Just have an arm that can extend either a urinal or bedpan. Any self respecting couchbot should be able to dump them out in the toliet and rinse them without mishap. Heck attach a little sprayer nozzle to it so you won't have to mess with TP either. I bet the Japanese would buy this crap up in a heartbeat. (The average japanese is even more gadget crazy than even the most raving Slashdotter. They even try to have more advanced toliets than their neighbors.)
True. I want KDE for my Debian desktops as well. At least when it finally does arrive it will have several months worth of bugfixes and polish. It's a pity that their release freeze came with all of this good stuff just down the road....Gnome 2, Mozilla 1.0, KDE 3.0.x......
Since Sid (unstable) is the gateway to Testing (Woody about to supplant Potato) it can't even be had from there. The real pisser is that once Woody releases and this all good stuff shows up in Sid it will be wise to wait for it to come to testing(next codename?). The very first truly new bits of software to go in are apt (heh heh!) to be flaky. Oh well, in this context its a GOOD thing that official Debian releases are infrequent.
Microsoft might write into its licenses "you may not run open source software on our platforms".
By all means pile on restrictions as to how people are allowed to use your products. Its yet another argument in favor classifing Microsoft products as "legacy". Since they've maxed out the desktop market we can expect to see Microsoft act more and more like the Scientologists.
The smart thing for them to do is to diversify in such a way that they are not abusing their desktop monopoly. If they play their cards right, they could get out of trouble with the government....permanently.....Shrub won't be around to shield them forever. And they could still continue to grow their revenues albeit at a less incredible rate. I don't expect them to be smart. I expect them to do things like bully school districts and like the Scientologists again fire howitzers at their feet.
Their ankle deep in the water now. If they start drowning, let's throw 'em an anvil. Nasty political campaign style ads with them admitting their software is unsafe at any speed to get out of antitrust charges would be a good start.
I don't have anything to say at the moment about the larger issues being debated in this thread but I do have something to say about random number sources. If I wanted to fill a CD with good random numbers/dev/urandom is not how I would go about it. The quality of/dev/urandom is reasonable as uses bits of fluff like the delay between keypresses and chatter from the device drivers to create an "entropy pool" to seed a pseudorandom algorithm with. The problem is that it is slooooowwww. Most goings on in a normal desktop PC are very very ordered and deterministic. The few that aren't represent a very small amount of entropy.
All of this means that the process that is generating your iso is going to see short bursts of data inbetween long periods of entropy gathering. That CD will probably take hours at least to generate. Also I said the quality of the data is "reasonable". If one means to keep the government or a well heeled corporate attacker out of the cyphertext it may not be good enough. Even the non-deterministic processes in a PC likely have a fair amount of order in them. In other words, that entropy pool is probably good enough to make a 2048 bit assymetric key. It probably wouldn't do for a 650MB iso. The longer the string of numbers, the more likely hidden order can be found.
The way I would is to sample the output of a white noise generator. The output of the ADC is then used to seed a good pseudorandom algorithm. The reason why we use the white noise as a seed is to obliterate any bias in the data caused by such factors as the slew rate, bandpass of the analog circuitry making the white noise or any subtle imperfection that may exist in the ADC. A reverse biased transistor is one source of analog noise. This would be a high speed generator of quality random numbers. The speed would only be limited by the clock rate of the ADC or rate at which the PC can process the output.
The FSF does not AFAIK own any copyrights on the Linux kernel itself. Just because something is GPLed doesn't mean that RMS has Godlike powers to dictate terms over it. The FSF is protective of the GNU tools which they do own the copyrights on and they can indeed haul people into court over those. Making something GPL doesn't make it a part of the GNU project.
The Polish government in exile was based in Britain and their surviving pilots flew with the RAF and the Army Air Corps. Most of these guys were tall, blond-haired, blue-eyed, dashing and had "cute" accents. Needless to say, they got all of the girls when they happened to be around. The piqued American personnel started telling Polack Jokes........
The problem with co-operative multitasking is that if an app hangs and doesn't hand control back to the kernel or the next app in line then the whole system freezes. Windows 3.x was badly affected by it to an obnoxious degree. And it could bite Win9x systems running 16-bit code. And its still a problem with MacOS 10.x. The woes of co-operative multitasking were my second biggest bitch with the old style MacOSen (lack of protected memory was the first). I'll gladly trade some speed for the ability to kill offending apps.
Now it may be that those Acorns had a higher code quality in general to mitigate that problem but it can still be a problem. For that matter, higher priority processes can be reniced on most pre-emptive systems to make the processes that matter snappy.
There is only one way a co-operative system can be tolerable. If the number of apps is restricted to a limited set that is highly polished and bug free then it might be okay. But forget just chucking any ole program into that environment unless you like looking at hourglasses, busy bees and ticking watches.
The developer can also tell the client that the more they want to own the code the more it is going to cost them. A completely closed app that a client owns ALL rights to means that the developer has to reinvent all of the little wheels and gears in it. The developer cannot even use his own private stock of functions much less anything GPLed.
Ironically, a developer coding such an app could use LGPLed libraries as long as he didn't have to patch the libraries himself (absolute ownership again). For that matter, the developer can also use the BSD and X stuff as long as all copyrights are acknowledged. The personal toolbox however would be completely out unless LGPLed or using one of the acknowledge copyright licenses.
It also means that the developer cannot add anything to his personal toolbox that comes out of an absolute ownership job. The developer is also in peril if he codes anything else that looks like something he did on an absolute ownership job.
That might be the best angle: "I can only give you a non-exclusive license to certain parts of the apps as I cannot put myself out of the market from having worked for you." This is a purely business argument and may go over better. Don't say "open source" at all. Just negotiate space for you to retain ownership of your generic functions and license them according to your inclinations be them Open Source or Proprietary.
We use Apples extensively in our elementaries. We replace them on a five year cycle. For the intended applications, the Apples last far longer than the Wintel boxes in the MS and HS. The way we use them, the price evens out. I'll also point out that in general, the quality of the Apples in a bit higher than el-cheapo Wintel boxes. The amount of hardware failure is much lower than some 233Mhz Gateways I used to service. It's that total cost of ownership thing again. For that matter, keeping the software in a sane state is easier as well. That's not to say that they don't crash from time to time but the recovery tends to be far less difficult than when a Windows machine gets it's registry and filesystem borked up.
Don't let the upfront cost of the Macs fool you. For elementary school classroom purposes, they mostly Just Work. Oh yeah, becoming an Apple Self-Servicing District wasn't very costly either and the technical support is EXCELLENT.
A quality Wintel box running 2000 pretty has these virtues as well...at least they're stable. They cost about as much too. We don't deploy eMachines or PowerSpecs here and with few exceptions there are no Macintosh equivalents to those.
The WORST offenders for mp3s where I work are the teachers. You're generous. We gave them a week before whacking the mp3s from the their network shares. We didn't bother with the warning when we found some 200 Dancing Baby AVIs......WHACK!!
It's really cute the way the p2p thing broke down. The High School and the Middle School share the same T1 line. The Middle Schoolers loved Gnutella. One teacher had something like 4 GB of MP3s in his share (quotas have since been put on the network storage). At the High School, Kazaa was King with AudioGalaxy running second. I'd sit there watching the network monitor when lunchtime rolled around. The P2P ports just absolutely spiked through the ceiling....greedy....greedy.
I'm fortunate enough to have an Administration with some clue. We unceremoniously blocked the ports and had an intercom announcement. There's already an AUP but it will be more heavily emphasized next year. It will also be made clear that the technological measures are only there to keep em honest. We don't intend to have an arms race. If someone gets busted then they're busted As I said, the Administration is with us on this one.
I just tell people: "Gnutella's cool but we don't use it at school. Do it at home or at your buddy's house. I like it too but I don't do it here."
My ass. The first thing I would do after an opensource migration is to make it known to former vendors that since "The Software" is not installed that all such agreements are null and void. If they persisted in an audit (there wouldn't be anything to find) then they can expect to be sued for wasting time and resources, barratry, persecution and anything else a good lawyer can think of.
Everyone will just start slamming their brakes at such intersections REGARDLESS of whether it is called for. If you get rear-ended it is almost always the other guys fault regardless of the situation anyway. Will they blame the increase in rearenders on their lights......no. I'll just make a bumper sticker that says "I brake HARD for camera lights." Fair warning for anybody who has to drive behind me.
Some infrared spectrometers do indeed use salt lenses to focus the light. Instruments intended to sample the air have to use dessicant filters because humid air will degrade the lenses.
What's needed is something like patch and diff for audio/video. If one had to follow a precise set of instructions to make their own Phantom Edit then indeed no one would watch it. However, you could stick your Phantom Menance DVD into your machine and uses a viewer that accepts a file of instructions to create a Phantom Edit in real time. It could be more than instructions as well. Replacement sound clips, video overlays, and video clips could be injected as well. As long as the "patches" do not themselves contain trademarked or copyrighted content then nothing legal could be done about this. One would have to own a copy of the original work to use it.
Perhaps boycott is the wrong word here. For instance, I am not boycotting Metallica. Whenever I happen to hear a Metallica tune; I feel disgust at yet another carriage maker bemoaning the automobile. I suppose that if I were a better person I could get past the behaivor of that greedy pighead Lars Ulrich but in this regard I'm sorry to admit I'm small and petty. I don't meet the ideal in this regard; so sue me....oh wait...someone probably will.
If an artist is loud and obnoxious enough in their public life then it WILL change the light in which their art is seen. Sometimes the art can survive the opproprium an artist brings on himself but sometimes not. Lars Ulrich could put out what could have been the best music I have ever heard in my life. He will never ever have to worry about me using PAN to download it from USENET but then he won't have to worry about my buying it either. I'm disgusted with the little prick. Why would I want to leech anything of his?
Microsoft has 21 fairly broad patents that cover among other things, integrating DRM into an OS. The worry is that an SSSCA type law would basically make a Microsoft OS or Microsoft licenced technology the only legal way to use computers in this country.
True. But presumably, the Democrats would have learned their lesson and would not field an Eisner pawn against him when his term comes up. They say that politics makes strange bedfellows for a reason. If the CB??? or SSSCA or whatever they're calling it this week really really offends you then you are going to have to hold your nose and help out the other guy. At this point, I'd support an Alfred E. Neumann candidacy if it would oust Hollings, Feinstein or other scum like that.
Another poster pointed out that the NRA doesn't really vote Republican, they vote Gun. The thing we really have to be wary of is forming alliances with other causes that are ultimately detrimental. Banding with Creationists for one term to get rid of someone like Hollings is an example of the type of thing we need to be prepared to do. There is no need to band with them for all time under any circumstances.
We have many many values that are important to us. Politics will demand that we sometimes accept setbacks in one area to gain victory in another. I don't like it either but that's reality folks.
I once had to retire a Mac LC II was the building fileserver. This thing had financials, the private records of students; you name it. I low-leveled the drive and wrote 0's to it. Once that was done, I drilled several holes through the platters. I broke the bit off the drill in the process. The drive with drill bit stub stuck in it looks like Count Datatula with a spike through his heart. We keep the spiked carcass around to show people how to make sure that sensitive data gets destroyed.
I don't believe that is what usually happens if some GPLed stuff winds up in a larger work. There are other options:
1. The GPLed portion can be replaced with non-infringing code.
2. Other consideration can be given to the original copyright holders. ie. relicensing of the affected code in exchange for money.
If by "wiggle out" you mean "get off scot-free and continue to distribute infringing code" then no that won't happen (at least not to FSF owned code.). At a minimum, they will have to put out a non-infringing mandatory upgrade since they didn't have the right to distribute the original code.
Remember, the GPL doesn't get to dictate penalties for copyright violation (although it may suggest them REALLY strongly). Either a judge does that or an accomodation is reached with the copyright holders in arbitration. Moglen is saying arbitration is their usual approach.
You've answered your own question.
on
GPL's Strength
·
· Score: 5, Informative
The GPL does not restrict or regulate the use of software at all. The only activity the GPL restricts is the conditions under which covered software can be redistributed.
You quoted, "let the user do things that are otherwise illegal (copy and redistribute software)". Most Open Source licenses don't regulate that activity either. The other activity that is regulated is preservation of the original copyright notice: you can't claim someone else's work as your own. The BSD, Apache, X style licenses say you can't plagiarize but can do anything else you want
You are correct, though, Microsoft's EULAs do restrict use of the software as well as copying, redistribution, reverse engineering and reselling.
The point Professor Moglen was trying to make is that in legal terms the GPL is SIMPLER than such a EULA and is therefore less prone to failure. I would imagine that is largely true of most other Open Source licenses.*
* The MPL/NPL, APSL, IBM's public license and other corporate style Open Source licenses tend to be complicated and fail the simplicity criterion.
SciAm's behaivor was completely uncalled for. Mims is a very credible source for electronic's hobbiests. His pencil drawn handbooks contain technical writing that is as clear and succinct as I've ever seen.
I would not take Mims seriously speaking as a creationist or Intelligent Designer or whatever they are going to call it next week. However, I take him very very seriously when it comes to electronics. Fair is fair, and there is nothing inappropriate about recognizing his electronics competence.
SciAm tarnished themselves by not recognizing this and gave creationists one hell of a talking point. Shame on them.
Slashdot Mirror
Just have an arm that can extend either a urinal or bedpan. Any self respecting couchbot should be able to dump them out in the toliet and rinse them without mishap. Heck attach a little sprayer nozzle to it so you won't have to mess with TP either. I bet the Japanese would buy this crap up in a heartbeat. (The average japanese is even more gadget crazy than even the most raving Slashdotter. They even try to have more advanced toliets than their neighbors.)
True. I want KDE for my Debian desktops as well. At least when it finally does arrive it will have several months worth of bugfixes and polish. It's a pity that their release freeze came with all of this good stuff just down the road....Gnome 2, Mozilla 1.0, KDE 3.0.x......
Since Sid (unstable) is the gateway to Testing (Woody about to supplant Potato) it can't even be had from there. The real pisser is that once Woody releases and this all good stuff shows up in Sid it will be wise to wait for it to come to testing(next codename?). The very first truly new bits of software to go in are apt (heh heh!) to be flaky. Oh well, in this context its a GOOD thing that official Debian releases are infrequent.
Microsoft might write into its licenses "you may not run open source software on our platforms".
By all means pile on restrictions as to how people are allowed to use your products. Its yet another argument in favor classifing Microsoft products as "legacy". Since they've maxed out the desktop market we can expect to see Microsoft act more and more like the Scientologists.
The smart thing for them to do is to diversify in such a way that they are not abusing their desktop monopoly. If they play their cards right, they could get out of trouble with the government....permanently.....Shrub won't be around to shield them forever. And they could still continue to grow their revenues albeit at a less incredible rate. I don't expect them to be smart. I expect them to do things like bully school districts and like the Scientologists again fire howitzers at their feet.
Their ankle deep in the water now. If they start drowning, let's throw 'em an anvil. Nasty political campaign style ads with them admitting their software is unsafe at any speed to get out of antitrust charges would be a good start.
I don't have anything to say at the moment about the larger issues being debated in this thread but I do have something to say about random number sources. If I wanted to fill a CD with good random numbers /dev/urandom is not how I would go about it. The quality of /dev/urandom is reasonable as uses bits of fluff like the delay between keypresses and chatter from the device drivers to create an "entropy pool" to seed a pseudorandom algorithm with. The problem is that it is slooooowwww. Most goings on in a normal desktop PC are very very ordered and deterministic. The few that aren't represent a very small amount of entropy.
All of this means that the process that is generating your iso is going to see short bursts of data inbetween long periods of entropy gathering. That CD will probably take hours at least to generate. Also I said the quality of the data is "reasonable". If one means to keep the government or a well heeled corporate attacker out of the cyphertext it may not be good enough. Even the non-deterministic processes in a PC likely have a fair amount of order in them. In other words, that entropy pool is probably good enough to make a 2048 bit assymetric key. It probably wouldn't do for a 650MB iso. The longer the string of numbers, the more likely hidden order can be found.
The way I would is to sample the output of a white noise generator. The output of the ADC is then used to seed a good pseudorandom algorithm. The reason why we use the white noise as a seed is to obliterate any bias in the data caused by such factors as the slew rate, bandpass of the analog circuitry making the white noise or any subtle imperfection that may exist in the ADC. A reverse biased transistor is one source of analog noise. This would be a high speed generator of quality random numbers. The speed would only be limited by the clock rate of the ADC or rate at which the PC can process the output.
The FSF does not AFAIK own any copyrights on the Linux kernel itself. Just because something is GPLed doesn't mean that RMS has Godlike powers to dictate terms over it. The FSF is protective of the GNU tools which they do own the copyrights on and they can indeed haul people into court over those. Making something GPL doesn't make it a part of the GNU project.
The Polish government in exile was based in Britain and their surviving pilots flew with the RAF and the Army Air Corps. Most of these guys were tall, blond-haired, blue-eyed, dashing and had "cute" accents. Needless to say, they got all of the girls when they happened to be around. The piqued American personnel started telling Polack Jokes........
I meant to say (less than) OS X. I think I just got bit by the so-called lameness filter. Sucker wacked off that "less than" looking character.
The problem with co-operative multitasking is that if an app hangs and doesn't hand control back to the kernel or the next app in line then the whole system freezes. Windows 3.x was badly affected by it to an obnoxious degree. And it could bite Win9x systems running 16-bit code. And its still a problem with MacOS 10.x. The woes of co-operative multitasking were my second biggest bitch with the old style MacOSen (lack of protected memory was the first). I'll gladly trade some speed for the ability to kill offending apps.
Now it may be that those Acorns had a higher code quality in general to mitigate that problem but it can still be a problem. For that matter, higher priority processes can be reniced on most pre-emptive systems to make the processes that matter snappy.
There is only one way a co-operative system can be tolerable. If the number of apps is restricted to a limited set that is highly polished and bug free then it might be okay. But forget just chucking any ole program into that environment unless you like looking at hourglasses, busy bees and ticking watches.
The developer can also tell the client that the more they want to own the code the more it is going to cost them. A completely closed app that a client owns ALL rights to means that the developer has to reinvent all of the little wheels and gears in it. The developer cannot even use his own private stock of functions much less anything GPLed.
Ironically, a developer coding such an app could use LGPLed libraries as long as he didn't have to patch the libraries himself (absolute ownership again). For that matter, the developer can also use the BSD and X stuff as long as all copyrights are acknowledged. The personal toolbox however would be completely out unless LGPLed or using one of the acknowledge copyright licenses.
It also means that the developer cannot add anything to his personal toolbox that comes out of an absolute ownership job. The developer is also in peril if he codes anything else that looks like something he did on an absolute ownership job.
That might be the best angle: "I can only give you a non-exclusive license to certain parts of the apps as I cannot put myself out of the market from having worked for you." This is a purely business argument and may go over better. Don't say "open source" at all. Just negotiate space for you to retain ownership of your generic functions and license them according to your inclinations be them Open Source or Proprietary.
We use Apples extensively in our elementaries. We replace them on a five year cycle. For the intended applications, the Apples last far longer than the Wintel boxes in the MS and HS. The way we use them, the price evens out. I'll also point out that in general, the quality of the Apples in a bit higher than el-cheapo Wintel boxes. The amount of hardware failure is much lower than some 233Mhz Gateways I used to service. It's that total cost of ownership thing again. For that matter, keeping the software in a sane state is easier as well. That's not to say that they don't crash from time to time but the recovery tends to be far less difficult than when a Windows machine gets it's registry and filesystem borked up.
Don't let the upfront cost of the Macs fool you. For elementary school classroom purposes, they mostly Just Work. Oh yeah, becoming an Apple Self-Servicing District wasn't very costly either and the technical support is EXCELLENT.
A quality Wintel box running 2000 pretty has these virtues as well...at least they're stable. They cost about as much too. We don't deploy eMachines or PowerSpecs here and with few exceptions there are no Macintosh equivalents to those.
The g in gcc DOES stand for GNU. The C compiler on old school Unices is called cc for C Compiler sensibly enough.
Say I wonder if that isn't the answer to the GNU/Linux bickering: glinux!
My table is almost complete as well. I've got several tons of U238 sitting around that I'll gladly trade for 20 pounds or so of Pu239.
Say, anybody know where I can find some polonium?
The WORST offenders for mp3s where I work are the teachers. You're generous. We gave them a week before whacking the mp3s from the their network shares. We didn't bother with the warning when we found some 200 Dancing Baby AVIs......WHACK!!
It's really cute the way the p2p thing broke down. The High School and the Middle School share the same T1 line. The Middle Schoolers loved Gnutella. One teacher had something like 4 GB of MP3s in his share (quotas have since been put on the network storage). At the High School, Kazaa was King with AudioGalaxy running second. I'd sit there watching the network monitor when lunchtime rolled around. The P2P ports just absolutely spiked through the ceiling....greedy....greedy.
I'm fortunate enough to have an Administration with some clue. We unceremoniously blocked the ports and had an intercom announcement. There's already an AUP but it will be more heavily emphasized next year. It will also be made clear that the technological measures are only there to keep em honest. We don't intend to have an arms race. If someone gets busted then they're busted As I said, the Administration is with us on this one.
I just tell people: "Gnutella's cool but we don't use it at school. Do it at home or at your buddy's house. I like it too but I don't do it here."
My ass. The first thing I would do after an opensource migration is to make it known to former vendors that since "The Software" is not installed that all such agreements are null and void. If they persisted in an audit (there wouldn't be anything to find) then they can expect to be sued for wasting time and resources, barratry, persecution and anything else a good lawyer can think of.
Everyone will just start slamming their brakes at such intersections REGARDLESS of whether it is called for. If you get rear-ended it is almost always the other guys fault regardless of the situation anyway. Will they blame the increase in rearenders on their lights......no. I'll just make a bumper sticker that says "I brake HARD for camera lights." Fair warning for anybody who has to drive behind me.
Some infrared spectrometers do indeed use salt lenses to focus the light. Instruments intended to sample the air have to use dessicant filters because humid air will degrade the lenses.
What's needed is something like patch and diff for audio/video. If one had to follow a precise set of instructions to make their own Phantom Edit then indeed no one would watch it. However, you could stick your Phantom Menance DVD into your machine and uses a viewer that accepts a file of instructions to create a Phantom Edit in real time. It could be more than instructions as well. Replacement sound clips, video overlays, and video clips could be injected as well. As long as the "patches" do not themselves contain trademarked or copyrighted content then nothing legal could be done about this. One would have to own a copy of the original work to use it.
Perhaps boycott is the wrong word here. For instance, I am not boycotting Metallica. Whenever I happen to hear a Metallica tune; I feel disgust at yet another carriage maker bemoaning the automobile. I suppose that if I were a better person I could get past the behaivor of that greedy pighead Lars Ulrich but in this regard I'm sorry to admit I'm small and petty. I don't meet the ideal in this regard; so sue me....oh wait...someone probably will.
If an artist is loud and obnoxious enough in their public life then it WILL change the light in which their art is seen. Sometimes the art can survive the opproprium an artist brings on himself but sometimes not. Lars Ulrich could put out what could have been the best music I have ever heard in my life. He will never ever have to worry about me using PAN to download it from USENET but then he won't have to worry about my buying it either. I'm disgusted with the little prick. Why would I want to leech anything of his?
Microsoft has 21 fairly broad patents that cover among other things, integrating DRM into an OS. The worry is that an SSSCA type law would basically make a Microsoft OS or Microsoft licenced technology the only legal way to use computers in this country.
True. But presumably, the Democrats would have learned their lesson and would not field an Eisner pawn against him when his term comes up. They say that politics makes strange bedfellows for a reason. If the CB??? or SSSCA or whatever they're calling it this week really really offends you then you are going to have to hold your nose and help out the other guy. At this point, I'd support an Alfred E. Neumann candidacy if it would oust Hollings, Feinstein or other scum like that.
Another poster pointed out that the NRA doesn't really vote Republican, they vote Gun. The thing we really have to be wary of is forming alliances with other causes that are ultimately detrimental. Banding with Creationists for one term to get rid of someone like Hollings is an example of the type of thing we need to be prepared to do. There is no need to band with them for all time under any circumstances.
We have many many values that are important to us. Politics will demand that we sometimes accept setbacks in one area to gain victory in another. I don't like it either but that's reality folks.
Do the schnozbuttons really taste like schnozbuttons?
I once had to retire a Mac LC II was the building fileserver. This thing had financials, the private records of students; you name it. I low-leveled the drive and wrote 0's to it. Once that was done, I drilled several holes through the platters. I broke the bit off the drill in the process. The drive with drill bit stub stuck in it looks like Count Datatula with a spike through his heart. We keep the spiked carcass around to show people how to make sure that sensitive data gets destroyed.
I don't believe that is what usually happens if some GPLed stuff winds up in a larger work. There are other options:
1. The GPLed portion can be replaced with non-infringing code.
2. Other consideration can be given to the original copyright holders. ie. relicensing of the affected code in exchange for money.
If by "wiggle out" you mean "get off scot-free and continue to distribute infringing code" then no that won't happen (at least not to FSF owned code.). At a minimum, they will have to put out a non-infringing mandatory upgrade since they didn't have the right to distribute the original code.
Remember, the GPL doesn't get to dictate penalties for copyright violation (although it may suggest them REALLY strongly). Either a judge does that or an accomodation is reached with the copyright holders in arbitration. Moglen is saying arbitration is their usual approach.
The GPL does not restrict or regulate the use of software at all. The only activity the GPL restricts is the conditions under which covered software can be redistributed.
You quoted, "let the user do things that are otherwise illegal (copy and redistribute software)". Most Open Source licenses don't regulate that activity either. The other activity that is regulated is preservation of the original copyright notice: you can't claim someone else's work as your own. The BSD, Apache, X style licenses say you can't plagiarize but can do anything else you want
You are correct, though, Microsoft's EULAs do restrict use of the software as well as copying, redistribution, reverse engineering and reselling.
The point Professor Moglen was trying to make is that in legal terms the GPL is SIMPLER than such a EULA and is therefore less prone to failure. I would imagine that is largely true of most other Open Source licenses.*
* The MPL/NPL, APSL, IBM's public license and other corporate style Open Source licenses tend to be complicated and fail the simplicity criterion.
SciAm's behaivor was completely uncalled for. Mims is a very credible source for electronic's hobbiests. His pencil drawn handbooks contain technical writing that is as clear and succinct as I've ever seen.
I would not take Mims seriously speaking as a creationist or Intelligent Designer or whatever they are going to call it next week. However, I take him very very seriously when it comes to electronics. Fair is fair, and there is nothing inappropriate about recognizing his electronics competence.
SciAm tarnished themselves by not recognizing this and gave creationists one hell of a talking point. Shame on them.