Nobody seems to have yet mentioned those little 2" blowtorches that attach to a keyring. I'm sure one of those would come in handy to someone with a subversive streak.
Maybe they won't help her in a class. Maybe they'll just help her have fun and gain the kind of out-of-the-classroom experience that college is important for.
I own a hammer, which I keep in my room at school. I bought it so I could nail some Teflon skid pads onto the legs of my chair, but it turned out that it already had metal pads so I couldn't use the hammer for that purpose. To date, the only thing I've ever used this hammer for (in the year and a half since I bought it) was to go outside with some friends and smash up an old broken motherboard. It served no useful purpose whatsoever but it was a novel way to have fun with friends.
If I owned a set of lockpicks, I'd probably use it to open the wiring box that contains my building's Ethernet switch, and take a look around. This isn't harmful (it's not like I'd go pulling wires out) but I'm sure it'd lead to some sort of creative fun.
And if an "LED flashlight" is one of those little keychain things with a single button battery and LED, I love mine. I can carry it in my pocket and it doesn't get in the way and it's really useful in lots of random situations.
College is about more than classes. Classes are no more than 50% of the picture, really. (For some people, drinking makes up the other 50%, but I don't drink at all so I must say that that's not true in all cases.)
No, it isn't just what they do with it. A patent is supposed to indicate that someone has uniquely invented something. From what people are saying here, neither of these patents seems to be anything new - the technologies are already widely used.
If the technologies are already widely used, that's prior art, and nobody should be able to patent them. Not Red Hat. Not Microsoft. Not me.
The reason they feel they need to do this is, presumably, to defend themselves. If they get the patents, nobody else can get them and use them as weapons. If the USPO finds they can't be granted because they're not novel, then (presumably) they can't be granted to anyone else either and therefore still can't be used as a weapon. This makes sense.
But it's still wrong, from a fundamental, moral point of view. No matter how it's used, nobody should be able to get a patent for something they didn't invent.
Software patents are bad because they facilitate "patent attacks" - companies that get a patent on some ludicrous thing (like pop-under ads) and then use it to sue other companies for infringement.
If Red Hat is applying for these patents in order to defend themselves against such an attack, (as another poster put it, pretty much the only way to prove prior art these days is to get the patent yourself), that's fine. What's troubling is the state of the political/business world in which they feel they have to do it.
One low-quality, distorted monochrome picture derived from nine seconds of audio doesn't exactly constitute "video". And "derived from" is the proper term - do some reading about what a spectrograph is and you'll understand how this picture arises.
(Attempt to) boot your computer from the CD, thus damaging it.
Get together and file a very large class-action lawsuit against the publisher or the people who made the protection mechanism, for selling a harmful product without adequate warning about its danger.
Actually it's the other way around: VHDL is better, and schematic entry is for wimps.
The reason is that what you draw in the graphic editor is not what actually gets put in the FPGA. I don't know much in detail about how FPGA's work internally, but basically, the compiler can look at a VHDL description and produce the most efficient gate-array implementation for it. Given a schematic design, it doesn't have a high-level sense of what the logic is supposed to do, so it's harder to produce an optimal FPGA implementation.
Your schematic design would be more efficient if it were implemented as you actually drew it, but not on an FPGA.
HTTP proxies can be very frustrating if they're not done "right".
At my high school, (this was several years ago), administration decided to put in an NT-based HTTP proxy, which required users to authenticate to it so it could track what sites they visited. Unfortunately, this meant it used NTLM authentication, so the Debian system which a friend and I (co-webmasters for the school) were running in the library. To update our packages, we had to manually download.deb packages and FTP them over to the Debian system.
Of course, this is the same school district that bought two quad-P2-Xeon servers for the sole purpose of running 30-odd copies of IE on 30-odd student workstations via Citrix...
As another poster in this thread pointed out, the speed limit is a law, not a contract. But even beyond this, there's a reasonable expectation that every licensed driver is aware that roads have speed limits, which the law says you must obey so there's no question of consent. So if you're driving on a road, it can be assumed that you know there's some speed which you're not allowed to exceed, and if you don't look at the sign to find out what that speed is, it's your own fault for being careless.
Unless you were shown the EULA before you paid for the software, the same can't be said of a software puchase.
IIS really isn't easier to configure - it's easier to not configure and just go with the defaults. My Apache logs still regularly show Nimda and even Code Red hits, because people just throw IIS on a box and put it on the 'net and then never mess with it again. When you install Apache, you have to actually open up httpd.conf and look through it, and this keeps you from pretending that everything is already configured appropriately for your use of the server. And most Linux distributions have packge systems - if a major security hole were to be found in Apache, I could just run "apt-get update; apt-get install apache" (or, more likely, "apt-get dist-upgrade") and get the fix. No having to surf to Windows Update and mess with checkboxes and then reboot when I'm done.
I know there are GUI interfaces for configuring Apache, but I've never used any of them because, really, httpd.conf isn't bad. You can read it top to bottom and know everything your server will and won't do. It's not like it's not documented - the default version of the file has explanatory comments for every option. And if you don't know what something means, you open the docs and look up the option in the list of directives, and you notice a bunch of other interesting and useful-looking directives at the same time, so you read about those too. You end up learning, rather than just wanting to get it over with.
If we look at real societies, we see that nowhere does the entire population actually become honest. The reason it happens in the simulation is that with a bunch of people acting honest for fear of being arrested, everyone else is forced to be honest too, because of the greater likelihood of being reported. But in a real society, a dishonest person who's acting honest to avoid arrest probably wouldn't report another dishonest person to the police. I have a feeling the system wouldn't "tip" if it weren't for this fact.
I wish I had the link... but I was doing some reading of my own about color spaces a few weeks ago, and one article I read said that the CIELAB color space was intended to make it possible to measure perceptual differences between colors, but was found to be inadequate for the purpose. But there's currently no other color space that's better for the purpose.
Yes, the market is dominated by Microsoft, largely because most people are intimidated by things like kernel configuration so they stick with the OS where they feel at least somewhat safe.
Most people, especially at the "average Joe" level, don't understand the inner workings of their OS. Heck, most of them fear their OS and assume that they'll break something if they tinker with the OS's inner settings.
And if they're running Windows, they're probably right, too. Let's see if we can dispel that fear by making something that's more approachable.
No, it's actually just about right. The limit used to be 2GB within 24 hours, but they split it to 1GB in 12 hours to help alleviate the afternoon peaks - you can download just as much, but you have to spread it out a bit. If you want to download two CDs, you can start one when you go to sleep at night, and start the second when you finish with classes the next day.
When one of my roommates moved out around the middle of last semester, I put in a second ethernet card and ran a second cable to the port where his computer had been plugged in, so I could switch over to it if I went over my limit with one card, but it turned out that I only did that once or twice. I do transfer enough data that I go over the limit from time to time, but with some forethought I can generally avoid it, and when it happens it's not a big annoyance because I know what I did to cause it.
I don't know what software is actually used to enforce the limits, unfortunately - I'm working on finding that out. The totals are checked every hour, and that determines who's in the penalty box for the following hour. I've heard another student suggest that it might be based on the software used by broadband providers to limit their users' bandwidth, tuned to provide something similar to a "brickwall filter" past a certain point. Or perhaps (my own speculation) the main Internet gateway forwards the packets to an alternate gateway, like a modem bank. I've heard secondhand that the penalty box is actually one or two 56k modems shared by all the people who are in it, and I could believe that - it's slow enough that I have a hard time maintaining a connection to AIM.
At my school (Lehigh University), we address bandwidth problems this way...
Each student (each MAC address, really) is allowed to transfer one gigabyte within a 12-hour period. If you go over the limit, you get put in a "penalty box" (basically sub-58k speeds) for a while until your transfer total for the past 12 hours is under a gig. Uploads and downloads are counted separately, and transfers that don't go off-campus don't count at all. One of the university's servers holds a list of what addresses are in the penalty box, and what their transfer totals are.
This is quite effective - it gives each student a reasonable amount of bandwidth, and it only punishes those who actually use too much of it. And our 45mbit internet connection is rarely maxed out.
People do generally like to be paid, but not everyone who writes a piece of software goes and sells it for profit - that's why we have free software. I'm sure there are people who would be willing to give their good ideas to the public for everyone's benefit.
Head on over to the Philips CD Logo page and download the zip file provided there. Inside is a PDF which gives the official rules for using the Compact Disc logo. The relevant text is on page 8:
This logo may only be used on discs complying with the CD-DA specification: IEC 60908 and/or the Philips-Sony Compact Disc Digital Audio System Description (also known as the RED Book).
Unfortunately you have to pay Philips to get the actual Red Book specification, but if this protection system works by putting "bad" data in the subchannels of the disc, I'd say it's reasonable to assume that this violates the standard. So if you see the "Compact Disc Digital Audio" logo anywhere on the disc or its cover, you can complain to your record store either that the advertising is misleading, or that the disc claims to be compliant but isn't so it must be defective.:-)
The problem with that is that AFAIK there's no legal protection for ideas like there is for actual creative works. Neither copyright nor patent applies to an idea. But if some way could be found to prevent companies from taking the free ideas and turning them into proprietary software, open-source brainstorming would be great.
These researchers have some very interesting ideas that could very well be the "next generation" (or a few generations down the line) in computing. The system they describe has a few problems that come to mind (mainly related to security), but the overall idea is sound.
However, I fail to see why a new operating system is needed for this, when a software layer would do just fine. Maybe they want to break backward compatibility. Maybe it's just Microsoft's compulsion to integrate everything with the OS. But this distributed system could be realized just fine as a layer on top of existing operating systems.
Reading this article makes me think of the Lisp programming language. The authors talk about creating a new level of abstraction where programmers concern themselves with the data they're working with instead of having to worry about the number of bits in an int or the amount of space they've allocated for a string. That's exactly what Lisp does. Now, this "distributed system" goes a bit beyond that and actually abstracts the entire Internet into a single computing environment, but that's very similar to what CORBA does.
Higher levels of abstraction have always built on lower ones; basic functions in a C library are written in assembly, and a Lisp interpreter uses the facilities of the language it's written in (likely C or something comparable) to create a new environment for the programmer. All we really need for a "distributed system" to exist is a new language and an interpreter that transparently does networking and data replication and all. Not a whole new OS.
Here's another angle. Given the diversity of hardware that exists, an OS that links every type of computer in the world would have to provide a whole new set of drivers for every piece of hardware, and those drivers and the interfaces to them would pretty much constitute an OS in themselves. Why not just leverage existing code?
I can imagine a Linux system that boots up in the usual way except that instead of starting gettys, it starts GnuMilennium and bingo, you're connected. Meanwhile all the other usual services run as separate processes. Meanwhile a Windows2010 system boots up and starts MSmilennium instead of Explorer. (Now, maybe when Microsoft talks about a "new OS" they really just mean a new frontend.) In both cases, you have a new software system running on the foundation of an existing kernel and set of services. No new OS.
If the open-source/academic community wants to get a head-start on Microsoft with this, a good place to start might be by thinking about how to combine the concepts behind Lisp with the concepts behind CORBA. And I don't mean writing a CORBA binding for Lisp - those already exist. I mean combining the data-centrism of Lisp with the network-transparency of CORBA. That would give you pretty much everything these Microsoft researchers are proposing.
Overall, I like these ideas. If such a system comes along, as long as it isn't proprietary and the security issues can be worked out, I'd love to give it a try.
You seem to be saying that because Zimmermann could forsee, when he wrote and released PGP, that it could be used for criminal purposes, he can be blamed for last Tuesday's attacks. In my mind this is very similar to the continuing debate about software like DeCSS which has legitimate uses as well as potentially illegal ones. I'm curious what your stance on those other issues is.
Many people in the Slashdot community support DeCSS and oppose the DMCA on the grounds that while it's obviously illegal to pirate copyrighted material, it makes no sense to hold programmers responsible for the actions of their users. I agree with this view, and by similar reasoning, I don't hold Phil responsible for the actions of people who use his software. Nor do I hold gun manufacturers responsible for every crime that gets committed with the weapons they produce. It's a fact of life that some people will break our social rules and commit crimes, petty or atrocious, and while it's certainly respectable to do one's part to make it harder for them, if we deny ourselves basic rights like copyright fair-use and cryptographic privacy to protect ourselves from criminals, we're living in fear and that's just not right.
The perfect solution to the world's problems hasn't been found yet, and might never be, but sticking our heads in the sand and outlawing anything that might possibly be used criminally definitely isn't it. Certainly, things would be much simpler if we just made the world a police state and took away all rights. But at that point we might as well get rid of all the people and replace everyone with robots, because without respect for people's emotions and desires, that's all we'd be. Civilization is based on people's ability to be creative and communicate freely, and we celebrate our human ability to do things like write poetry and music, study the world we live in and learn about it, invent automobiles and airplanes, and, I may add, write software. Any of these (well, maybe not poetry) can be used to do harm, but I, for one, am not willing to start giving up my humanity just because the world isn't always a nice place.
This is something I've wondered about too... how does it work, legally, to have the EULA inside the software box so it can't be read until you've purchased the product?
Stores generally don't take returns on opened software because otherwise people would buy the product, copy the CD, and then return it for a refund. But if you don't accept the terms of the EULA, what options are you left with? You can't get your money back but you can't use the product you bought. It seems like having the license inside the box is a way of pressuring the customer into accepting it; maybe it should be required to have a copy of it somewhere that it can be read before the purchase is made? (Even though most people don't read them anyway, I know.)
Last I heard (and this was from an old book called "two-minute mysteries"), mice actually don't eat cheese all that much - it has a tendency to overheat their blood or something and kill them. At least, that was the solution to the mystery (the mice were dying because the lab guy was feeding them cheese). I know this is offtopic but I'd be interested in knowing if this is actually true.
Slashdot Mirror
Nobody seems to have yet mentioned those little 2" blowtorches that attach to a keyring. I'm sure one of those would come in handy to someone with a subversive streak.
Maybe they won't help her in a class. Maybe they'll just help her have fun and gain the kind of out-of-the-classroom experience that college is important for.
I own a hammer, which I keep in my room at school. I bought it so I could nail some Teflon skid pads onto the legs of my chair, but it turned out that it already had metal pads so I couldn't use the hammer for that purpose. To date, the only thing I've ever used this hammer for (in the year and a half since I bought it) was to go outside with some friends and smash up an old broken motherboard. It served no useful purpose whatsoever but it was a novel way to have fun with friends.
If I owned a set of lockpicks, I'd probably use it to open the wiring box that contains my building's Ethernet switch, and take a look around. This isn't harmful (it's not like I'd go pulling wires out) but I'm sure it'd lead to some sort of creative fun.
And if an "LED flashlight" is one of those little keychain things with a single button battery and LED, I love mine. I can carry it in my pocket and it doesn't get in the way and it's really useful in lots of random situations.
College is about more than classes. Classes are no more than 50% of the picture, really. (For some people, drinking makes up the other 50%, but I don't drink at all so I must say that that's not true in all cases.)
Not at all. If you're a musician, the DMCA does not restrict in any way your right to perform, record, and distribute your own music.
yEnc isn't all that great. See http://www.exit109.com/~jeremy/news/yenc.html.
No, it isn't just what they do with it. A patent is supposed to indicate that someone has uniquely invented something. From what people are saying here, neither of these patents seems to be anything new - the technologies are already widely used.
If the technologies are already widely used, that's prior art, and nobody should be able to patent them. Not Red Hat. Not Microsoft. Not me.
The reason they feel they need to do this is, presumably, to defend themselves. If they get the patents, nobody else can get them and use them as weapons. If the USPO finds they can't be granted because they're not novel, then (presumably) they can't be granted to anyone else either and therefore still can't be used as a weapon. This makes sense.
But it's still wrong, from a fundamental, moral point of view. No matter how it's used, nobody should be able to get a patent for something they didn't invent.
Software patents are bad because they facilitate "patent attacks" - companies that get a patent on some ludicrous thing (like pop-under ads) and then use it to sue other companies for infringement.
If Red Hat is applying for these patents in order to defend themselves against such an attack, (as another poster put it, pretty much the only way to prove prior art these days is to get the patent yourself), that's fine. What's troubling is the state of the political/business world in which they feel they have to do it.
One low-quality, distorted monochrome picture derived from nine seconds of audio doesn't exactly constitute "video". And "derived from" is the proper term - do some reading about what a spectrograph is and you'll understand how this picture arises.
OK, everyone who owns an iMac:
Actually it's the other way around: VHDL is better, and schematic entry is for wimps.
The reason is that what you draw in the graphic editor is not what actually gets put in the FPGA. I don't know much in detail about how FPGA's work internally, but basically, the compiler can look at a VHDL description and produce the most efficient gate-array implementation for it. Given a schematic design, it doesn't have a high-level sense of what the logic is supposed to do, so it's harder to produce an optimal FPGA implementation.
Your schematic design would be more efficient if it were implemented as you actually drew it, but not on an FPGA.
HTTP proxies can be very frustrating if they're not done "right".
At my high school, (this was several years ago), administration decided to put in an NT-based HTTP proxy, which required users to authenticate to it so it could track what sites they visited. Unfortunately, this meant it used NTLM authentication, so the Debian system which a friend and I (co-webmasters for the school) were running in the library. To update our packages, we had to manually download .deb packages and FTP them over to the Debian system.
Of course, this is the same school district that bought two quad-P2-Xeon servers for the sole purpose of running 30-odd copies of IE on 30-odd student workstations via Citrix...
Actually, I saw a segment about Opera on CNN the other day. I was surprised, and pleased.
As another poster in this thread pointed out, the speed limit is a law, not a contract. But even beyond this, there's a reasonable expectation that every licensed driver is aware that roads have speed limits, which the law says you must obey so there's no question of consent. So if you're driving on a road, it can be assumed that you know there's some speed which you're not allowed to exceed, and if you don't look at the sign to find out what that speed is, it's your own fault for being careless.
Unless you were shown the EULA before you paid for the software, the same can't be said of a software puchase.
IIS really isn't easier to configure - it's easier to not configure and just go with the defaults. My Apache logs still regularly show Nimda and even Code Red hits, because people just throw IIS on a box and put it on the 'net and then never mess with it again. When you install Apache, you have to actually open up httpd.conf and look through it, and this keeps you from pretending that everything is already configured appropriately for your use of the server. And most Linux distributions have packge systems - if a major security hole were to be found in Apache, I could just run "apt-get update; apt-get install apache" (or, more likely, "apt-get dist-upgrade") and get the fix. No having to surf to Windows Update and mess with checkboxes and then reboot when I'm done.
I know there are GUI interfaces for configuring Apache, but I've never used any of them because, really, httpd.conf isn't bad. You can read it top to bottom and know everything your server will and won't do. It's not like it's not documented - the default version of the file has explanatory comments for every option. And if you don't know what something means, you open the docs and look up the option in the list of directives, and you notice a bunch of other interesting and useful-looking directives at the same time, so you read about those too. You end up learning, rather than just wanting to get it over with.
If we look at real societies, we see that nowhere does the entire population actually become honest. The reason it happens in the simulation is that with a bunch of people acting honest for fear of being arrested, everyone else is forced to be honest too, because of the greater likelihood of being reported. But in a real society, a dishonest person who's acting honest to avoid arrest probably wouldn't report another dishonest person to the police. I have a feeling the system wouldn't "tip" if it weren't for this fact.
I wish I had the link... but I was doing some reading of my own about color spaces a few weeks ago, and one article I read said that the CIELAB color space was intended to make it possible to measure perceptual differences between colors, but was found to be inadequate for the purpose. But there's currently no other color space that's better for the purpose.
Yes, the market is dominated by Microsoft, largely because most people are intimidated by things like kernel configuration so they stick with the OS where they feel at least somewhat safe.
And if they're running Windows, they're probably right, too. Let's see if we can dispel that fear by making something that's more approachable.
No, it's actually just about right. The limit used to be 2GB within 24 hours, but they split it to 1GB in 12 hours to help alleviate the afternoon peaks - you can download just as much, but you have to spread it out a bit. If you want to download two CDs, you can start one when you go to sleep at night, and start the second when you finish with classes the next day.
When one of my roommates moved out around the middle of last semester, I put in a second ethernet card and ran a second cable to the port where his computer had been plugged in, so I could switch over to it if I went over my limit with one card, but it turned out that I only did that once or twice. I do transfer enough data that I go over the limit from time to time, but with some forethought I can generally avoid it, and when it happens it's not a big annoyance because I know what I did to cause it.
I don't know what software is actually used to enforce the limits, unfortunately - I'm working on finding that out. The totals are checked every hour, and that determines who's in the penalty box for the following hour. I've heard another student suggest that it might be based on the software used by broadband providers to limit their users' bandwidth, tuned to provide something similar to a "brickwall filter" past a certain point. Or perhaps (my own speculation) the main Internet gateway forwards the packets to an alternate gateway, like a modem bank. I've heard secondhand that the penalty box is actually one or two 56k modems shared by all the people who are in it, and I could believe that - it's slow enough that I have a hard time maintaining a connection to AIM.
At my school (Lehigh University), we address bandwidth problems this way...
Each student (each MAC address, really) is allowed to transfer one gigabyte within a 12-hour period. If you go over the limit, you get put in a "penalty box" (basically sub-58k speeds) for a while until your transfer total for the past 12 hours is under a gig. Uploads and downloads are counted separately, and transfers that don't go off-campus don't count at all. One of the university's servers holds a list of what addresses are in the penalty box, and what their transfer totals are.
This is quite effective - it gives each student a reasonable amount of bandwidth, and it only punishes those who actually use too much of it. And our 45mbit internet connection is rarely maxed out.
People do generally like to be paid, but not everyone who writes a piece of software goes and sells it for profit - that's why we have free software. I'm sure there are people who would be willing to give their good ideas to the public for everyone's benefit.
Head on over to the Philips CD Logo page and download the zip file provided there. Inside is a PDF which gives the official rules for using the Compact Disc logo. The relevant text is on page 8:
Unfortunately you have to pay Philips to get the actual Red Book specification, but if this protection system works by putting "bad" data in the subchannels of the disc, I'd say it's reasonable to assume that this violates the standard. So if you see the "Compact Disc Digital Audio" logo anywhere on the disc or its cover, you can complain to your record store either that the advertising is misleading, or that the disc claims to be compliant but isn't so it must be defective. :-)
The problem with that is that AFAIK there's no legal protection for ideas like there is for actual creative works. Neither copyright nor patent applies to an idea. But if some way could be found to prevent companies from taking the free ideas and turning them into proprietary software, open-source brainstorming would be great.
IANAL.
These researchers have some very interesting ideas that could very well be the "next generation" (or a few generations down the line) in computing. The system they describe has a few problems that come to mind (mainly related to security), but the overall idea is sound.
However, I fail to see why a new operating system is needed for this, when a software layer would do just fine. Maybe they want to break backward compatibility. Maybe it's just Microsoft's compulsion to integrate everything with the OS. But this distributed system could be realized just fine as a layer on top of existing operating systems.
Reading this article makes me think of the Lisp programming language. The authors talk about creating a new level of abstraction where programmers concern themselves with the data they're working with instead of having to worry about the number of bits in an int or the amount of space they've allocated for a string. That's exactly what Lisp does. Now, this "distributed system" goes a bit beyond that and actually abstracts the entire Internet into a single computing environment, but that's very similar to what CORBA does.
Higher levels of abstraction have always built on lower ones; basic functions in a C library are written in assembly, and a Lisp interpreter uses the facilities of the language it's written in (likely C or something comparable) to create a new environment for the programmer. All we really need for a "distributed system" to exist is a new language and an interpreter that transparently does networking and data replication and all. Not a whole new OS.
Here's another angle. Given the diversity of hardware that exists, an OS that links every type of computer in the world would have to provide a whole new set of drivers for every piece of hardware, and those drivers and the interfaces to them would pretty much constitute an OS in themselves. Why not just leverage existing code?
I can imagine a Linux system that boots up in the usual way except that instead of starting gettys, it starts GnuMilennium and bingo, you're connected. Meanwhile all the other usual services run as separate processes. Meanwhile a Windows2010 system boots up and starts MSmilennium instead of Explorer. (Now, maybe when Microsoft talks about a "new OS" they really just mean a new frontend.) In both cases, you have a new software system running on the foundation of an existing kernel and set of services. No new OS.
If the open-source/academic community wants to get a head-start on Microsoft with this, a good place to start might be by thinking about how to combine the concepts behind Lisp with the concepts behind CORBA. And I don't mean writing a CORBA binding for Lisp - those already exist. I mean combining the data-centrism of Lisp with the network-transparency of CORBA. That would give you pretty much everything these Microsoft researchers are proposing.
Overall, I like these ideas. If such a system comes along, as long as it isn't proprietary and the security issues can be worked out, I'd love to give it a try.
You seem to be saying that because Zimmermann could forsee, when he wrote and released PGP, that it could be used for criminal purposes, he can be blamed for last Tuesday's attacks. In my mind this is very similar to the continuing debate about software like DeCSS which has legitimate uses as well as potentially illegal ones. I'm curious what your stance on those other issues is.
Many people in the Slashdot community support DeCSS and oppose the DMCA on the grounds that while it's obviously illegal to pirate copyrighted material, it makes no sense to hold programmers responsible for the actions of their users. I agree with this view, and by similar reasoning, I don't hold Phil responsible for the actions of people who use his software. Nor do I hold gun manufacturers responsible for every crime that gets committed with the weapons they produce. It's a fact of life that some people will break our social rules and commit crimes, petty or atrocious, and while it's certainly respectable to do one's part to make it harder for them, if we deny ourselves basic rights like copyright fair-use and cryptographic privacy to protect ourselves from criminals, we're living in fear and that's just not right.
The perfect solution to the world's problems hasn't been found yet, and might never be, but sticking our heads in the sand and outlawing anything that might possibly be used criminally definitely isn't it. Certainly, things would be much simpler if we just made the world a police state and took away all rights. But at that point we might as well get rid of all the people and replace everyone with robots, because without respect for people's emotions and desires, that's all we'd be. Civilization is based on people's ability to be creative and communicate freely, and we celebrate our human ability to do things like write poetry and music, study the world we live in and learn about it, invent automobiles and airplanes, and, I may add, write software. Any of these (well, maybe not poetry) can be used to do harm, but I, for one, am not willing to start giving up my humanity just because the world isn't always a nice place.
This is something I've wondered about too... how does it work, legally, to have the EULA inside the software box so it can't be read until you've purchased the product?
Stores generally don't take returns on opened software because otherwise people would buy the product, copy the CD, and then return it for a refund. But if you don't accept the terms of the EULA, what options are you left with? You can't get your money back but you can't use the product you bought. It seems like having the license inside the box is a way of pressuring the customer into accepting it; maybe it should be required to have a copy of it somewhere that it can be read before the purchase is made? (Even though most people don't read them anyway, I know.)
Last I heard (and this was from an old book called "two-minute mysteries"), mice actually don't eat cheese all that much - it has a tendency to overheat their blood or something and kill them. At least, that was the solution to the mystery (the mice were dying because the lab guy was feeding them cheese). I know this is offtopic but I'd be interested in knowing if this is actually true.