You need the ability to make additional phone charges, the way 900 numbers can, on home phone numbers. You could have a pre-ring recording warn callers that, at your discretion, you can charge them some arbitrary amount of money. (Sure, you could zap non-telemarketers too, but you'd just be punishing yourself by scaring away your friends.) -------
First you'd need to express the function "used to work for" in the binary boolean calculus, and then you'd need to express that ^2 function in the same calculus. Actually, ^2 and the boolean algebra already share common formal systems (Pascal, for example, or an extended lambda calculus), so I suppose the real weak link here is reducing "used to work for" to some formal system. Best of luck! -------
It's hardly likely to happen without an AI of some sort directing it.
Is HIV intelligent? Was the bubonic plague sentient? Could ice-nine, if it were real, pass the Turing test?
For it to be a menace it would have to be moderately intelligent as well.
No, it most certainly wouldn't. Joy's point was that, with nanotechnology, the whole world becomes like a computer system. Ever read virus source code? It's not even remotely interesting from an AI point of view. The ONLY requirement for destructive code is that it properly exploits weaknesses in the security of a system. Really and truly, honest to god. -------
Self-replicating nanotech doesn't need any "AI agenda" to cause mass destruction. Like any computer virus, or like Vonnegut's 'ice-nine', the very act of replication can be devastating. -------
Your point about bank robbery was interesting -- how does one distinguish performance art from crime? The key factor is that "speech" (the subset of all actions) exists on an abstract level; speech is the manipulation of an abstract world sandboxed from the "real world." So we have can have crime in speech (pretending to kill someone) until it affects things in life (actually killing someone, and filming it).
I think issues about pornography (rape, kiddie porn, etc.), violence (no "Mewtwo guns down his classmates" Pokemon episode), drug use, etc. in speech fall back to this. You have unlimited freedom in speech, as long as it's sufficiently distanced from the real world.
This is the rule of thumb implied by "source code as free speech." Of course there's really no such thing as "source code", but perhaps the idea is that some layer of abstraction (even if it is recursively enumerable) be put between the average computer user and the average computer. -------
The whole idea of "source code" is an historical accident, stemming from the way Von Neumann architectures are built. For example, some folks have built LISP machines -- chips that execute LISP code directly in hardware. Is this compiled code, or source code? I could take that same LISP code and compile it to x86 machine language. What is that LISP code now? Is it source code, or compiled code?
All sufficently general formal languages are identical in power. Whether a formal statement is written in binary boolean calculus, or that same boolean statement in an electrical diagram, or that same statement sculpted out of transistors, is irrelevant. -------
While infrastructure -- the invisible technological foundations -- should be standardized for interoperability's sake, homogenized interfaces are bad all-around. This is because the interfaces that an individual works most efficiently with vary from person to person. A novice will set up TCP/IP on one machine most efficiently with a hand-holding gui, while an expert will set TCP/IP up most efficiently (and possibly on a large number of machines, quickly) with text files and scripts. Same infrastructure, different interfaces.
Why can't you keep all the physics (or whatever; the data and computation needed to do worthwhile cheating) on the server and put only bare interface stuff on the client? Would this be a much higher data transfer requirement; like would the graphics be a lot harder to compute? That seems like a viable solution.
If this is too much of a bandwidth hog, make some computational sandbox for the server to download cheat-worthy computation to.
For every function F, there's an inifinite set X of programs X1, X2,... that compute that function. (No, X is not recursively enumerable, but there exist subsets of X that are.) Write a program Y that generates some (obscure) binary Xi that computes F each time it is run (where i is some positive integer). Release the source to Y. When a client logs in, have her download the binary for Xi. You can verify that this is a legitimate binary. (I think yacc or lex dealt with similar issues.) You wouldn't even need a sandbox, necessarily, although that might improve portability.
You might not even need for X to be inifinite, although someone could take an Xi and make a Xj with the same hash value; if X were infinite you (in theory) could discard used Xi's.
Remember all those great RSA-in-perl decryption gimmics that were (and may still be) floating around? Someone needs to condense DeCSS to few enough lines to print on trinkets. I want my dirty underwear to violate the MPAA's intellectual property "rights", dammit!
If the assembly language is interpreted, can't it (slowly) appear to be a true 128-bit processor? Indeed, 128-bits is the processor's I/O size; in theory the processor width can be anything. (Although this doesn't take into account the associated speed hit of emulating a width higher than the one used by the native language. Of course, none of that emulation-speed-hit stuff has been discussed in sufficient detail by Transmeta, nor probably will be until the real assembly language is reverse engineered.)
Opening up the source code is an offtopic issue. Finkployd asserted that moderation of articles wouldn't work because, in his opinion, the low-brow Jerry-Springer-loving masses don't have the brains to properly evaluate the articles' merits. (I replied that the comment moderation system relies on the wisdom of those low-brow masses, and that ironically the trailer trash had ranked M{s|r}. Ployd's comments highly.) This has nothing to do with the release of source code, as you discuss, or with whether I should "go off and start my own site", as Fink discusses. The issue at hand is whether the Great Unwashed are mentally fit to rank the articles themselves; by their ability to properly rank replies to the articles, they have demonstrated this mental fitness.
If you give/. to the masses, it will ruin it. Are you aware how stupid these masses are?
(Score:4)
Gee, Finkployd, those faceless nabobs seem to like what you write. Appealing to the lowest common denominator, were we?
The current system of moderation and metamoderation encourages thoughtful posting and ranking. There is no reason (your elitist fear of the Great Unwashed aside) to think that the same could not be applied to story postings.
Recently, I've read several articles that fondly recall the days of mainframe programs, which tended to be less buggy and more open than programs of today. I recall from documentaries that Jobs had a strong disdain for mainframe programmers, viewing them as uptight and uncreative. Today, MacOS developers are hard at work retrofitting the existing (IMHO, feature-bloated) MacOS to BSD; an open, unsexy operating system built by mainframe programmers. How do you think Jobs' manager-as-cowboy style of software development management has affected the overall computer industry, if at all? And what are some ways that the current preference of glitz over substance in software can be changed? (Yes, this question is framed in a biased way -- that stems from my fond memories of the solid, flexible design of the ][.)
What you abolitionists don't understand is that we Southern plantation owners have a legal, enforceable duty to our families and estates to protect our plantations from exploitation by wage-greedy workers. Furthermore, that same duty compels us to maximize the output and value of our estates. Like it or not, in today's business and legal climate, my fellow plantation owners and I must not only keep negro slaves, but also fight to protect that right, lest the entire economy of the South collapse and we be held liable.
--J. Random Plantation Owner, 1855
Re:About mouse usage with keyboard
on
Interface Zen
·
· Score: 2
I actually posted a comment like this a couple of weeks ago. The only reply I got was that "tongue mice" (mouth joysticks) are used for paraplegics and amputees; however, those mice must be molded to their mouths like retainers. (That would not be necessary for those with the manual ability to put a mouse back in, should it fall out.) After meditating on the subject and discussing it with various people, here are my thoughts:
1) People are totally grossed out by the idea of "tongue mice." It evokes the image of licking a mouse, not the cleanest thing around. Also, people I've talked to expressed fear that these devices would become like retainers, reeking and covered with old spit. The portion of these devices put into one's mouth MUST be replacable and disposable. No commercial success of such a device would be otherwise possible. (Also, a name other than "tongue mouse" would be needed -- I think "hook" would be a good name; it would evoke the idea of a hookah or fish hook, not of a retainer.) But the main thing is to eliminate fear of putting a nasty, public thing in one's mouth; making the intra-oral parts of these things disposable would do that.
2) Rather than a touchpad, I think a good design would be a small joystick with a single button, the approximate size and shape of a cigarette holder. Cigarette holders are gripped between the teeth; a (roughly) ball-shaped button could be put here for bite-clicking. The actual joystick part could be put on the end, where the smoke would come out of the cigarette holder, as it were. It could be the approximate size and shape of a clitoris (:>), large enough to get one's tongue around, small enough not to get in the way. The intra-oral part probably could be designed cheaply as a disposable tip, as not much of the device would actually go in the mouth.
All in all, I think it's pretty marketable if done correctly. People tend to forget how cool smoking is -- you can bite contemplatively on a cigarette, you can use it to gesture, etc. -- and this would be a non-eating substitute for that oral fixation. Plus it would totally eliminate RSI from pointing.
I've searched around, but no one is doing this as a mass-consumption product. (Several vendors have feet mice.) I found no relevant patents on the IBM patent search.
Doctors amputate Turkish earthquake survivor's arm [This story contains video]
Any recursively enumerable set of hunks of code can be checked for by a virus scanner, regardless of the size of those hunks of code. The code you describe is not too small for a virus scanner to search for, and is probably (these things can never be exact) unique enough not to conflict with existing code.
I recall this issue having come up in Phrack, in essays on "mutating" code. A way to make viruses "mutate," it was argued, would be to keep the main virus instructions "encrypted" (obfuscated, really), and wrap encryption/decryption code around that (usually this was very small XOR "encryption", not very large code at all). The problem was that a virus scanner could check for this encryption code and thus detect the virus. The same dillema would exist with bootstrapping code.
Doctors amputate Turkish earthquake survivor's arm [This story contains video]
Some of the questions for Schneier seemed, on the surface, a little insulting. The one I'm thinking of said (paraphrased): "You don't have a Phd., and therefore don't have the mathematical background to be a true expert in cryptography. Why should anyone listen to you?" [The question was a lot more polite than that, but that was the rough gist.] Schneier responded by listing the requirements of an experienced cryptographer, and arguing that a cryptographer needs deep knowledge of many different fields (not just mathematics). He used it as an opportunity to demonstrate a deep understanding of his field, and why he could consider himself a qualified expert in it. That seemingly-condescending question was really a good opportunity for Schneier to toot his own horn, like the Onion's standard "Why should anyone buy your record?" band interview question.
These "Why did you..." questions are of the same breed. "Here are some of the bad things said about you, now address them and make them go away." If JP *had* acted in a defensible way, he should have offered a good defense. Indeed, if he had acted in a genuinely valiant manner, he could have come off looking great from those questions. But by side-stepping them, he just looks like a wannabe. Doctors amputate Turkish earthquake survivor's arm [This story contains video]
But if you password-protected your MP3 files, the RIAA/judge/etc. would never learn of their existence. If they did, it would probably (although, I suppose, not quite axiomatically...) indicate less-than-totally-secure web design, which would make you liable anyway. I guess a snooping sysadmin would be an exception, but limited obfuscation would prevent the obvious 'find / -name "*.mp3"'. In any event, your liability would probably correspond closely with your competence in security/obscurity. This seems fair to me. Doctors amputate Turkish earthquake survivor's arm [This story contains video]
What CS degrees do you have? For how many years have you worked as an administrator/programmer? Under whom did you work, and can they vouch for your expertise? Doctors amputate Turkish earthquake survivor's arm [This story contains video]
Remember, any decent page layout involves a maximum of info-stimulation. Perhaps Tux could be surfing...with a skateboard...on the moon! And fisting a goat! No, wait, scratch that. Spiking a football! Yeah, extreme! And the words "Open Source In-your-face POSIX-conformance 2DX-TREME!!!!!" should be written somewhere.
But for god's sake, whatever you do, don't go with a simple, tasteful design that subtly conveys professionalism. Doctors amputate Turkish earthquake survivor's arm [This story contains video]
A long-standing problem with HTML is its non-turing-completeness. By not allowing extensions to HTML to be written in HTML (as they are in TeX), language extensions must be written in hacked-on plugin languages like Java and ActiveX (that are often proprietary), or must be written into the HTML interpreter itself. Is XML turing-complete? If so, that could resolve many of these problems. Doctors amputate Turkish earthquake survivor's arm [This story contains video]
The term "operating system" is not a rigorous one. There is no clear definition of what does and does not constitute an OS. Is the shell part of the OS? Is the GUI part of the OS? How about networking code -- both code bundled originally with the OS and, say, networking code bought from another vendor? Certainly the kernel is part of the operating system, yet the FSF plans to call their system "GNU/HURD", totally ignoring the Mach kernel at the center of the Hurd (almost as if the kernel weren't really the OS, but the things surrounding it were).
If an operating system is code designed to load programs and provide mechanisms for things like task-switching, file management, and power cycling, then even Windows 3.1 is an OS. It provided (non-preemptive) task-switching, a mechanism for starting and stopping programs (double-clicking the icon), file management, and cycling power.
Not to get too cosmic, but I'm reminded of what Tanenbaum once wrote, that there really is no disctinction between software and hardware. Similarly, operating system layers (like Windows 3.1, BASH, or the DOS shell itself) become the operating system. That's all an operating system every really is: An abstraction layer.
(None of this is meant to imply that MS wasn't being predatory with IE-bundling in Win98. Putting something as unintuitive as a web browser interface on a file manager is a totally monopolistic move, regardless of how "operating system" is defined.) Doctors amputate Turkish earthquake survivor's arm [This story contains video]
This "slowest machine that actually does reboot" is an NP-complete problem, given that the space (memory/hard drive) that the NT box has to work with is finite. It trivially reduces (I assume in polynomial time) to an NP-complete problem -- for a turing machine with N transitions (or instructions, whatever they're called, my automata lingo is rusty), find the longest possible finite sequence of '1's that can be printed out (with a '0' and '1' alphabet) before halting.
(And now, I'll see massive holes ripped in this postulate. Bring it on!)
"So, what do you want to hack for, Bickle?" "I can't sleep nights." "They got porno theaters for that."
This is a little off-topic, but since he was talking about the difficulty of installing things like apache and php under NT, I feel justified.
To what extent can the GNU tools be wrapped around NT? If the POSIX.1 layer of NT is still intact (maybe), then couldn't the other layers of POSIX be built on top of that? I know that a version of Bash has been ported to NT (but then it's been ported to DOS as well, and since I doubt that DOS could support things like coprocesses, I'm guessing the ported version isn't identical to the UNIX version.)
(I suppose I should finally learn the specifications for the Debian standard base. Could the NT files be moved around to make space for, say,/usr/local and still work? Hmmm...to get devices to show up as files, you'd probably need kernel support... I'm showing my ignorace of kernel internals here...HURD moves device stuff out of kernel space into daemon space...)
But in any event, a (perhaps limited) collection of the GNU tools for NT could make many a windows-imprisoned administrator's life easier. Does such a thing exist?
"So, what do you want to hack for, Bickle?" "I can't sleep nights." "They got porno theaters for that."
Slashdot Mirror
You need the ability to make additional phone charges,
the way 900 numbers can, on home phone numbers. You
could have a pre-ring recording warn callers that,
at your discretion, you can charge them some arbitrary
amount of money. (Sure, you could zap non-telemarketers
too, but you'd just be punishing yourself by scaring
away your friends.)
-------
First you'd need to express the function "used
to work for" in the binary boolean calculus,
and then you'd need to express that ^2 function
in the same calculus. Actually, ^2 and the
boolean algebra already share common formal
systems (Pascal, for example, or an extended
lambda calculus), so I suppose the real weak link here
is reducing "used to work for" to some formal
system. Best of luck!
-------
Is HIV intelligent? Was the bubonic plague sentient? Could ice-nine, if it were real, pass the Turing test?
For it to be a menace it would have to be moderately intelligent as well.
No, it most certainly wouldn't. Joy's point was that, with nanotechnology, the whole world becomes like a computer system. Ever read virus source code? It's not even remotely interesting from an AI point of view. The ONLY requirement for destructive code is that it properly exploits weaknesses in the security of a system. Really and truly, honest to god.
-------
Self-replicating nanotech doesn't need any "AI agenda" to cause mass destruction. Like any computer virus, or like Vonnegut's 'ice-nine', the very act of replication can be devastating.
-------
Your point about bank robbery was interesting -- how does one distinguish performance art from crime? The key factor is that "speech" (the subset of all actions) exists on an abstract level; speech is the manipulation of an abstract world sandboxed from the "real world." So we have can have crime in speech (pretending to kill someone) until it affects things in life (actually killing someone, and filming it).
I think issues about pornography (rape, kiddie porn, etc.), violence (no "Mewtwo guns down his classmates" Pokemon episode), drug use, etc. in speech fall back to this. You have unlimited freedom in speech, as long as it's sufficiently distanced from the real world.
This is the rule of thumb implied by "source code as free speech." Of course there's really no such thing as "source code", but perhaps the idea is that some layer of abstraction (even if it is recursively enumerable) be put between the average computer user and the average computer.
-------
The whole idea of "source code" is an historical accident, stemming from the way Von Neumann architectures are built. For example, some folks have built LISP machines -- chips that execute LISP code directly in hardware. Is this compiled code, or source code? I could take that same LISP code and compile it to x86 machine language. What is that LISP code now? Is it source code, or compiled code?
All sufficently general formal languages are identical in power. Whether a formal statement is written in binary boolean calculus, or that same boolean statement in an electrical diagram, or that same statement sculpted out of transistors, is irrelevant.
-------
While infrastructure -- the invisible technological foundations -- should be standardized for interoperability's sake, homogenized interfaces are bad all-around. This is because the interfaces that an individual works most efficiently with vary from person to person. A novice will set up TCP/IP on one machine most efficiently with a hand-holding gui, while an expert will set TCP/IP up most efficiently (and possibly on a large number of machines, quickly) with text files and scripts. Same infrastructure, different interfaces.
Why can't you keep all the physics (or whatever; the data and computation needed to do worthwhile cheating) on the server and put only bare interface stuff on the client? Would this be a much higher data transfer requirement; like would the graphics be a lot harder to compute? That seems like a viable solution.
... that compute that function. (No, X is not recursively enumerable, but there exist subsets of X that are.) Write a program Y that generates some (obscure) binary Xi that computes F each time it is run (where i is some positive integer). Release the source to Y. When a client logs in, have her download the binary for Xi. You can verify that this is a legitimate binary. (I think yacc or lex dealt with similar issues.) You wouldn't even need a sandbox, necessarily, although that might improve portability.
If this is too much of a bandwidth hog, make some computational sandbox for the server to download cheat-worthy computation to.
For every function F, there's an inifinite set X of programs X1, X2,
You might not even need for X to be inifinite, although someone could take an Xi and make a Xj with the same hash value; if X were infinite you (in theory) could discard used Xi's.
What am I missing here?
Remember all those great RSA-in-perl decryption gimmics that were (and may still be) floating around? Someone needs to condense DeCSS to few enough lines to print on trinkets. I want my dirty underwear to violate the MPAA's intellectual property "rights", dammit!
If the assembly language is interpreted, can't it
(slowly) appear to be a true 128-bit processor?
Indeed, 128-bits is the processor's I/O size; in theory the
processor width can be anything. (Although
this doesn't take into account the associated
speed hit of emulating a width higher than the
one used by the native language. Of course, none
of that emulation-speed-hit stuff has been discussed
in sufficient detail by Transmeta, nor probably will
be until the real assembly language is reverse engineered.)
Opening up the source code is an offtopic issue.
Finkployd asserted that moderation of articles
wouldn't work because, in his opinion, the
low-brow Jerry-Springer-loving masses don't have
the brains to properly evaluate the articles'
merits. (I replied that the comment moderation
system relies on the wisdom of those low-brow
masses, and that ironically the trailer trash
had ranked M{s|r}. Ployd's comments highly.) This
has nothing to do with the release of source code,
as you discuss, or with whether I should "go off
and start my own site", as Fink discusses. The
issue at hand is whether the Great Unwashed are
mentally fit to rank the articles themselves; by
their ability to properly rank replies to the
articles, they have demonstrated this mental
fitness.
.
(Score:4)
Gee, Finkployd, those faceless nabobs seem to like what you write. Appealing to the lowest common denominator, were we?
The current system of moderation and metamoderation encourages thoughtful posting and ranking. There is no reason (your elitist fear of the Great Unwashed aside) to think that the same could not be applied to story postings.
Recently, I've read several articles that fondly recall the days of mainframe programs, which tended to be less buggy and more open than programs of today. I recall from documentaries that Jobs had a strong disdain for mainframe programmers, viewing them as uptight and uncreative. Today, MacOS developers are hard at work retrofitting the existing (IMHO, feature-bloated) MacOS to BSD; an open, unsexy operating system built by mainframe programmers. How do you think Jobs' manager-as-cowboy style of software development management has affected the overall computer industry, if at all? And what are some ways that the current preference of glitz over substance in software can be changed? (Yes, this question is framed in a biased way -- that stems from my fond memories of the solid, flexible design of the ][.)
--J. Random Plantation Owner, 1855
1) People are totally grossed out by the idea of "tongue mice." It evokes the image of licking a mouse, not the cleanest thing around. Also, people I've talked to expressed fear that these devices would become like retainers, reeking and covered with old spit. The portion of these devices put into one's mouth MUST be replacable and disposable. No commercial success of such a device would be otherwise possible. (Also, a name other than "tongue mouse" would be needed -- I think "hook" would be a good name; it would evoke the idea of a hookah or fish hook, not of a retainer.) But the main thing is to eliminate fear of putting a nasty, public thing in one's mouth; making the intra-oral parts of these things disposable would do that.
2) Rather than a touchpad, I think a good design would be a small joystick with a single button, the approximate size and shape of a cigarette holder. Cigarette holders are gripped between the teeth; a (roughly) ball-shaped button could be put here for bite-clicking. The actual joystick part could be put on the end, where the smoke would come out of the cigarette holder, as it were. It could be the approximate size and shape of a clitoris (:>), large enough to get one's tongue around, small enough not to get in the way. The intra-oral part probably could be designed cheaply as a disposable tip, as not much of the device would actually go in the mouth.
All in all, I think it's pretty marketable if done correctly. People tend to forget how cool smoking is -- you can bite contemplatively on a cigarette, you can use it to gesture, etc. -- and this would be a non-eating substitute for that oral fixation. Plus it would totally eliminate RSI from pointing.
I've searched around, but no one is doing this as a mass-consumption product. (Several vendors have feet mice.) I found no relevant patents on the IBM patent search.
Doctors amputate Turkish earthquake survivor's arm [This story contains video]
I recall this issue having come up in Phrack, in essays on "mutating" code. A way to make viruses "mutate," it was argued, would be to keep the main virus instructions "encrypted" (obfuscated, really), and wrap encryption/decryption code around that (usually this was very small XOR "encryption", not very large code at all). The problem was that a virus scanner could check for this encryption code and thus detect the virus. The same dillema would exist with bootstrapping code.
Doctors amputate Turkish earthquake survivor's arm [This story contains video]
These "Why did you..." questions are of the same breed. "Here are some of the bad things said about you, now address them and make them go away." If JP *had* acted in a defensible way, he should have offered a good defense. Indeed, if he had acted in a genuinely valiant manner, he could have come off looking great from those questions. But by side-stepping them, he just looks like a wannabe.
Doctors amputate Turkish earthquake survivor's arm [This story contains video]
But if you password-protected your MP3 files, the RIAA/judge/etc. would never learn of their existence. If they did, it would probably (although, I suppose, not quite axiomatically...) indicate less-than-totally-secure web design, which would make you liable anyway. I guess a snooping sysadmin would be an exception, but limited obfuscation would prevent the obvious 'find / -name "*.mp3"'. In any event, your liability would probably correspond closely with your competence in security/obscurity. This seems fair to me.
Doctors amputate Turkish earthquake survivor's arm [This story contains video]
What CS degrees do you have? For how many years have you worked as an administrator/programmer? Under whom did you work, and can they vouch for your expertise?
Doctors amputate Turkish earthquake survivor's arm [This story contains video]
But for god's sake, whatever you do, don't go with a simple, tasteful design that subtly conveys professionalism.
Doctors amputate Turkish earthquake survivor's arm [This story contains video]
A long-standing problem with HTML is its non-turing-completeness. By not allowing extensions to HTML to be written in HTML (as they are in TeX), language extensions must be written in hacked-on plugin languages like Java and ActiveX (that are often proprietary), or must be written into the HTML interpreter itself. Is XML turing-complete? If so, that could resolve many of these problems.
Doctors amputate Turkish earthquake survivor's arm [This story contains video]
If an operating system is code designed to load programs and provide mechanisms for things like task-switching, file management, and power cycling, then even Windows 3.1 is an OS. It provided (non-preemptive) task-switching, a mechanism for starting and stopping programs (double-clicking the icon), file management, and cycling power.
Not to get too cosmic, but I'm reminded of what Tanenbaum once wrote, that there really is no disctinction between software and hardware. Similarly, operating system layers (like Windows 3.1, BASH, or the DOS shell itself) become the operating system. That's all an operating system every really is: An abstraction layer.
(None of this is meant to imply that MS wasn't being predatory with IE-bundling in Win98. Putting something as unintuitive as a web browser interface on a file manager is a totally monopolistic move, regardless of how "operating system" is defined.)
Doctors amputate Turkish earthquake survivor's arm [This story contains video]
(And now, I'll see massive holes ripped in this postulate. Bring it on!)
"So, what do you want to hack for, Bickle?"
"I can't sleep nights."
"They got porno theaters for that."
To what extent can the GNU tools be wrapped around NT? If the POSIX.1 layer of NT is still intact (maybe), then couldn't the other layers of POSIX be built on top of that? I know that a version of Bash has been ported to NT (but then it's been ported to DOS as well, and since I doubt that DOS could support things like coprocesses, I'm guessing the ported version isn't identical to the UNIX version.)
(I suppose I should finally learn the specifications for the Debian standard base. Could the NT files be moved around to make space for, say, /usr/local and still work? Hmmm...to get devices to show up as files, you'd probably need kernel support... I'm showing my ignorace of kernel internals here...HURD moves device stuff out of kernel space into daemon space...)
But in any event, a (perhaps limited) collection of the GNU tools for NT could make many a windows-imprisoned administrator's life easier. Does such a thing exist?
"So, what do you want to hack for, Bickle?"
"I can't sleep nights."
"They got porno theaters for that."