In the US, the law is required to allow you to express your personal ideology, up to limits where doing so causes others harm. The Bill of Rights is one part of the law that she really should value as a citizen.
Plagiarism, of course, is a matter of degree. A quote or a few quotes is not plagiarism. Properly credited quotes can be supporting evidence for a position, although it's often better to write one's own explanation of something with a footnote (or endnote, internal reference, etc) to where the fact was found. It shows much better understanding, and makes a piece of prose flow better since it's written by one author and not assembled like Frankenstein's monster. Quotes should be left to actual speech, firsthand witness reporting, or examples reinforcing a point made in the author's own words as well.
One good way to mitigate the problems of term papers is to require fewer multiple choice and more essay answer tests throughout the term, so the students have a sample of their writing on record already.
A good way to measure student ability without a term paper at all is to have academic projects instead. Students learn much more about a topic from interviewing people on camera or on audio recorder than by searching the web or reading an encyclopedia entry. Interviewing scientific researchers makes science seem much more interesting than reading science-for-laypeople magazines. Interviewing military vets is a very educational experience, too. Science projects are a good idea. Civics projects, in which students for example write letters to government officials or help (maybe even start) nonprofit groups and document their unique experience, are very educational too. History projects for periods with no living survivors are a bit more difficult, as having a whole class approach a handful of historians at the local universities and museums could be overwhelming, especially if the same topic is required of all students. Projects in lieu of papers altogether may not be a solution, but they'd make a nice addition for the students and for those concerned about plagiarism.
Plagiarism is much like any other negative activity, or indeed any activity at all (leaving morality ad ethics out of it for now). If the rewards outweigh the risks, many people are going to choose the rewards. Rewards for plagiarizing a term paper are high: good grade, less time invested, and the paper often makes up an inordinate amount of the grade. The risks are pretty low if there's no time to evaluate someone's writing. Despite what many people think, it's very difficult to determine a person's writing style from his or her speaking style, too. Only a sample of earlier writings really makes it difficult to rip off another's writing. Sifting through millions or billions of other written works to find a match is much less likely to work than to simply find a mismatch between that student's earlier writing and the current project. Making the term paper less important in the overall grade structure creates a smaller incentive to cheat. Evaluating students in a balanced way across papers, projects, quizzes, and large tests make it much more difficult to game the system, and less rewarding to do so.
When I mod, I purposely look for topics interesting enough to read but about which I can resist commenting. That way, not only do i use all my mod points (most times), but I find I'm much more objective. When your first instinct is to rip into some idiot with a post, it's hard not to find a virtual "-1, pinhead" in the moderations list. Likewise, it's far too easy to up-mod a post with which I just want to agree.
I find that with this method, I sometimes mod a post up because the author really did make an insightful post with a good point even if I disagree with their conclusion. I also sometimes mod posts down when I agree with the conclusion but find the argument faulty or trolling. If I'm overwhelmingly drawn to post, I don't think I can be so impartial in my moderation.
In any case, caves are mostly formed by erosion and not tectonic activity. Water isn't necessary, but some sort of fluid is (notice I said fluid and not liquid -- air is a fluid).
The types of caves we normally think about on Earth are almost all formed by the flow of liquid water, dissolution of rock in water, and/or widening of cracks in rock by ice. If we find the types of caves on Mars that we have on Earth, there's a good chance they were formed by the presence of lots of liquid water at one time or another.
Does your Renault have more than 190,000 miles (305775.36 km) on it without major repair work? My wife's 1996 Dodge Intrepid does. My 1998 Pontiac Grand Am SE (2.4L DOHC, which got 35-38MPG on the highway when tuned a little leaner than factory settings and driven carefully) had over 180k miles on it (within five years) with only the alternator and battery replaced and a couple of tune-ups when some jackass rear-ended me in stop-and-go traffic near Chicago.
As for safety... My car was undrivable and the insurance company paid me off for value instead of repairing it. The larger car that did all the damage had a scratch or two and the guy drove it away from the accident. I was stranded for two hours waiting for a police officer and a car hauler while the other guy was able to drive away from the scene. I drive a larger car now, because it has a much better safety rating.
I turn off most or all of XP's eye candy, and I'm pretty sure you can do the same with Vista. On Linux or the BSDs, you can choose a window manager that doesn't do that at all or one that's highly configurable.
You make a good point about graphics power being used just for the sake of using it. Physics engines just like graphics engines could be put to good use, bad use, and silly use.
I was saying that the projects which state v2 or later can still be considered licensed as v2. The fact that some v2 projects are going to also be licensed as v3 doesn't stop them all from having been licensed as v2.
There has been some misunderstanding on/. that v2 or later means automatic revocation of v2 rights and automatic change to v3 exclusively. I was attempting to dispel that.
Of course you think you're right, but... ever heard of a distribution agreement? Ever heard of royalty arrangements? Just because distribution happens under contract doesn't mean just anyone can distribute copies of something.
As for used book stores, they do work under the first sale doctrine, because they are buying from after the first sale. Hence the word "used".
Someone distributing on behalf of the copyright holder or by arrangement with the copyright holder is simply not a buyer, and is not distributing based on some intrinsic right.
The distributor is not the end buyer. As I said, IANAL, but I don't believe first sale doctrine has ever meant that a wholesale distributor can, for instance, break street dates, make changes to a product, or change the license between the copyright holder and the first _retail_ buyer.
A distributor is not a buyer. The distributor is, by definition, distributing. Many distribution agreements actually work without the distributor ever owning a copy, even as an intermediary. The distributor just gets a cut of the sale, or distributes for a fee based on the number of copies. Regardless of how the transactions actually happen, the distributor is hardlyu ever considered the "buyer". You don't buy a car as used because the dealer lot owned it before you,`for example.
Hopefully to license GPU technology to Intel as an outside think tank, while making motherboard chipsets and high-speed PCI Express add-in cards for things that still aren't integrated onto the CPU. They have experience in making some pretty nice chipsets, after all, and more experience than most in making high-performance PCI Express peripherals.
PCI Express is 2.5Gbps per lane each way, so x16 means 40Gbps full duplex. I haven't seen any x32 anywhere, but there's supposed to be specs for it. That's 80Gbps full duplex on one interface. The companies that are still patting their own backs about the jump from ISA to PCI are not likely to be the leaders in putting peripherals other than video on x8, x16, and x32 connections. NVidia could be if it positions itself well.
Imagine a server with a 16x PCIe four port 10Gbps Ethernet card. (Sure, there's overhead, but the chances of sustaining the maximum on all four ports simultaneously for very long is small, and large FIFOs could make the issues pretty much moot. Make it three ports if that's such a concern.) As the processors scale, memory limits soar, and flash memory (maybe even holographic in a few years) replaces slower hard drives, machines will be able to satisfy more network requests as long as there is bandwidth into and out of the machine that can keep the data flowing.
Also, it'd make sense for NVidia to do physics, encryption, and other things which could use some acceleration outside the CPU.
Don't rule out a merger with the perpetual back runner, either. Via/S3/NVidia might just have enough know-how together to mount a cost/performance attack for low-end desktops and the entire mobile space.
I think you have the right idea in your bolded statement, but the wording could be slightly clearer. It's not so much making non-free software "with" free software as making non-free software "from" free software. The same distinction can be made between "using" and "containing".
There's nothing in the GPL that even hints that you can't compile a closed-source program using gcc. You entirely own the rights to whatever code you run gcc against. You can't make gcc itself closed source, though. That's a fine line of language, but a huge difference in concepts.
I know all this distinction sounds silly to some, but there are (or at least were) things in this world as crazy as compilers which carry a per-project or per-copy royalty on compiled programs or stipulations on how or where the resulting binaries could be used.
People who received the code under GPL v2 still have every right to copy, modify, and distribute v2 code as v2. The (or any later version) is clearly at the licensee's option.
I fail to see how granting license under a second license (v3) takes away rights granted under the first (v2). Retroactively enforcing something, if that was even a legal possibility in any reasonable jurisdiction, could not be done without revoking the rights already granted.
What part of a "distributor" undertaking actions to "distribute" copies of a work does not sound like "distribution" to you?
Yes, distribution is in fact one of the rights held only by the copyright holder and those licensed by the copyright holder. In fact, distribution is one of the rights afforded exclusively to the copyright holder and licensees under US copyright law and under the WIPO and WCT treaties. So if you don't follow this license, you can't _distribute_.
The entry for the terms in the Table of Contents for the GPL v2 is called "TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION".
Here's the fourth paragraph of Preamble of the GPL v2, and notice it doesn't say "if you are the one to make the copies you distribute" anywhere:
For example, if you distribute copies of such a program, whether gratis or for a fee, you must give the recipients all the rights that you have. You must make sure that they, too, receive or can get the source code. And you must show them these terms so they know their rights.
Notice that it doesn't say you have to have modified it to be bound by the license.
Here's paragraph 5 of the license proper (emphasis mine):
5. You are not required to accept this License, since you have not signed it. However, nothing else grants you permission to modify or distribute the Program or its derivative works. These actions are prohibited by law if you do not accept this License. Therefore, by modifying or distributing the Program (or any work based on the Program), you indicate your acceptance of this License to do so, and all its terms and conditions for copying, distributing or modifying the Program or works based on it.
The only reason the license repeatedly says "copy and distribute" is that it is granting both rights. It is not because the two are separable and you must agree to the license only if you do both.
X/Open claims to own the Unix certification mark, and they also claim to own the rights to the definition of standards and interfaces. If they publish a paper with the IP they got from Novell that SCO didn't, and someone uses that to make a clone, what foot is SCO standing on in the first place?
or the manufacturer or the mechanic. Generalizations being what they are, blaming the driver is a good place to start. However, brakes do sometimes fail even with proper maintenance. I've seen the pedal cover fall off the pedal on a car still under warranty, and if that happens at the wrong time (the driver's foot goes to the floor with the cover usually) it could cause an accident. Anti-lock brakes have been known to lock and cause an increased stopping distance. Defective tires have been known to blow out under normal inflation pressures, and cars with one flat front tire are difficult to stop or steer in a controllable fashion.
Yes, the car being the problem is the exception, but it can happen. Likewise, if you happen to have an accounting package running on certain older Pentium chips, or your PC comes with non-ECC RAM for some reason and is being used for important data... I think you can see the point here. It's an exception, and probably exceedingly rare. Rare doesn't mean nonexistent, though.
A gun that jams and backfires despite proper maintenance might really kill a person, even. However, the gun should never get the blame when a person fires it and the gun doesn't malfunction. That's the holder's fault, or a parent's fault if they let their kid play with the gun.
Of course, in all these situations that the object seems to carry part of the blame independent of the end user, that blame really rests on the manufacturer or maintainer.
but it's not because you like to support the artists instead of the middlemen. I go to show, pay the cover (unless a buddy of mine is the promoter in which case I still might pay if I like the band), and then buy the CD from the band at the show.
I am a deviant, but I still maintain the two are unrelated.;-) I think most people would prefer the money go to the band.
I don't think even IBM is claiming that SCO turned over all rights to Xenix... er, SCO Unix, er, I mean Unixware. IBM's just saying that any rights to any property SCO had in Linux is part of the UnitedLinux effort, since Caldera/SCO/Caldera/SCO/Canopy/whomever they want to be this week signed off on distributing everything that was in Linux through that.
Not all datacenters are wired the same, of course. Some have step-downs to 48v DC, and wires are screwed to terminals on the equipment. Some have 220V or 440v outlets mounted on the rack, and UPS plugged into those, with servers and such actually plugged into the UPSes. These are not wall sockets, but are only different in that they are not on the walls. Of course, anyone who lets a cleaner with no clearance into the data center is an idiot. Cleaning server rooms is what interns are for.
I used to work for an ISP which used the UPS approach. We'd have two separate 3000VA UPSes plugged into two separate circuits, and have the redundant power on each server go into two (or three, if it was a three-PS server) into different UPSes. Servers which duplicated a service for load balancing got their power sources spread out to different source circuits from one another. It was all still three-prong, 115v@60Hz (US) outlets, but within the racks themselves.
The janitor was never in the server room, and the air conditioner maintenance guys/telephone company guys were never in their unaccompanied. That goes a long way towards accountability for cords not being moved.
OK, so "vast majority" in terms of raw bit count - of course you're right. "Vast majority" in terms of pictures? Absolutely http(s)
Don't be so quick to jump to that conclusion -- the alt.binaries groups are huge, including alt.binaries.pictures.erotica etc. The sorting ability and automatic download can more than make up for differences in picture count.
The fact that I can download images of DVDs, TV shows, music, or anything else is irrelevant because the majority of views on the intarweb are through a browser, and filtering that would accomplish my goals, even if bittorrent and nntp are huge bandwidth hogs.
It's not irrelevant at all. If your goal is to minimize access, you can't simply shift access to another service. It may help quite a bit with accidentally stumbling over porn to block it from HTTP only, but it doesn't stop a kid who's looking for it, or who clicks on a link that loads a non-HTTP URL.
FWIW, I'm not going to let my kids have access to unrestricted nntp feeds, either. There's lots of crap out there.
That's a good thing to know. We're talking about things besides HTTP now.
If your goal is to filter the Interweb, PICS labels were designed for this purpose, Great. How many sites use that?
And while I'm comfortable with my implementation and management of squid/dansguardian plus custom filters, most parents could not possibly figure out how that works. I work in IT at a Fortune 100 company, and most of my coworkers couldn't do it. (Outside of the small percentage of really geeky ones.)
NetNanny is install and go. Cybersitter is install and go. K9 web protection I believe is install and go (and free for home use). ICRAplus is free and really, really easy.
I get that people get ticked when some piddly group like a state government tries to define technology standards for the whole world, but porn and violent content are a major problem and should be filtered from people who need the filters.
It's not defining. It's redefining. Redefining a system that's already in place and used around the world. That's an entirely different thing which requires an entirely different amount of work.
Current technology is pretty crappy when it comes to filtering, and it makes sense to me that passing some laws about how to characterize the content is an attempt to address the real issue.
Current filtering technology is really, really good when it's used.
Attempting to address an issue that's important to your constituents is a good thing. Botching that attempt badly because you ignore real issues with your proposed solution is another.
Requiring that adult content has a flag in the image format makes as much sense as all this -- more in fact. New image formats could have an explicit porn flag, and older ones could be required by the state of Utah to carry a this-is-porn watermark or this-is-porn ID string steganographically inserted into the image stream.
An HTTP header such as "Content-age-rating: 13+" or "Age-rating: 18+" or something makes more sense than changing the ports, too. This sort of thing could even be done by proxying the outside web and inserting the header as needed at the ISP. It's an onerous mandate on an ISP, but at least it'd only be for the jurisdiction in question.
The idea that all adult content gets moved off of port 80 onto some other port suggests someone gets to enforce that all web servers on the Internet are no longer sending "adult" material over port 80. Who enforces this in Belgium? In Australia? In the Netherlands? In South Africa?
It makes more sense, if you really want to do this by port, to start with a clean port from the beginning. Let port 80 be, as that's where the smut you're worried about is, and require that all traffic on port -- or, say, 4321 or so -- is clean from the start. Or that it's labelled from the start. As a tech-savvy person, you personally could writ
Oddly enough, there's a name for a random collection of people doing their own self-interested thing that creates value. It's called capitalism. Ballmer doesn't understand this because in capitalism there's this thing called competition, and that's a dirty word in his little world.
Slashdot Mirror
In the US, the law is required to allow you to express your personal ideology, up to limits where doing so causes others harm. The Bill of Rights is one part of the law that she really should value as a citizen.
Plagiarism, of course, is a matter of degree. A quote or a few quotes is not plagiarism. Properly credited quotes can be supporting evidence for a position, although it's often better to write one's own explanation of something with a footnote (or endnote, internal reference, etc) to where the fact was found. It shows much better understanding, and makes a piece of prose flow better since it's written by one author and not assembled like Frankenstein's monster. Quotes should be left to actual speech, firsthand witness reporting, or examples reinforcing a point made in the author's own words as well.
One good way to mitigate the problems of term papers is to require fewer multiple choice and more essay answer tests throughout the term, so the students have a sample of their writing on record already.
A good way to measure student ability without a term paper at all is to have academic projects instead. Students learn much more about a topic from interviewing people on camera or on audio recorder than by searching the web or reading an encyclopedia entry. Interviewing scientific researchers makes science seem much more interesting than reading science-for-laypeople magazines. Interviewing military vets is a very educational experience, too. Science projects are a good idea. Civics projects, in which students for example write letters to government officials or help (maybe even start) nonprofit groups and document their unique experience, are very educational too. History projects for periods with no living survivors are a bit more difficult, as having a whole class approach a handful of historians at the local universities and museums could be overwhelming, especially if the same topic is required of all students. Projects in lieu of papers altogether may not be a solution, but they'd make a nice addition for the students and for those concerned about plagiarism.
Plagiarism is much like any other negative activity, or indeed any activity at all (leaving morality ad ethics out of it for now). If the rewards outweigh the risks, many people are going to choose the rewards. Rewards for plagiarizing a term paper are high: good grade, less time invested, and the paper often makes up an inordinate amount of the grade. The risks are pretty low if there's no time to evaluate someone's writing. Despite what many people think, it's very difficult to determine a person's writing style from his or her speaking style, too. Only a sample of earlier writings really makes it difficult to rip off another's writing. Sifting through millions or billions of other written works to find a match is much less likely to work than to simply find a mismatch between that student's earlier writing and the current project. Making the term paper less important in the overall grade structure creates a smaller incentive to cheat. Evaluating students in a balanced way across papers, projects, quizzes, and large tests make it much more difficult to game the system, and less rewarding to do so.
When I mod, I purposely look for topics interesting enough to read but about which I can resist commenting. That way, not only do i use all my mod points (most times), but I find I'm much more objective. When your first instinct is to rip into some idiot with a post, it's hard not to find a virtual "-1, pinhead" in the moderations list. Likewise, it's far too easy to up-mod a post with which I just want to agree.
I find that with this method, I sometimes mod a post up because the author really did make an insightful post with a good point even if I disagree with their conclusion. I also sometimes mod posts down when I agree with the conclusion but find the argument faulty or trolling. If I'm overwhelmingly drawn to post, I don't think I can be so impartial in my moderation.
In any case, caves are mostly formed by erosion and not tectonic activity. Water isn't necessary, but some sort of fluid is (notice I said fluid and not liquid -- air is a fluid).
The types of caves we normally think about on Earth are almost all formed by the flow of liquid water, dissolution of rock in water, and/or widening of cracks in rock by ice. If we find the types of caves on Mars that we have on Earth, there's a good chance they were formed by the presence of lots of liquid water at one time or another.
Does your Renault have more than 190,000 miles (305775.36 km) on it without major repair work? My wife's 1996 Dodge Intrepid does. My 1998 Pontiac Grand Am SE (2.4L DOHC, which got 35-38MPG on the highway when tuned a little leaner than factory settings and driven carefully) had over 180k miles on it (within five years) with only the alternator and battery replaced and a couple of tune-ups when some jackass rear-ended me in stop-and-go traffic near Chicago.
As for safety... My car was undrivable and the insurance company paid me off for value instead of repairing it. The larger car that did all the damage had a scratch or two and the guy drove it away from the accident. I was stranded for two hours waiting for a police officer and a car hauler while the other guy was able to drive away from the scene. I drive a larger car now, because it has a much better safety rating.
Yay! No more pedestrian lawsuits after my front bumper finishes them off!
I turn off most or all of XP's eye candy, and I'm pretty sure you can do the same with Vista. On Linux or the BSDs, you can choose a window manager that doesn't do that at all or one that's highly configurable.
You make a good point about graphics power being used just for the sake of using it. Physics engines just like graphics engines could be put to good use, bad use, and silly use.
I wasn't arguing what you apparently think I was.
/. that v2 or later means automatic revocation of v2 rights and automatic change to v3 exclusively. I was attempting to dispel that.
I was saying that the projects which state v2 or later can still be considered licensed as v2. The fact that some v2 projects are going to also be licensed as v3 doesn't stop them all from having been licensed as v2.
There has been some misunderstanding on
Of course you think you're right, but... ever heard of a distribution agreement? Ever heard of royalty arrangements? Just because distribution happens under contract doesn't mean just anyone can distribute copies of something.
As for used book stores, they do work under the first sale doctrine, because they are buying from after the first sale. Hence the word "used".
Someone distributing on behalf of the copyright holder or by arrangement with the copyright holder is simply not a buyer, and is not distributing based on some intrinsic right.
The distributor is not the end buyer. As I said, IANAL, but I don't believe first sale doctrine has ever meant that a wholesale distributor can, for instance, break street dates, make changes to a product, or change the license between the copyright holder and the first _retail_ buyer.
A distributor is not a buyer. The distributor is, by definition, distributing. Many distribution agreements actually work without the distributor ever owning a copy, even as an intermediary. The distributor just gets a cut of the sale, or distributes for a fee based on the number of copies. Regardless of how the transactions actually happen, the distributor is hardlyu ever considered the "buyer". You don't buy a car as used because the dealer lot owned it before you,`for example.
Hopefully to license GPU technology to Intel as an outside think tank, while making motherboard chipsets and high-speed PCI Express add-in cards for things that still aren't integrated onto the CPU. They have experience in making some pretty nice chipsets, after all, and more experience than most in making high-performance PCI Express peripherals.
PCI Express is 2.5Gbps per lane each way, so x16 means 40Gbps full duplex. I haven't seen any x32 anywhere, but there's supposed to be specs for it. That's 80Gbps full duplex on one interface. The companies that are still patting their own backs about the jump from ISA to PCI are not likely to be the leaders in putting peripherals other than video on x8, x16, and x32 connections. NVidia could be if it positions itself well.
Imagine a server with a 16x PCIe four port 10Gbps Ethernet card. (Sure, there's overhead, but the chances of sustaining the maximum on all four ports simultaneously for very long is small, and large FIFOs could make the issues pretty much moot. Make it three ports if that's such a concern.) As the processors scale, memory limits soar, and flash memory (maybe even holographic in a few years) replaces slower hard drives, machines will be able to satisfy more network requests as long as there is bandwidth into and out of the machine that can keep the data flowing.
Also, it'd make sense for NVidia to do physics, encryption, and other things which could use some acceleration outside the CPU.
Don't rule out a merger with the perpetual back runner, either. Via/S3/NVidia might just have enough know-how together to mount a cost/performance attack for low-end desktops and the entire mobile space.
I think you have the right idea in your bolded statement, but the wording could be slightly clearer. It's not so much making non-free software "with" free software as making non-free software "from" free software. The same distinction can be made between "using" and "containing".
There's nothing in the GPL that even hints that you can't compile a closed-source program using gcc. You entirely own the rights to whatever code you run gcc against. You can't make gcc itself closed source, though. That's a fine line of language, but a huge difference in concepts.
I know all this distinction sounds silly to some, but there are (or at least were) things in this world as crazy as compilers which carry a per-project or per-copy royalty on compiled programs or stipulations on how or where the resulting binaries could be used.
People who received the code under GPL v2 still have every right to copy, modify, and distribute v2 code as v2. The (or any later version) is clearly at the licensee's option.
I fail to see how granting license under a second license (v3) takes away rights granted under the first (v2). Retroactively enforcing something, if that was even a legal possibility in any reasonable jurisdiction, could not be done without revoking the rights already granted.
Yes, distribution is in fact one of the rights held only by the copyright holder and those licensed by the copyright holder. In fact, distribution is one of the rights afforded exclusively to the copyright holder and licensees under US copyright law and under the WIPO and WCT treaties. So if you don't follow this license, you can't _distribute_.
Bitlaw page about copyright
US Copyright Office
Wikipedia page on copyright
Findlaw's copyright page
Wikipedia WIPO page
Dutch copyright law page on Wikipedia (in English)
Japanese copyright law chapter II (note section 3, subsection 3) (translated to English, obviously)
The entry for the terms in the Table of Contents for the GPL v2 is called "TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION".
Here's the fourth paragraph of Preamble of the GPL v2, and notice it doesn't say "if you are the one to make the copies you distribute" anywhere:
Notice that it doesn't say you have to have modified it to be bound by the license.
Here's paragraph 5 of the license proper (emphasis mine):
The only reason the license repeatedly says "copy and distribute" is that it is granting both rights. It is not because the two are separable and you must agree to the license only if you do both.
They are only within the GPL if they are actually distributing the source to people who request it. Are they doing that?
Probably because the people bring the votes, but the big businesses bring the money.
X/Open claims to own the Unix certification mark, and they also claim to own the rights to the definition of standards and interfaces. If they publish a paper with the IP they got from Novell that SCO didn't, and someone uses that to make a clone, what foot is SCO standing on in the first place?
or the manufacturer or the mechanic. Generalizations being what they are, blaming the driver is a good place to start. However, brakes do sometimes fail even with proper maintenance. I've seen the pedal cover fall off the pedal on a car still under warranty, and if that happens at the wrong time (the driver's foot goes to the floor with the cover usually) it could cause an accident. Anti-lock brakes have been known to lock and cause an increased stopping distance. Defective tires have been known to blow out under normal inflation pressures, and cars with one flat front tire are difficult to stop or steer in a controllable fashion.
Yes, the car being the problem is the exception, but it can happen. Likewise, if you happen to have an accounting package running on certain older Pentium chips, or your PC comes with non-ECC RAM for some reason and is being used for important data... I think you can see the point here. It's an exception, and probably exceedingly rare. Rare doesn't mean nonexistent, though.
A gun that jams and backfires despite proper maintenance might really kill a person, even. However, the gun should never get the blame when a person fires it and the gun doesn't malfunction. That's the holder's fault, or a parent's fault if they let their kid play with the gun.
Of course, in all these situations that the object seems to carry part of the blame independent of the end user, that blame really rests on the manufacturer or maintainer.
but it's not because you like to support the artists instead of the middlemen. I go to show, pay the cover (unless a buddy of mine is the promoter in which case I still might pay if I like the band), and then buy the CD from the band at the show.
;-) I think most people would prefer the money go to the band.
I am a deviant, but I still maintain the two are unrelated.
I don't think even IBM is claiming that SCO turned over all rights to Xenix... er, SCO Unix, er, I mean Unixware. IBM's just saying that any rights to any property SCO had in Linux is part of the UnitedLinux effort, since Caldera/SCO/Caldera/SCO/Canopy/whomever they want to be this week signed off on distributing everything that was in Linux through that.
Not all datacenters are wired the same, of course. Some have step-downs to 48v DC, and wires are screwed to terminals on the equipment. Some have 220V or 440v outlets mounted on the rack, and UPS plugged into those, with servers and such actually plugged into the UPSes. These are not wall sockets, but are only different in that they are not on the walls. Of course, anyone who lets a cleaner with no clearance into the data center is an idiot. Cleaning server rooms is what interns are for.
I used to work for an ISP which used the UPS approach. We'd have two separate 3000VA UPSes plugged into two separate circuits, and have the redundant power on each server go into two (or three, if it was a three-PS server) into different UPSes. Servers which duplicated a service for load balancing got their power sources spread out to different source circuits from one another. It was all still three-prong, 115v@60Hz (US) outlets, but within the racks themselves.
The janitor was never in the server room, and the air conditioner maintenance guys/telephone company guys were never in their unaccompanied. That goes a long way towards accountability for cords not being moved.
OK, so "vast majority" in terms of raw bit count - of course you're right. "Vast majority" in terms of pictures? Absolutely http(s)
Don't be so quick to jump to that conclusion -- the alt.binaries groups are huge, including alt.binaries.pictures.erotica etc. The sorting ability and automatic download can more than make up for differences in picture count.
The fact that I can download images of DVDs, TV shows, music, or anything else is irrelevant because the majority of views on the intarweb are through a browser, and filtering that would accomplish my goals, even if bittorrent and nntp are huge bandwidth hogs.
It's not irrelevant at all. If your goal is to minimize access, you can't simply shift access to another service. It may help quite a bit with accidentally stumbling over porn to block it from HTTP only, but it doesn't stop a kid who's looking for it, or who clicks on a link that loads a non-HTTP URL.
FWIW, I'm not going to let my kids have access to unrestricted nntp feeds, either. There's lots of crap out there.
That's a good thing to know. We're talking about things besides HTTP now.
If your goal is to filter the Interweb, PICS labels were designed for this purpose,
Great. How many sites use that?
And while I'm comfortable with my implementation and management of squid/dansguardian plus custom filters, most parents could not possibly figure out how that works. I work in IT at a Fortune 100 company, and most of my coworkers couldn't do it. (Outside of the small percentage of really geeky ones.)
NetNanny is install and go. Cybersitter is install and go. K9 web protection I believe is install and go (and free for home use). ICRAplus is free and really, really easy.
I get that people get ticked when some piddly group like a state government tries to define technology standards for the whole world, but porn and violent content are a major problem and should be filtered from people who need the filters.
It's not defining. It's redefining. Redefining a system that's already in place and used around the world. That's an entirely different thing which requires an entirely different amount of work.
Current technology is pretty crappy when it comes to filtering, and it makes sense to me that passing some laws about how to characterize the content is an attempt to address the real issue.
Current filtering technology is really, really good when it's used.
Attempting to address an issue that's important to your constituents is a good thing. Botching that attempt badly because you ignore real issues with your proposed solution is another.
Requiring that adult content has a flag in the image format makes as much sense as all this -- more in fact. New image formats could have an explicit porn flag, and older ones could be required by the state of Utah to carry a this-is-porn watermark or this-is-porn ID string steganographically inserted into the image stream.
An HTTP header such as "Content-age-rating: 13+" or "Age-rating: 18+" or something makes more sense than changing the ports, too. This sort of thing could even be done by proxying the outside web and inserting the header as needed at the ISP. It's an onerous mandate on an ISP, but at least it'd only be for the jurisdiction in question.
The idea that all adult content gets moved off of port 80 onto some other port suggests someone gets to enforce that all web servers on the Internet are no longer sending "adult" material over port 80. Who enforces this in Belgium? In Australia? In the Netherlands? In South Africa?
It makes more sense, if you really want to do this by port, to start with a clean port from the beginning. Let port 80 be, as that's where the smut you're worried about is, and require that all traffic on port -- or, say, 4321 or so -- is clean from the start. Or that it's labelled from the start. As a tech-savvy person, you personally could writ
Oddly enough, there's a name for a random collection of people doing their own self-interested thing that creates value. It's called capitalism. Ballmer doesn't understand this because in capitalism there's this thing called competition, and that's a dirty word in his little world.
I don't think Ballmer could be emotionally detached from yesterday's breakfast. The man rants, yells, or cries abuot nearly anything.
Build a set for the next CSI? They have so much equipment in local labs on those shows, you'd think spectrographs were always $300.
I'm going to assume you just skimmed and responded based on that. The poster did, indeed, mention "as much or more" in that sentence.