Ternary computing could be quite useful for asynchronous logic. The three states would be 1/yes, 0/no, and n/(no answer yet).
Basically you want the truth table to be, in order of precedence: 0 AND * = 0 n AND 1 = n n AND n = n 1 AND 1 = 1 (OR is the reverse, swap 0 and 1) NOT 1 = 0, NOT 0 = 1, NOT n = n
Gates can be actually made to follow the right truth tables by diddling your substrate voltages in an otherwise fairly standard CMOS design; this has the effect of making your circuit twice as slow or quadrupling its power consumption though, which sucks. You also have to watch your noise thresholds here, transients can be nasty although they are unlikely to propagate far through a network of n's. This can be mostly fixed by further tinkering with thresholds, but then the leakage current becomes prohibitively high.
You can also just design extra logic with standard gates and watch your glitches very carefully.
There may be cleverer ways to do this, or the savings of asynch might be enough to make it useful anyway.
It can't be a Warhol worm. That name only applies to worms which use a remote vulnerability, and hence can infect a vulnerable computer immediately.
A mail worm is different. Sure, it can be more or less efficient, but it can't be a Warhol worm. It can't ever spread across the internet in 15 minutes (or even less, like the Flash worm).
Yeah, I thought that would be really cool too, and maybe you could even get diamond heatsinks to top it off [shiny, but expen$ive], but remember that light does weird things to electronics, most notably increase leakages. Those chip packages are black for a reason.
I've often wondered whether we couldn't attach a Stirling engine directly to the surface of the processor and recover some of the energy being lost as heat and turn it into electricity.
Well, basically because of thermodynamics. Recovering energy from a heat gradient slows heat flow along that gradient. If your CPU runs hot enough to need a fan, this would melt it for sure. If not, you're probably not going to get much energy out of it anyway, and your CPU is probably already pretty efficient.
Maybe if they ever build CPUs out of diamond (as some previous article mentioned was being researched), you could do something like this. But I doubt it would be practical. I think in the end, you'd be better off using your CPU or its heat exchanger as a grill or space-heater.
[slightly offtopic] On the subject of diamond CPUs, it would be really cool if someone (it would probably be Apple) could eventually make a system whose chipset was all transparent diamond, with translucent PCBs. I don't think it'll ever happen because of heatsinks and light absorbtion problems, but it would look so damn cool...
Now, I may be more than a bit biased in this regard (my girlfriend is homeschooled), but I have to take exception to your comment. While the homeschoolers I know are more introverted than many of my friends (blame it on the lack of "socialization" if you want, but most of them are geeks and geeks are not generally outgoing), they are certainly not screwed up either academically or socially. Most of them are extremely smart, both in the amount of information they know and in their skills in applying it.
I'm sure that I'm not typical in my experiences with homeschoolers. Some go to my church, and some former homeschoolers went to my high school. The others I met through math competitions (USA Math Olympiad), or through my girlfriend (whom I also met through math stuff; we both scored in the top 12 nationally on the USAMO). While this is clearly a biased slice, you might notice that homeschooled children have disproportionately high representation in all sorts of academic competitions (national spelling bee, math competitions, etc).
And yes, many of them have teachers for parents. But I would guess this is true of a large proportion of homeschoolers, as teachers are often critical of public schools and can't afford private ones.
Unfortunately, SAT scores are a really lousy metric. I wasn't able to find any info on other standardized tests, such as those given in elementary school; these would probably be better indicators. Students who homeschool all the way through high school probably have a better time than those who just homeschool for a year or two.
You're right. It's not flamebait. It's a troll. If I got to metamod that, it'd go down as unfair.
He just took an old troll post and changed the machine name to G5. As a hint of this, there is no dual G5 1.6GHz; the 1.6GHz machine is single-processor and has no PCI-X. Furthermore, what the hell version of Netscape is he running? The boxes ship with MSIE and possibly Safari; there hasn't been a new version of Netscape in years, and he'd have to run it in classic.
But hey, at least it's better than Father Randy "Pudge" O'Day.
You're right that Mach is a microkernel. However, a microkernel does not sit on top of a kernel. What distinguishes a microkernel architecture from a monolithic kernel (which is most OSes) is that it does not have one unified address space and thread state. Instead, it is distributed among several services (such as virtual memory, filesystems, drivers, network stack...) which pass messages to each other.
There is an ongoing debate as to the virtues of microkernel architectures, but stereotypically, microkernels are slower but more stable and secure, and they are harder to create in the first place but easier to extend. Some of this is just stereotype, especially given kernel modules on the one side and improvements in message passing and context switching on the other.
MacOS X, QNX, Mach and the Hurd are microkernels; Linux, *BSD and Windows are monolithic. OS 9 is neither, as the whole system has a unified address space (which makes it fast but unstable).
Have you actually had to use one of those bargain-basement computer systems?
Yes. I used an eMac all year at college.
They're torturously slow,
Well, the POV-Ray rendering speeds and the time to compile X11 from fink were pretty slow, but for most other stuff it was plenty fast.
plagued with instability problems,
Yeah, it panicked maybe 3 times all year, but it turned out this was Norton's POS antivirus software which Harvard recommended... I uninstalled that and it hasn't crashed since. My uptimes were usually a couple weeks, generally broken up by tired nights when I got pissed at the (admittedly loud) fan and the obnoxious "breathing" power LED.
and in general are something I'd rather not have to deal with on a regular basis.
Hmm. Different strokes, I guess.
PS: if you, like some Harvard admins, take a first-generation iMac, retrofit it with OSX 10.1, add a poorly-setup netboot image, and add a bunch of third-party hacks in the name of "security" (despite having isolated users and a netboot image that refreshes every reboot), then yes, it is pretty much unbearable.
By settling, they are effectively saying "We don't think we were completely in the wrong, but we see how some might think so and would rather come to an amiable solution and move on".
No. They said it themselves: they think (or at least claim) that they're entirely in the right, but that they don't want to spend the legal fees fighting this off.
A tic-tac-toe program? Dude, you have to be kidding me. The game tree for tic-tac-toe is so small (there's only 200-some thousand, and that's discounting symmetry and pruning) that it is trivial to play perfectly.
Checkers, maybe? Chess? Something just a bit more complicated???
I've seen basically two kinds of posts in resoponse the the article. There's been Mac zealots who are lauding Apple for "taking responsibility," and there have been people slamming them for false advertising. There have also been several "OMG APPLE IS TEH SUX" trolls and jokes of various degrees of wittiness, but that's to be expected.
First things first. Apple is not "taking responsibility" for all those users out there with poor OS X support. They are refunding them, and doing so to settle a lawsuit. Taking responsibility would be adding full support to OS X. Their settlement sounds pretty fair as far as class action suits go, but it's not done out of benevolence. As has been pointed out before, Apple is a company, and its motivation is primarily money.
Second, Apple's advertising in this matter was not exceptionally shady. They said they'd support G3s, and it turns out that they do not fully support all of them. The OS runs on these G3s (which are below the recommended minimum specs), but poorly, and doesn't fully support their hardware.
Fine. All companies advertise in language like this. Plenty of laptops are in various Linux distros' supported lists, but don't have power management support. If you think promising to support something and then doing a half-assed job is bad, look at advertisers that get away with what would be called a lie anywhere outside a court.
This case is neither a big win nor a big loss for Apple. It does not show that their character is particularly good or evil. Give it a rest.
You apparently don't know Latin. First, the plural of cactus is "cacti," and by the same logic, the plural of "virus" should be "viri," not "virii," which would be the plural of "virius" (which is not a word in Latin or English).
Second, Latin is annoying in that it has more than one declension table, and no shortage of complex rules. For instance, the plural of "genus" is "genera." "Viri" is already a Latin plural; it means "men" (plural of "vir"), and it is also already the genetive singular of "virus." The tricky thing is that "virus" is one of only a few words in its declension/gender/ending group, and all these words are mass-nouns (ie, they refer to a substance and therefore do not have plurals, just as "milk" does not have a plural in English). "Virus" itself means "poison," and in Latin it is a mass noun and has no (recorded) plural.
See also corpus->corpora, status->status (with a long u instead of short), octopus->octopodes (in Greek).
Come on, if you're gonna s/Mac/BSD/gi, at least be creative about it. You did change G3 to PIII, but remember that BSD is an OS, while Mac is a hardware architecture, so phases like "the BSD machine's faster chip architecture" don't make sense anymore, and since it's free, there's no "cheaper."
Some of that stuff you request is here in OS X. Not all of it, of course, but I can address most of the software stuff...
* Software that works. Office is a good example of bad things. Why the hell do features that worked in an earlier version of office get broken in later editions. I would think software would evolve in such a way that stuff gets better, not stuff added on and and the old stuff worse.
Can't help you much there. Office for Mac is almost as broken as Office for Windows. OpenOffice.org pretty much sucks. AppleWorks is, ironically, the epitome of a bad Carbon port. I used TeXshop most of the time, but the learning curve on TeX is rather steep.
In other areas though, Mail is the best email client I've ever used (beats out KMail, Outlook Express, Entourage, Eudora, Claris, Mozilla, webmail and ssh+pine), and Safari is the best browser I've used (beats out Konq, MSIE, Netscape/Mozilla, Chimera/Camino, and Phoenix/Firebird). Neither one is perfect (brushed metal???), but the clients are fast and the user interfaces are very clean. Most of the utilities that ship with the system (basically wrappers around BSD tools such as cdrecord, hdutil, fdisk, etc) are very well-designed also.
* Better GUI... obviously. I've used KDE, Knome, Win2000 (like Win98), and WinXP... XP still messes me up everytime. Why did they change the start menu. I know, go and change the scheme.
OS X has an excellent GUI. It could use a few additions, such as multiple desktops, and the Finder needs to be fixed (happening in 10.3). FFM would be nice too, but it's incompatible with Fitt's-Law-compliant menubars.
However, applications integrate very nicely in Cocoa and in Carbon, with AppleEvents and such, and the system control GUIs are excellent. Compare, for instance, difficulty in setting up a firewall on Windows, Linux, *BSD, and OS X. Or a webserver. Or whatever.
* Connectivity. 'tis getting better. Google is good. P2P is good. Email is ok. IM is good. Video phone anyone?
iSight + built-in mic + iChat = video and voice chat.
* Related to tech: telecommunications. fucking joke. With lots dark fiber out there, phone services should be a dirt cheap commindity. land lines are a joke. Everyone, please get broadband, if you can, and dump your landline. The baby bells need to suffer.
You can get a VoIP client for Mac, of course, but the real hurdle here is for everyone else to get VoIP. (The Ukraine, btw, is going for landlines + VoIP. It's actually pretty scary how far ahead they are in terms of technological infrastructure. Nobody uses checks anymore, everything is wired.)
* Music and Video on demand. There is no good technical reason that I shouldn't be able to purchase and instantly listen to any audio or video thing ever created. Big media blows, I hope they bankrupt with the telecoms.
The iTMS will get you music on demand. They don't have every label, but they have a pretty large selection. $1 a song. Now, video, that would be nice...
* Backups. Consumer level PC need a VERY GOOD inexpensive method of backing up stuff... I'm talking the whole hard drive in a manner of minutes. Cheap. Often.
I don't know about this, as I've only looked at free tools. I made incremental backups to CD (cheapest you can get) every week at school, and a full backup to DVD before going home for the summer. However, I wouldn't say that Joe User can do this very easily.
I accomplished it with a cronjob that runs a Perl script I wrote. Every Friday, it would check for changed files, handle the files in there according to an RC (most of them got hfstar -czvpf, but the more private ones were encrypted with gpg), and collect them in one place. When I got back from class, I'd add anything special I wanted to backup (usually a network-account image), and it'd make an ISO and burn them to multisession CDR.
I might upload my backup utils, but they'd be just one package among many.
Ah, yes, I misread. I had thought it was 168,000, plus several thousand before.
But still, you probably can't just comment out all those lines of code; you'd probably have to rewrite a fair amount of code to remove it, and you might end up reimplementing what SCO did. Even if it is all SMP stuff, implementing SMP (NUMA, whatever they added) in your kernel can have an effect on the single-proc version too (for instance, the threading and synch models).
Slashdot Mirror
Ternary computing could be quite useful for asynchronous logic. The three states would be 1/yes, 0/no, and n/(no answer yet).
Basically you want the truth table to be, in order of precedence:
0 AND * = 0
n AND 1 = n
n AND n = n
1 AND 1 = 1
(OR is the reverse, swap 0 and 1)
NOT 1 = 0, NOT 0 = 1, NOT n = n
Gates can be actually made to follow the right truth tables by diddling your substrate voltages in an otherwise fairly standard CMOS design; this has the effect of making your circuit twice as slow or quadrupling its power consumption though, which sucks. You also have to watch your noise thresholds here, transients can be nasty although they are unlikely to propagate far through a network of n's. This can be mostly fixed by further tinkering with thresholds, but then the leakage current becomes prohibitively high.
You can also just design extra logic with standard gates and watch your glitches very carefully.
There may be cleverer ways to do this, or the savings of asynch might be enough to make it useful anyway.
With some practice you can even handle the unusual states like 21 and 27 easily (I use my thumb as 2^0).
Yeah, but watch out for 132.
Yes you do, as at some point you'll have a connection between the + and - rail and get 6.6V.
It can't be a Warhol worm. That name only applies to worms which use a remote vulnerability, and hence can infect a vulnerable computer immediately.
A mail worm is different. Sure, it can be more or less efficient, but it can't be a Warhol worm. It can't ever spread across the internet in 15 minutes (or even less, like the Flash worm).
So what do you call getting numerous AOL installation CDs
Free coasters.
Yeah, I thought that would be really cool too, and maybe you could even get diamond heatsinks to top it off [shiny, but expen$ive], but remember that light does weird things to electronics, most notably increase leakages. Those chip packages are black for a reason.
Hm. I don't hear any hum in my dad's (gigantic, bitching) solid-state amps. And I have very sensitive hearing.
You sure it's there?
(having never heard a tube-amp system, I can't really compare tho)
Was there ever a virus exclusively for linux?
Yeah, the Linux Slapper Worm. It used a remote root hole in Apache (IIRC) to cause havoc.
I've often wondered whether we couldn't attach a Stirling engine directly
to the surface of the processor and recover some of the energy being lost
as heat and turn it into electricity.
Well, basically because of thermodynamics. Recovering energy from a heat gradient slows heat flow along that gradient. If your CPU runs hot enough to need a fan, this would melt it for sure. If not, you're probably not going to get much energy out of it anyway, and your CPU is probably already pretty efficient.
Maybe if they ever build CPUs out of diamond (as some previous article mentioned was being researched), you could do something like this. But I doubt it would be practical. I think in the end, you'd be better off using your CPU or its heat exchanger as a grill or space-heater.
[slightly offtopic] On the subject of diamond CPUs, it would be really cool if someone (it would probably be Apple) could eventually make a system whose chipset was all transparent diamond, with translucent PCBs. I don't think it'll ever happen because of heatsinks and light absorbtion problems, but it would look so damn cool...
Now, I may be more than a bit biased in this regard (my girlfriend is homeschooled), but I have to take exception to your comment. While the homeschoolers I know are more introverted than many of my friends (blame it on the lack of "socialization" if you want, but most of them are geeks and geeks are not generally outgoing), they are certainly not screwed up either academically or socially. Most of them are extremely smart, both in the amount of information they know and in their skills in applying it.
I'm sure that I'm not typical in my experiences with homeschoolers. Some go to my church, and some former homeschoolers went to my high school. The others I met through math competitions (USA Math Olympiad), or through my girlfriend (whom I also met through math stuff; we both scored in the top 12 nationally on the USAMO). While this is clearly a biased slice, you might notice that homeschooled children have disproportionately high representation in all sorts of academic competitions (national spelling bee, math competitions, etc).
And yes, many of them have teachers for parents. But I would guess this is true of a large proportion of homeschoolers, as teachers are often critical of public schools and can't afford private ones.
I googled briefly for numbers, and here's what I found:
"The average SAT scores of home-schooled students were 568 Verbal and 532 Math, above the national averages of 505 Verbal and 514 Math."
Unfortunately, SAT scores are a really lousy metric. I wasn't able to find any info on other standardized tests, such as those given in elementary school; these would probably be better indicators. Students who homeschool all the way through high school probably have a better time than those who just homeschool for a year or two.
You're right. It's not flamebait. It's a troll. If I got to metamod that, it'd go down as unfair.
He just took an old troll post and changed the machine name to G5. As a hint of this, there is no dual G5 1.6GHz; the 1.6GHz machine is single-processor and has no PCI-X. Furthermore, what the hell version of Netscape is he running? The boxes ship with MSIE and possibly Safari; there hasn't been a new version of Netscape in years, and he'd have to run it in classic.
But hey, at least it's better than Father Randy "Pudge" O'Day.
Heck, a *helium* cylinder can kill you if mishandled.
<voice="high;sqeaky">You mean heeeleeeum can be dangerous tooooo... no waaaay...</voice>
You're right that Mach is a microkernel. However, a microkernel does not sit on top of a kernel. What distinguishes a microkernel architecture from a monolithic kernel (which is most OSes) is that it does not have one unified address space and thread state. Instead, it is distributed among several services (such as virtual memory, filesystems, drivers, network stack...) which pass messages to each other.
There is an ongoing debate as to the virtues of microkernel architectures, but stereotypically, microkernels are slower but more stable and secure, and they are harder to create in the first place but easier to extend. Some of this is just stereotype, especially given kernel modules on the one side and improvements in message passing and context switching on the other.
MacOS X, QNX, Mach and the Hurd are microkernels; Linux, *BSD and Windows are monolithic. OS 9 is neither, as the whole system has a unified address space (which makes it fast but unstable).
Have you actually had to use one of those bargain-basement computer systems?
Yes. I used an eMac all year at college.
They're torturously slow,
Well, the POV-Ray rendering speeds and the time to compile X11 from fink were pretty slow, but for most other stuff it was plenty fast.
plagued with instability problems,
Yeah, it panicked maybe 3 times all year, but it turned out this was Norton's POS antivirus software which Harvard recommended... I uninstalled that and it hasn't crashed since. My uptimes were usually a couple weeks, generally broken up by tired nights when I got pissed at the (admittedly loud) fan and the obnoxious "breathing" power LED.
and in general are something I'd rather not have to deal with on a regular basis.
Hmm. Different strokes, I guess.
PS: if you, like some Harvard admins, take a first-generation iMac, retrofit it with OSX 10.1, add a poorly-setup netboot image, and add a bunch of third-party hacks in the name of "security" (despite having isolated users and a netboot image that refreshes every reboot), then yes, it is pretty much unbearable.
rm -f -- --help
By settling, they are effectively saying "We don't think we were completely in the wrong, but we see how some might think so and would rather come to an amiable solution and move on".
No. They said it themselves: they think (or at least claim) that they're entirely in the right, but that they don't want to spend the legal fees fighting this off.
A tic-tac-toe program? Dude, you have to be kidding me. The game tree for tic-tac-toe is so small (there's only 200-some thousand, and that's discounting symmetry and pruning) that it is trivial to play perfectly.
Checkers, maybe? Chess? Something just a bit more complicated???
I've seen basically two kinds of posts in resoponse the the article. There's been Mac zealots who are lauding Apple for "taking responsibility," and there have been people slamming them for false advertising. There have also been several "OMG APPLE IS TEH SUX" trolls and jokes of various degrees of wittiness, but that's to be expected.
First things first. Apple is not "taking responsibility" for all those users out there with poor OS X support. They are refunding them, and doing so to settle a lawsuit. Taking responsibility would be adding full support to OS X. Their settlement sounds pretty fair as far as class action suits go, but it's not done out of benevolence. As has been pointed out before, Apple is a company, and its motivation is primarily money.
Second, Apple's advertising in this matter was not exceptionally shady. They said they'd support G3s, and it turns out that they do not fully support all of them. The OS runs on these G3s (which are below the recommended minimum specs), but poorly, and doesn't fully support their hardware.
Fine. All companies advertise in language like this. Plenty of laptops are in various Linux distros' supported lists, but don't have power management support. If you think promising to support something and then doing a half-assed job is bad, look at advertisers that get away with what would be called a lie anywhere outside a court.
This case is neither a big win nor a big loss for Apple. It does not show that their character is particularly good or evil. Give it a rest.
You apparently don't know Latin. First, the plural of cactus is "cacti," and by the same logic, the plural of "virus" should be "viri," not "virii," which would be the plural of "virius" (which is not a word in Latin or English).
Second, Latin is annoying in that it has more than one declension table, and no shortage of complex rules. For instance, the plural of "genus" is "genera." "Viri" is already a Latin plural; it means "men" (plural of "vir"), and it is also already the genetive singular of "virus." The tricky thing is that "virus" is one of only a few words in its declension/gender/ending group, and all these words are mass-nouns (ie, they refer to a substance and therefore do not have plurals, just as "milk" does not have a plural in English). "Virus" itself means "poison," and in Latin it is a mass noun and has no (recorded) plural.
See also corpus->corpora, status->status (with a long u instead of short), octopus->octopodes (in Greek).
ref
If you can get it at a physical Barnes & Noble in your own town, you could easily save $12 in shipping.
Come on, if you're gonna s/Mac/BSD/gi, at least be creative about it. You did change G3 to PIII, but remember that BSD is an OS, while Mac is a hardware architecture, so phases like "the BSD machine's faster chip architecture" don't make sense anymore, and since it's free, there's no "cheaper."
Some of that stuff you request is here in OS X. Not all of it, of course, but I can address most of the software stuff...
* Software that works. Office is a good example of bad things. Why the hell do features that worked in an earlier version of office get broken in later editions. I would think software would evolve in such a way that stuff gets better, not stuff added on and and the old stuff worse.
Can't help you much there. Office for Mac is almost as broken as Office for Windows. OpenOffice.org pretty much sucks. AppleWorks is, ironically, the epitome of a bad Carbon port. I used TeXshop most of the time, but the learning curve on TeX is rather steep.
In other areas though, Mail is the best email client I've ever used (beats out KMail, Outlook Express, Entourage, Eudora, Claris, Mozilla, webmail and ssh+pine), and Safari is the best browser I've used (beats out Konq, MSIE, Netscape/Mozilla, Chimera/Camino, and Phoenix/Firebird). Neither one is perfect (brushed metal???), but the clients are fast and the user interfaces are very clean. Most of the utilities that ship with the system (basically wrappers around BSD tools such as cdrecord, hdutil, fdisk, etc) are very well-designed also.
* Better GUI... obviously. I've used KDE, Knome, Win2000 (like Win98), and WinXP... XP still messes me up everytime. Why did they change the start menu. I know, go and change the scheme.
OS X has an excellent GUI. It could use a few additions, such as multiple desktops, and the Finder needs to be fixed (happening in 10.3). FFM would be nice too, but it's incompatible with Fitt's-Law-compliant menubars.
However, applications integrate very nicely in Cocoa and in Carbon, with AppleEvents and such, and the system control GUIs are excellent. Compare, for instance, difficulty in setting up a firewall on Windows, Linux, *BSD, and OS X. Or a webserver. Or whatever.
* Connectivity. 'tis getting better. Google is good. P2P is good. Email is ok. IM is good. Video phone anyone?
iSight + built-in mic + iChat = video and voice chat.
* Related to tech: telecommunications. fucking joke. With lots dark fiber out there, phone services should be a dirt cheap commindity. land lines are a joke. Everyone, please get broadband, if you can, and dump your landline. The baby bells need to suffer.
You can get a VoIP client for Mac, of course, but the real hurdle here is for everyone else to get VoIP. (The Ukraine, btw, is going for landlines + VoIP. It's actually pretty scary how far ahead they are in terms of technological infrastructure. Nobody uses checks anymore, everything is wired.)
* Music and Video on demand. There is no good technical reason that I shouldn't be able to purchase and instantly listen to any audio or video thing ever created. Big media blows, I hope they bankrupt with the telecoms.
The iTMS will get you music on demand. They don't have every label, but they have a pretty large selection. $1 a song. Now, video, that would be nice...
* Backups. Consumer level PC need a VERY GOOD inexpensive method of backing up stuff... I'm talking the whole hard drive in a manner of minutes. Cheap. Often.
I don't know about this, as I've only looked at free tools. I made incremental backups to CD (cheapest you can get) every week at school, and a full backup to DVD before going home for the summer. However, I wouldn't say that Joe User can do this very easily.
I accomplished it with a cronjob that runs a Perl script I wrote. Every Friday, it would check for changed files, handle the files in there according to an RC (most of them got hfstar -czvpf, but the more private ones were encrypted with gpg), and collect them in one place. When I got back from class, I'd add anything special I wanted to backup (usually a network-account image), and it'd make an ISO and burn them to multisession CDR.
I might upload my backup utils, but they'd be just one package among many.
Yes, it's a fact... 10 years later, it is undeniable that *BSD is dying. Netcraft confirms...
Ah, yes, I misread. I had thought it was 168,000, plus several thousand before.
But still, you probably can't just comment out all those lines of code; you'd probably have to rewrite a fair amount of code to remove it, and you might end up reimplementing what SCO did. Even if it is all SMP stuff, implementing SMP (NUMA, whatever they added) in your kernel can have an effect on the single-proc version too (for instance, the threading and synch models).
Just release the infringing code lines so we can comment it out, doom your profits, and get on with our lives!
Riiiiiight... like you're going to comment out some 200,000 lines of code in your kernel.