All it's going to take to kick the iPhone squarely in the balls is for someone to make a very sleak Android-based phone that has no developer restrictions on it.
It had better be VERY sleek...
Don't discount how well imlemented the iPhone is. I don't own one, but I tried one in the apple store, and the touch screen and user interface is amazingly well done. It feels totally natural to use. An iPhone competitor can't just copy the iPhone features (incl. multi-touch) - it needs to copy the slickness and attention to detail of implementation (think how different a crappy ATM touch screen is from a well done one).
Like it or not, the iPhone does also have that Apple cachet. The brand advantage creates an unfair playing field - an iPhone competitor needs to be significantly better or cheaper than iPhone to compete with it.
Office is certainly a cash cow, but the the document format lock-in that keeps it so is disappearing. Things like OpenOffice have pretty good interopability and Microsoft seem to be getting increasingly forced to open up their standards.
Don't forget that Google is also sticking it to them on this front. For 95% of home users Google Docs (supports MS.doc,.ppt,.xls formats) is all you need. I guess it's karma from killing Netscape that is coming back to Microsoft.
If it makes sense for Microsoft commercially then bully for them, but for anyone wanting to develop for Windows (or Linux+Windows) using free software there's already plently of choices:
MinGW -- minimal GNU toolchain for Windows (incl g++/gcc) Qt (not just GUI, but also threads, networking,...) Cygwin -- Linux API DLL for porting Linux apps to Windows (I prefer to just use cross-platform libraries) Cygwin/X -- X windows libs, server, utilities for Windows (presumably can be used with GTK, although I havn't tried it) pthreads_win32 Boost.Theads - portable threads library (if pthreads, or QThread doesn't fit your needs)
Using MinGW and Qt it's easy to develop apps that can just be recompiled across Linus/Solaris/Windows/Mac.
What's currently missing is a strong cross-platform multimedia toolkit, but that will be changing with KDE's Phonon framework which will be supported by Qt 4.4. Phonon will support Linux, Windows and Mac via (respectively) GStreamer, DirectShow and Quicktime backends.
I'm not sure if GNOME has any equivalent in the works, or to what degree Phonon is tied to Qt. Having alternate desktops and even GUI libraries is not a bad thing, but it'd be helpful if Linux could standardize on core stuff like multimedia support rather than having competing camps there too.
Try reading the article... The judge overturned the jury verdict because the plaintiffs lawyers had deliberately mislead them by deliberately ignoring his instructions.
If you're such a fan of trial by jury, then you should indeed be thankful that there are judges who slap down laywers who try to make an end run around the system!
Microsoft already has a high-powered research center (in fact more than one).
If you want to pick a research center to strive after, Xerox PARC might be the worst! AT&T Bell Labs or IBM Watson Labs might be better examples. Xerox PARC, while chock full of great people coming up with great innovations also had a reputation as a place where ideas never turned into commercial products. e.g. Xerox invented the modern GUI (from the bitmapped display and mouse on up), but it took another company - Apple - to actually commercialise it.
2) Where the benefit of splitting the task is greater than the disadvantages of doing so.
In the case of software projects, there are severe constraints on both of these. If a project has sufficiently large susbtasks, with sufficiently thin interfaces between them, then doing these in parallel may give a net gain. However there is a limit to this, and putting more people on a project or subtask past a certain point will in fact slow the project down (essentially due to communications, and miscommunications) rather than speed it up. A two man job is best, and faster, done by two men rather than ten.
I'm sure Microsoft is more interested in buying the customer/client base, both for Yahoo!'s web properties as well as advertising services, as well as eliminating a competitor in these areas so they can concentrate the fight on Google.
There may be some search expertise in Yahoo they can use, but really I doubt Microsoft is lacking in software talent, and I'm sure Microsoft research is more than up to the task of providing any necessary technology. The reason Microsoft is falling behind Google is surely because they are not so nimble (although I wonder how long Google can keep it up, if indeed they still are, given their crazy growth rate). Microsoft have become a giant slow moving behemoth, and apparently have horrible software management practices. The years of delay and scaled back feature set of Vista says it all. Adding masses more Yahoo! software engineers and managers to the mix is not the solution. Microsoft need to totally rethink the way they manage software projects - cut the burocracy and layers of management and inter-team back biting and get back to start-up type get-it-done environment.
IMO, the spin that this is about aquiring great technology is presumably because that sounds better than saying they're trying to remove a competitor and remove user choice - FORCINC people to become Microsoft customers.
Well, that's OK since Newtons law/fact of gravity is in fact wrong, and has been replaced by Einsteins theory of relativity!
But anyways...
The problem is that evolution most certainly is NOT a scientific theory!
Darwin's hypothesis that there was an hereditory mechanism for individual traits could be considered a theory, but certainly not any more since DNA has been discovered is certainly a fact.
Since evolutionary fitness is defined as the capability to preferentially survive and leave numerous offspring in a given environment, "survival of he fittest" is merely a shothand tautology, as is evolution towards greater fitness, given genetic coding of traits. Nothing remotely theoretical about it.
Speciation is nothing more than evolution of (geographically, culturally, or whatever) seperated subgroups of an existing species to the point that their DNA has diverged sufficient that they can no longer interbreed. By definition if you can't interbreed you are a new species, and lack of interbreeding means seperate evolutionary paths from that initial no-going-back forking point. Whoopee, new species, and the tree of life.
Darwin's "Origin of Species" is therefore, given the existence of DNA, most certainly a fact.
I think most people don't really have a clue what the "theory of evolution" is anyway. Darwin never used such a title, and I think in the popular imagination this is equated with a scientific origin of life in the first place, which is a issue. Sure evolution could have given rise to life, and almost certainly did, but speciiation is bound to ensue regardless of where DNA based life came from in the first place.
You wouldn't get that much energy from coming downstairs on that weight-lifting platform.
Say you want to run a 5W LED light (from which you may get the 600 lumen output of a 40W incandescent bulb)... For each hour you want to run the light, you'd need 3600 * 5 = 18,000 Joules of energy.
Now, with a 100Kg person and 3m drop from upstairs to downstairs, and g approximated to 10m/s/s, each trip downstairs would generate 100 * 3 * 10 = 3,000 Joules, so you'd need to make 18,000 / 3,000 = 6 trips downstairs to run that dim lamp for an hour.
OTOH hooking your exercise bike up to a generator and battery charger might be a slightly better bet. A person on a bike generates about 100-200W, so 30 min on the bike would keep that same 5W LED burning for 10-20hr.
So that 0.01W LED might give you a whopping 1.3 Lumens light output, rather than the claimed 600-800. To improve that by the needed factor of 500 you'd have to increase the weightxheight by that amount. A 2000' high lamp or 2.5 ton weight (about 20 cu ft of Gold) should do the trick.
I think the "transcend the himan condition" comment may have been inspired by Roger Penrose's book "The Emperors New Mind", or a reaction to it. Penrose wants to believe that computers arn't capable of human thought, and supports this by claiming (against all evidence to the contrary) that the human brain and conciousness work at a quantum level rather than in the realm of classical physics and neuro transmitters. Quantum computers would at least remove this source of "incapable of human level thought & consiousness" quackery, and in fact given that the human brain almost certainly *does* operate as the neural network it appears to be, quantum computers may indeed be theoretically capable of beyond-human computation (not just the speed advantage of conventional computers).
There is a lot of talk about computers surpassing, or not surpassing, humans at various tasks - does it not bother anyone that computers don't actually posses any intelligence? By any definition of intelligence you'd like? Every problem that a computer can "solve" is in reality solved by a human using that computer as a tool.
Did humans design our own DNA and brain architecture?
Why are you willing to attribute to a human the intelligence exhibited by our evolutionarily designed wetware brains, when you wouldn't afford the same to a computer exhibited by its human designed silicon brain?
I think more fundamentaly he doesn't realize that thread != core (i.e. that a three, or 42 for that matter, thread program can run on an N-core CPU, where N = 1, 2, 3, 4,....). Number of cores just means how much genuine parallelism may be occuring at run time.
His claim thay threads are useful in powers of two is of course complete junk since threads are usually used one at a time for specific tasks (data aquisition thread, rendering thread, etc), or in groups (maybe of run-time configurable size) to provide thread pools for specific tasks - e.g. server threads.
Let's not forget also that the OS itself will be competing with whatever application(s) you are running for the CPU, so even a single single-threaded program will benefit from a multi-core CPU by way of not having to compete with the OS as much for the CPU cores.
It's now Jan. 28 - D Day. Heins has modified his test so the effects observed are difficult to deny. He holds a permanent magnet a few centimetres away from the driveshaft of an electric motor, and the magnetic field it creates causes the motor to accelerate. It went well.
Contacted by phone a few hours after the test, [MIT professor] Zahn is genuinely stumped - and surprised. He said the magnet shouldn't cause acceleration. "It's an unusual phenomena I wouldn't have predicted in advance. But I saw it. It's real. Now I'm just trying to figure it out."
There's no talk of perpetual motion. No whisper of broken scientific laws or free energy. Zahn would never go there - at least not yet. But he does see the potential for making electric motors more efficient, and this itself is no small feat.
"To my mind this is unexpected and new, and it's worth exploring all the possible advantages once you're convinced it's a real effect," he added. "There are an infinite number of induction machines in people's homes and everywhere around the world. If you could make them more efficient, cumulatively, it could make a big difference."
Seems like we have a respected scientist confronted with an unexpected phenomena that he wants to understand. What part of that makes him a jackass? He's not saying it's a perpetual motion machine - he's saying it's maybe an (unexpected) way to make a more efficient induction motor.
The point isn't that Stallman wrote all of GNU himself, so I'm not sure what your point is. The fact that GCC was forked into ECGS then readopted as the offical GNU C compiler is testatment to the power of the free software model that Stallman created. Stallman is a decent hacker but his real claim to fame is creating the free software project and GNU project and being it's driving force over the years, notwithstanding blow-hards like Eric Raymond trying get credit for it.
I wouldn't characterize Linus as a brilliant programmer. A brilliant software manager perhaps, but no more than a strong programmer. Most of the Linux kernel has been written be people other than Linux, and the Linux operating system owes 1000% more to Stallman as a driving force than Linus.
Stallman's FSF and GNU project have got nothing to do with BSD. The BSD licence originated with the BSD Unix distribution and has since, in modified form, taken on a life of it's own as a permissive type of free software licence.
Stallman's goal with the FSF was originally to create an entirely free Unix replacement - the GNU (= GNU's Not Unix) project, starting from the ground up with tools like Emacs, GCC, the GNU C library, Bison (Yak replacement), replace ments for all the Unix user space tools, etc.
Stallman didn't hijack a BSD initiated free softare movement - he created the movement in the first place and created an entirely free Unix implementation (minus the kernel - intended to be HURD) that is an alternative to BSD, Sys V, etc. He's quite right to assert that "Linux" should be called GNU/Linux, since a Linux distribution is essentially GNU with Linus's kernel. You can even have a GNU system with another kernl (such as Mach), but without GNU (and Richard Stallman) "Linux" would not exist - you'd just have some sad hacker in Finland with a toy kernel and maybe dreams of building an operating system and user space tools around it one day.
If an infant hears the word cat more in the presence of a cat than a bird, and hears the word apple more in the presence of an apple than a cat, then the association cat-word == cat-thing is going to be stonger than cat-word == bird-thing, apple-word == cat-thing.
i.e. (A, B) => assoc(A, B)++ ---- The ONE MILLION DOLLAR formula!
It doesn't make any difference how many words / objects or other sensory inputs are presented together - just strengthen the associations of all co-present stimuli (this is simply how the brain works - no thought required), and those that occur most frequently will naturally end up with the strongest associations. Brains presumably evolved this way for cause-effect association (i.e. environment prediction), but it works just as well for associating words with objects etc.
This is simple strenthening of associations by repeated exposure (of course many other mechanisms such as focus also come into play), although if you can get a $1M grant by calling it data mining then I guess more power to you.
Anyone who has seen a a regular cat (1-2ft with tail, 10-15lb) jump up a 5-6ft wall should realize that a 10ft tiger weighing 350lb certainly has a fighting chance of jumping up a 12ft wall. In fact I'm sure they can do much more than than. If it can clean the thing ballistically as this story notes, then one can only imagine what they can do by "running up the wall" normal cat-style, using their paw-pads to grip the wall and gain additional upward momentum through muscle power.
Only in America would someone whine "but no-one told me it was unsafe to be stupid, boo-hoo".
These morons got precisely what they might expect to have coming to them, regardless of whether the zoo should have built a higher wall. The only real tragedy is that the innocent tiger - en endangered species - was killed.
The kernel is the lowest level software that interacts directly with the hardware (processor, memory controller, etc) and provides a more abstract virtual machine in terms of processes, threads, memory allocation, inter-process communication. A micro kernel would essentially stop there and let things like device drivers and file systems be implemented at user level, whereas a monolithic kernel would include stuff like that as part of the kernel itself.
Of the stuff you list there (just going on the names, without knowing specifically what these are in the HP world), it's a mix of user-space stuff (strXXX C-library, daemons), possibly kernel (swapper may be part of memory subsystem), and some stuff that (ttisr = teletype interrupt service routine?, lvmkd = logical volume make device?) that could be considered as either depending on the kernel design... in this case kernel given that it's a Unix box.
Slashdot Mirror
All it's going to take to kick the iPhone squarely in the balls is for someone to make a very sleak Android-based phone that has no developer restrictions on it.
It had better be VERY sleek...
Don't discount how well imlemented the iPhone is. I don't own one, but I tried one in the apple store, and the touch screen and user interface is amazingly well done. It feels totally natural to use. An iPhone competitor can't just copy the iPhone features (incl. multi-touch) - it needs to copy the slickness and attention to detail of implementation (think how different a crappy ATM touch screen is from a well done one).
Like it or not, the iPhone does also have that Apple cachet. The brand advantage creates an unfair playing field - an iPhone competitor needs to be significantly better or cheaper than iPhone to compete with it.
That's all I see this gimmic as, is just pushing the heat around.
That's all any heatsink is doing!
Office is certainly a cash cow, but the the document format lock-in that keeps it so is disappearing. Things like OpenOffice have pretty good interopability and Microsoft seem to be getting increasingly forced to open up their standards.
.doc, .ppt, .xls formats) is all you need. I guess it's karma from killing Netscape that is coming back to Microsoft.
Don't forget that Google is also sticking it to them on this front. For 95% of home users Google Docs (supports MS
http://docs.google.com/
With these nanoparticle coatings, car owners could make their own hydrogen, either in their garage or even when driving
If you power the car from methane, then you don't need any nanoparticles - most drivers can make their own methane already.
The new "gas" stations would be Taco Bell drive thrus.
If it makes sense for Microsoft commercially then bully for them, but for anyone wanting to develop for Windows (or Linux+Windows) using free software there's already plently of choices:
...)
MinGW -- minimal GNU toolchain for Windows (incl g++/gcc)
Qt (not just GUI, but also threads, networking,
Cygwin -- Linux API DLL for porting Linux apps to Windows (I prefer to just use cross-platform libraries)
Cygwin/X -- X windows libs, server, utilities for Windows (presumably can be used with GTK, although I havn't tried it)
pthreads_win32
Boost.Theads - portable threads library (if pthreads, or QThread doesn't fit your needs)
Using MinGW and Qt it's easy to develop apps that can just be recompiled across Linus/Solaris/Windows/Mac.
What's currently missing is a strong cross-platform multimedia toolkit, but that will be changing with KDE's Phonon framework which will be supported by Qt 4.4. Phonon will support Linux, Windows and Mac via (respectively) GStreamer, DirectShow and Quicktime backends.
I'm not sure if GNOME has any equivalent in the works, or to what degree Phonon is tied to Qt. Having alternate desktops and even GUI libraries is not a bad thing, but it'd be helpful if Linux could standardize on core stuff like multimedia support rather than having competing camps there too.
Try reading the article... The judge overturned the jury verdict because the plaintiffs lawyers had deliberately mislead them by deliberately ignoring his instructions.
If you're such a fan of trial by jury, then you should indeed be thankful that there are judges who slap down laywers who try to make an end run around the system!
Microsoft already has a high-powered research center (in fact more than one).
If you want to pick a research center to strive after, Xerox PARC might be the worst! AT&T Bell Labs or IBM Watson Labs might be better examples. Xerox PARC, while chock full of great people coming up with great innovations also had a reputation as a place where ideas never turned into commercial products. e.g. Xerox invented the modern GUI (from the bitmapped display and mouse on up), but it took another company - Apple - to actually commercialise it.
That only applies for tasks that are:
1) Further parallelizable, and
2) Where the benefit of splitting the task is greater than the disadvantages of doing so.
In the case of software projects, there are severe constraints on both of these. If a project has sufficiently large susbtasks, with sufficiently thin interfaces between them, then doing these in parallel may give a net gain. However there is a limit to this, and putting more people on a project or subtask past a certain point will in fact slow the project down (essentially due to communications, and miscommunications) rather than speed it up. A two man job is best, and faster, done by two men rather than ten.
I'm sure Microsoft is more interested in buying the customer/client base, both for Yahoo!'s web properties as well as advertising services, as well as eliminating a competitor in these areas so they can concentrate the fight on Google.
There may be some search expertise in Yahoo they can use, but really I doubt Microsoft is lacking in software talent, and I'm sure Microsoft research is more than up to the task of providing any necessary technology. The reason Microsoft is falling behind Google is surely because they are not so nimble (although I wonder how long Google can keep it up, if indeed they still are, given their crazy growth rate). Microsoft have become a giant slow moving behemoth, and apparently have horrible software management practices. The years of delay and scaled back feature set of Vista says it all. Adding masses more Yahoo! software engineers and managers to the mix is not the solution. Microsoft need to totally rethink the way they manage software projects - cut the burocracy and layers of management and inter-team back biting and get back to start-up type get-it-done environment.
IMO, the spin that this is about aquiring great technology is presumably because that sounds better than saying they're trying to remove a competitor and remove user choice - FORCINC people to become Microsoft customers.
Well, that's OK since Newtons law/fact of gravity is in fact wrong, and has been replaced by Einsteins theory of relativity!
But anyways...
The problem is that evolution most certainly is NOT a scientific theory!
Darwin's hypothesis that there was an hereditory mechanism for individual traits could be considered a theory, but certainly not any more since DNA has been discovered is certainly a fact.
Since evolutionary fitness is defined as the capability to preferentially survive and leave numerous offspring in a given environment, "survival of he fittest" is merely a shothand tautology, as is evolution towards greater fitness, given genetic coding of traits. Nothing remotely theoretical about it.
Speciation is nothing more than evolution of (geographically, culturally, or whatever) seperated subgroups of an existing species to the point that their DNA has diverged sufficient that they can no longer interbreed. By definition if you can't interbreed you are a new species, and lack of interbreeding means seperate evolutionary paths from that initial no-going-back forking point. Whoopee, new species, and the tree of life.
Darwin's "Origin of Species" is therefore, given the existence of DNA, most certainly a fact.
I think most people don't really have a clue what the "theory of evolution" is anyway. Darwin never used such a title, and I think in the popular imagination this is equated with a scientific origin of life in the first place, which is a issue. Sure evolution could have given rise to life, and almost certainly did, but speciiation is bound to ensue regardless of where DNA based life came from in the first place.
You wouldn't get that much energy from coming downstairs on that weight-lifting platform.
Say you want to run a 5W LED light (from which you may get the 600 lumen output of a 40W incandescent bulb)... For each hour you want to run the light, you'd need 3600 * 5 = 18,000 Joules of energy.
Now, with a 100Kg person and 3m drop from upstairs to downstairs, and g approximated to 10m/s/s,
each trip downstairs would generate 100 * 3 * 10 = 3,000 Joules, so you'd need to make 18,000 / 3,000 = 6 trips downstairs to run that dim lamp for an hour.
OTOH hooking your exercise bike up to a generator and battery charger might be a slightly better bet. A person on a bike generates about 100-200W, so 30 min on the bike would keep that same 5W LED burning for 10-20hr.
The lamp is 48" high = 1.219m, and the weight is 50lb = 22.679Kg
Potential energy = weight * height * g = 22.679 * 1.219 * 9.8 = 270.9J
Expeding that 270.9 Joules over the claimed 4 hr runtime, gives a continuous power output of 270.9 / (4 * 3600) = 0.01W
According to this story the record for Lumens per Watt from a LED stood last year at 131 Lumens/Watt :
http://www.dvorak.org/blog/?p=5822
So that 0.01W LED might give you a whopping 1.3 Lumens light output, rather than the claimed 600-800. To improve that by the needed factor of 500 you'd have to increase the weightxheight by that amount. A 2000' high lamp or 2.5 ton weight (about 20 cu ft of Gold) should do the trick.
What am I missing?
Kinda retro... You'da thunk that they'd hook the the moon up with internet and just use VOIP.
Houston: Where are you? Why arn't you on IM?
Astronaut: I'm in ur base, dialing into ur network
I think the "transcend the himan condition" comment may have been inspired by Roger Penrose's book "The Emperors New Mind", or a reaction to it. Penrose wants to believe that computers arn't capable of human thought, and supports this by claiming (against all evidence to the contrary) that the human brain and conciousness work at a quantum level rather than in the realm of classical physics and neuro transmitters. Quantum computers would at least remove this source of "incapable of human level thought & consiousness" quackery, and in fact given that the human brain almost certainly *does* operate as the neural network it appears to be, quantum computers may indeed be theoretically capable of beyond-human computation (not just the speed advantage of conventional computers).
There is a lot of talk about computers surpassing, or not surpassing, humans at various tasks - does it not bother anyone that computers don't actually posses any intelligence? By any definition of intelligence you'd like? Every problem that a computer can "solve" is in reality solved by a human using that computer as a tool.
Did humans design our own DNA and brain architecture?
Why are you willing to attribute to a human the intelligence exhibited by our evolutionarily designed wetware brains, when you wouldn't afford the same to a computer exhibited by its human designed silicon brain?
I think more fundamentaly he doesn't realize that thread != core (i.e. that a three, or 42 for that matter, thread program can run on an N-core CPU, where N = 1, 2, 3, 4, ....). Number of cores just means how much genuine parallelism may be occuring at run time.
His claim thay threads are useful in powers of two is of course complete junk since threads are usually used one at a time for specific tasks (data aquisition thread, rendering thread, etc), or in groups (maybe of run-time configurable size) to provide thread pools for specific tasks - e.g. server threads.
Let's not forget also that the OS itself will be competing with whatever application(s) you are running for the CPU, so even a single single-threaded program will benefit from a multi-core CPU by way of not having to compete with the OS as much for the CPU cores.
An astronaut told me her biggest concern was a drop of hot sauce floating around and getting in her eye
I'd be more concerned about a fellow astronaut getting explosive diarrhea before he'd fully docked his backside with the space-crapper.
Naturally it'd be even worse if he'd just been eating hot sauce.
I think you're the jackass here.
Let's look at the FULL quote:
It's now Jan. 28 - D Day. Heins has modified his test so the effects observed are difficult to deny. He holds a permanent magnet a few centimetres away from the driveshaft of an electric motor, and the magnetic field it creates causes the motor to accelerate. It went well.
Contacted by phone a few hours after the test, [MIT professor] Zahn is genuinely stumped - and surprised. He said the magnet shouldn't cause acceleration. "It's an unusual phenomena I wouldn't have predicted in advance. But I saw it. It's real. Now I'm just trying to figure it out."
There's no talk of perpetual motion. No whisper of broken scientific laws or free energy. Zahn would never go there - at least not yet. But he does see the potential for making electric motors more efficient, and this itself is no small feat.
"To my mind this is unexpected and new, and it's worth exploring all the possible advantages once you're convinced it's a real effect," he added. "There are an infinite number of induction machines in people's homes and everywhere around the world. If you could make them more efficient, cumulatively, it could make a big difference."
Seems like we have a respected scientist confronted with an unexpected phenomena that he wants to understand. What part of that makes him a jackass? He's not saying it's a perpetual motion machine - he's saying it's maybe an (unexpected) way to make a more efficient induction motor.
The Network is the Computer
Same shit at Sun.
The point isn't that Stallman wrote all of GNU himself, so I'm not sure what your point is. The fact that GCC was forked into ECGS then readopted as the offical GNU C compiler is testatment to the power of the free software model that Stallman created. Stallman is a decent hacker but his real claim to fame is creating the free software project and GNU project and being it's driving force over the years, notwithstanding blow-hards like Eric Raymond trying get credit for it.
I wouldn't characterize Linus as a brilliant programmer. A brilliant software manager perhaps, but no more than a strong programmer. Most of the Linux kernel has been written be people other than Linux, and the Linux operating system owes 1000% more to Stallman as a driving force than Linus.
Stallman's FSF and GNU project have got nothing to do with BSD. The BSD licence originated with the BSD Unix distribution and has since, in modified form, taken on a life of it's own as a permissive type of free software licence.
Stallman's goal with the FSF was originally to create an entirely free Unix replacement - the GNU (= GNU's Not Unix) project, starting from the ground up with tools like Emacs, GCC, the GNU C library, Bison (Yak replacement), replace ments for all the Unix user space tools, etc.
Stallman didn't hijack a BSD initiated free softare movement - he created the movement in the first place and created an entirely free Unix implementation (minus the kernel - intended to be HURD) that is an alternative to BSD, Sys V, etc. He's quite right to assert that "Linux" should be called GNU/Linux, since a Linux distribution is essentially GNU with Linus's kernel. You can even have a GNU system with another kernl (such as Mach), but without GNU (and Richard Stallman) "Linux" would not exist - you'd just have some sad hacker in Finland with a toy kernel and maybe dreams of building an operating system and user space tools around it one day.
If an infant hears the word cat more in the presence of a cat than a bird, and hears the word apple more in the presence of an apple than a cat, then the association cat-word == cat-thing is going to be stonger than cat-word == bird-thing, apple-word == cat-thing.
i.e. (A, B) => assoc(A, B)++ ---- The ONE MILLION DOLLAR formula!
It doesn't make any difference how many words / objects or other sensory inputs are presented together - just strengthen the associations of all co-present stimuli (this is simply how the brain works - no thought required), and those that occur most frequently will naturally end up with the strongest associations. Brains presumably evolved this way for cause-effect association (i.e. environment prediction), but it works just as well for associating words with objects etc.
This is simple strenthening of associations by repeated exposure (of course many other mechanisms such as focus also come into play), although if you can get a $1M grant by calling it data mining then I guess more power to you.
I'll bet Dr.J could have done it.
http://www.youtube.com/watch?v=FPQUBY-HadY
Anyone who has seen a a regular cat (1-2ft with tail, 10-15lb) jump up a 5-6ft wall should realize that a 10ft tiger weighing 350lb certainly has a fighting chance of jumping up a 12ft wall. In fact I'm sure they can do much more than than. If it can clean the thing ballistically as this story notes, then one can only imagine what they can do by "running up the wall" normal cat-style, using their paw-pads to grip the wall and gain additional upward momentum through muscle power.
Only in America would someone whine "but no-one told me it was unsafe to be stupid, boo-hoo".
These morons got precisely what they might expect to have coming to them, regardless of whether the zoo should have built a higher wall. The only real tragedy is that the innocent tiger - en endangered species - was killed.
The kernel is the lowest level software that interacts directly with the hardware (processor, memory controller, etc) and provides a more abstract virtual machine in terms of processes, threads, memory allocation, inter-process communication. A micro kernel would essentially stop there and let things like device drivers and file systems be implemented at user level, whereas a monolithic kernel would include stuff like that as part of the kernel itself.
Of the stuff you list there (just going on the names, without knowing specifically what these are in the HP world), it's a mix of user-space stuff (strXXX C-library, daemons), possibly kernel (swapper may be part of memory subsystem), and some stuff that (ttisr = teletype interrupt service routine?, lvmkd = logical volume make device?) that could be considered as either depending on the kernel design... in this case kernel given that it's a Unix box.