under the GPL is likely to want to encourage, although there's a slim chance that people would choose the GPL over the BSD license out of curiousity, hoping to see the source to modified and distributed versions they can't actually run.
It wouldn't have to change the user's settings. It would only have to make the information available to people's phones. People could set their phone to have a special ring for when the user is at a show, much like special rings based on caller ID. And phones could come pre-programmed with a silent ring for this condition regardless of caller. So you don't change the phone settings; a single configuration would cause the phone to behave differently in response to where you happened to be.
That is, however, an optional part of the license, so anyone intending to release somethign as Free Software can simply choose not to use it. I believe I've seen that debian legal finally decided that it's okay for the GFDL to support invariant sections, so long as documents debian ships have no such sections. (But other issues with the GFDL are more serious and apply to all documents.) I'd guess that no project team that wants to keep from being kicked out would accept the addition of an invariant section to their project, given that it would require debian to fork their project.
Of course, if you consider the users of a web service program to be the people running the web service rather than the people using the service provided by those people, then it is fundamentally non-free to put any restriction on what the people running the service can do. (Like writing a compiler that can't be used to compile certain programs would be, or writing a word processor which required the produced documents to be put in the public domain.)
I'm a bit mystified as to why they didn't have the option be: "If this optional section is included, then the use of the software to provide a service to others is only permitted if the terms of this license for copying, distribution, and modification, as given above, are followed." It seems foolish to protect the mechanism for sending the source code, rather than the right of the service users to get the source code through whatever mechanism is appropriate.
If you can dual-boot (which shouldn't be any harder than getting Windows to boot at all), you need one fewer computer. Or you might want $1000 of thermal engineering and design. How many vendors are selling Intel hardware in consumer-level packaging that puts the entire computer inside a flat panel display? Chances are that the iMac is also the quietest off-the-shelf Intel machine available now.
But even without open content, why doesn't the London Times license summaries from Brittanica to use as the background information in their articles? Or, at least, have a well-researched in-house archive of background info that they use whenever it's relevant?
Wikipedia is, in fact, licensed such that it would be completely fine on Wikipedia's end if a newspaper used a paragraph or two, provided they mentioned this fact.
It is a bit of a mystery why newspapers rarely cite sources for background information, and why this is supposed to be a good thing. The background information in an article is, by convention, a brief quotation from an expert who probably wasn't thinking about the topic before the reporter asked, or a reasonable reference work that's been paraphrased by a reporter who doesn't know anything about the topic, or whatever the reporter happens to think is right. Why wouldn't the reader prefer a summary directly from a reference work, cited in the byline? Certainly, for the portion of the article which is news, we want the reporter's words, because the reporter will presumably tell us something that no previously-existing text says, but I don't see a benefit to having the old information rewritten for the article.
If a gas station in your state doesn't accept passports or any form of ID your state provides, I bet they won't sell much booze, cigarettes, or porn. It's possible that gas stations in states that do the Real ID thing will stop accepting out-of-state ID (assuming they accept it now), but it's a bit unlikely that a business would cut off almost all of their customers and stay around.
When is the last time spyware was able to execute using Firefox with NoScript (whitelisted javascript), reading the EULA's [aka don't download 5,500 'free' games] and not installing kazaa [aka reem my computer please]?
January 2nd?
Actually, writing open source software is one of the least expensive forms of entertainment for someone who needs to have a computer for other reasons already and has the necessary skills. Commercializing a piece of software requires a certain amount of money to be invested, and involves a certain amount of additional work. Open source doesn't have the potential monetary payoff, but it doesn't have the risk of monetary loss, either, and it's a better value proposition for someone with some code and no business or marketting experience. There's no way to reliably come out ahead if you have some random code, and your choices for breaking even are to just keep it or release it, and releasing it has some chance of generating interest from potential employers.
For in-house software, it is generally economically better, also. Consider Apache: it was wriiten by a set of sysadmins who were each paid by their employers to have a web server running with some extra features. None of them could have individually have made a product which would have been able to compete with the original NCSA reference version, so the wasn't a market for their work. As a group, they were able to make something worthwhile for a lot of people, but the copyright owners were too widely spread to form a company to make money on it. Open source allowed each participant to give away something worthless to them (the distribution rights to little features) in return for something valuable (the distribution rights to little features they'd otherwise have to implement themselves). Obviously, companies probably won't make any money by releasing open-source software, but they don't make any money on in-house software anyway (they make money with it, instead), they probably won't enable others to outcompete them with it, and they have a chance of eliciting useful improvements from others.
If you actually go to the real article instead of the slashdot-style blurb that slashdot linked to for some reason, you'll find that the study had people attempt to avoid a stopped car and drive "as fast as they were comfortable" through some cones. Surely drivers ought to be able to avoid traffic (especially if they're driving in California on the highway, where this seems to happen every day), and they should be able to find some safe speed to drive around cones. They did perfectly fine when driving their own cars. And chances are that they would have done fine with the 15-year old car 15 years ago (not as good results, but not messing up badly like they did now).
One minor issue with the first test is that they didn't know the road was wet, which means that they were misled about road conditions. Generally, being clueless like this indicates that the driver has already made a mistake, because you're supposed to watch out for that sort of thing, but it's a bit hard if somebody's sprayed down the road on a sunny day.
The real issue with the study is that, in order to pass, you'd have to drive in ways that are bad for driving the cars they ordinarily drive. If you've got ABS, you're supposed to not pump the brakes, because that just gives you less stopping power. If you've got traction control, you should turn tighter if you have to swerve. The point of these features is not that you can do stupid things and keep control of the vehicle; it's that you can do necessary things in more extreme situations and keep control of the vehicle. Nobody will be impressed if you crash into another car instead of swerving because you would have skidded in a 1990 BMW if you'd tried to avoid it.
When I drive a Subaru on snow, I have a tendancy to understeer on turns, because I expect to fishtail to complete the turn. This is kind of a bad habit, because I don't really drive anything else on snow these days, and it means I turn less gracefully than I could. (On the other hand, I could drive fine one day when there was thin snow over black ice. I only went out because I knew I could handle fishtailing, and the car actually skidded the way I expected; people who didn't learn to drive locally before traction control stayed home that day.)
Drivers haven't gotten worse; they've gotten accustomed to being able to drive in a way that is safer in modern cars.
Major worms tend to happen after full disclosure, because full disclosure allows people with no particular interest in exploiting a flaw to exploit it, and major worms are general of little or no value to their creators. This is not to say that any particular unreported flaw is being exploited, but if nefarious types know about a flaw, chances are that the result will be more effective targetted phishing, data theft, etc., not a major worm, and nobody will ever realize that the flaw had been exploited, because professionals are more likely to find a flaw (simply by time spent looking), and they're likely to use it for financial gain (either by reporting it, if they're working for a security company, or by exploiting it).
Chances are that they wouldn't want to put back doors in published source, particularly with all of the tracking of origins of patches in, at least, the Linux kernel this days.
It's not like the government will be the only people looking at the code, and the government generally doesn't want to publish clear documentation of domestic spying.
For that matter, the NSA is already a contributor to the Linux kernel, employs a maintainer (Stephen Smalley), and hosts a mailing list and web site on their module. But you can bet that a number of people review any changes they make.
I'll buy that commercial education-specific software is rarely, if ever, useful for education. (For that matter, any software which is not actually evaluated on its success at performing its stated function is likely to be bad, and educational software generally has this disconnect.)
On the other hand, there is software and there are sites which are actually very useful for learning. For example, I was recently curious as to whether a US president had ever been impeached when his party controlled Congress. Without a computer, I would have had a hard time determining whether the Johnson impeachment, which is generally reported as a separation-of-powers issue, also occurred with an opposing party holding a majority in the Senate (and, if so, whether it was a two-thirds majority). A quick check on Wikipedia reveals that the House was controlled by the Republicans (as of the 1866 election) and Johnson was a Democrat (although Lincoln was a Republican, and Johnson was his Vice-President; thus there is a party-affiliation link to the fact that Congress wanted to prevent Johnson from replacing Lincoln's cabinet appointees). A couple of links further gets the Senate's chronological list of Senators, which would tell me the exact membership of the Senate at the relevant times if it weren't 3 in the morning and I was willing to look through the couple of pages and add up parties. For that matter, it's worth noting that Congress had recently passed a law such that the successor to the presidency would be the (Republican, Lincoln-appointed) Secretary of State.
But the point is that computers can store and index a lot more information than can be conveniently managed in books, which means that students can do primary research for themselves, and investigate cause and effect for themselves, rather than having it reported to them by textbooks. Furthermore, they can essentially skip "knowledge", because the computer can answer for them more questions than any person could hope to know the answer to, and they can build the rest of Bloom's taxonomy on (easy) research skills instead of laboriously gained knowledge.
There's no replacement for a book in presenting a detailed argument on a particular topic, but a computer is far better for researching a topic than any single book or short list of books.
The weird thing isn't that they have those positions. The weird thing is that they have those titles. Why is the ITD Director also called the CIO, when the Commissioner of Revenue isn't called the CFO, and the Governor isn't called the CEO? The only thing I can think of is that they're trying to attract applicants from the business IT world who would like a CIO title. But this still seems like an odd desire, since the organizations are likely to have different sorts of demands on technology.
Actually, there's been a bit of work on a couple of Linux system calls, "splice" and "tee", which would allow programs to connect devices together without copying the data through a task's address space. With the appropriate driver support, you could literally splice a network connection to a video device, and the network card would start DMAing the data straight to the video card. Of course, if you need to do anything to the data in software in between (such as using a video codec that you don't have hardware support for), it needs to go to a program in between, and UNIX has been "streaming" in this sense for decades, using pipes (which also got some optimization in Linux at the same time).
Of course, to really do streaming video with the OS out of the way, you'd want to have microcontroller/FPGA/DSP coprocessors on your video and sound cards that could be loaded with the codec of the week, so you could actually get the OS out of the loop in real life situations.
In particular, it uses the OpenGL engine for everything, instead of only using it for 3D. And, actually, I don't think OS X or Vista is using OpenGL as the common layer; I thought they had these things rendering the windowing system with drivers for the particular video card, rather than having per-card drivers that implement an independant 3D standard, and rendering the windowing system with that standard. What isn't new is having the video drivers use the 3D functionality of the cards; what is new is using this functionality only through a standard interface. It doesn't matter for the quality of the eye-candy, but it does mean that there's less stuff happening in the proprietary drivers, and those drivers are providing a standardized API.
What I would like to see would be closed-source nVidia OpenGL drivers that run in userspace, using a generic API for access to the device. (For that matter, it would be great if they'd specify their card's bus behavior, so that the kernel could present nVidia cards to userspace as a character device that nobody knows how to use, but which will at least not screw up memory or lock the bus).
The discontinuity is only in the assignment of seconds of linear time (i.e., the number of seconds that have elapsed since some reference) to days and time of day. For purposes like GPS, where you don't care at all about making the time relevant to humans, the linear time is perfectly sufficient, and there aren't leap seconds or leap year days because those concepts are only relevant to the alignment of units larger than seconds. That site is a bit misleading, actually. GPS time doesn't use an hour:minute:second form. And the conversion of the GPS time values to human-readable clock values can include leap seconds, because the GPS signal includes a field for the number of accumulated leap seconds since GPS time started. So if you'd been watching the display on your GPS at midnight this new years, you probably would have seen it behave oddly as the leap second count field changed, most likely in the form of showing 12:59:59 for two seconds, because knowing to show 12:59:60 would require a bit of extra state tracking (you need to notice that the leap second count is different between two consecutive messages) and practically nobody is going to be paying attention and care.
If the car only steered you slightly in response to markings, it wouldn't be any worse than when they leave the bumpy reflectors in the old lane markings. The subjective experience would be like if there were slight speed bumps along the edges of lanes, so that you'd need to push against them slightly. It would just be a bit more force feedback.
Stanley would probably do fine at avoiding obstacles, but it wouldn't have any clue how other drivers may be expected to behave. Also, they'd need to extend its visual range; the cameras only see 80 meters, which gives it 7 seconds to react at 25 mph, but only 3 seconds at 55, which isn't really enough time to deal with some potential hazards, even if it reacts instantly. And, of course, surface roads will be difficult in other ways, because the car has to deal with a variety of regulations for driver behavior beyond not crashing into things, like stopping for pedestrians who aren't actually in your path and rules for right-of-way between vehicles.
But if it's just a talking head, it's only worth the same 1000 words, 900 times a minute. Somehow, I don't think I want to watch something with only 0.02% new content after the first 70 milliseconds.
2122 pts/3 S+ 0:03 gimp
Of course, the lines should be (2.6.13 af_unix.c, lines 292-293):
I looked at AWT slightly, and Swing even less, and I think I'm going to wait until it is actually cool and the version is official before I look more closely. (I've mainly being using servlets, or, rather, a very little bit of the servlet API and a lot more of my own library, because the servlet API is mostly wrong, but it's barely good enough that you can write something good over it.)
The improvements in Java 5 are mostly old JSRs from long before C#; it's possible, however, that C# was responsible for getting Sun to stop endless debating improvements and actually include them.
It's actually just a method that takes as its last argument an array of Object, but it's more convenient to call now, because you can leave out the "new Object[] { ... }" part. The only variability allowed is the length of the array. This enables features like a type-safe printf that's convenient to use.
The language changes in Java 5 are sufficiently significant that they eliminate most of the ways Java was awkward up to that point. Actually using the earlier versions involves a whole lot of annoying kludges which make it unnecessarily hard to learn and use. I think that Java is a better design overall, but they're similar enough that you may as well learn whichever has a more expressive version being taught at your school.
What I actually object to is the personification of whatever process is responsible for genetic change. Calling this God, with all of the semantic associations that word has, makes certain suggestions which are not clearly laid out.
For example, implies that religious people wishing to do the Will of God ought to use antibiotics to speed the spread of drug-resistant bacteria, because bacteria gaining drug resistance is clearly a genetic process. There is the assumption that all acts of creation must align with the religious concept of goodness, even though our understanding of disease is that some genetic changes are made to facilitate the slaughter of innocents.
Furthermore, there is the idea that the process running genetic variation has an understanding of human thought. There is, so far as I can tell, nothing to suggest that humanity is any more comprehensible to whatever creates life than the other way around.
Now, if you remove from ID the suggestions that a study of religion would reveal something about genetic change, it is probably a fine idea: clearly massive trial and error is a possible design process, and generate-and-test and genetic algorithms are studied as artificial intelligence. If ID advocates were willing to accept the possibility that the intelligent designer is an idiot savant with a special aptitude for protein folding and long chains of dependencies, but lacking the ability to use language, recognize emotion, or form unified plans, then ID could be a well-supported theory. But you shouldn't assume that whatever is responsible for the structure of the human eye could play a competant game of go.