The online rate is rather unreasonable.
I don't know why they do that since the compact dead tree version (with
magnifying glass) is more
reasonably priced (at least for what you get).
If memory serves it can be had for under $300.
Be sure to be specific when Fortran bashing. Fortran 77 should
justifiably elicit shudders and shouts, but
Fortran 90 and later are very nice for
scientific programming. In fact, modern
Fortran doesn't look a whole lot like Fortran 77.
If I had moderator points right now, I'd bump this up. Usenet reached a low for a few years there, but it has gotten better as more people have moved to the web.
Slashdot can be fun, but web based forums just don't work as well for extended discussions. Unless you're really fast, your chances of having several meaningful back-and-forth posts with someone here are negligible. By the time you've made a post and someone has responded to it, it's yesterdays Slashdot and you might as well forget about posting a followup. For what it does Usenet is still the best.
Unfortunately, matlab is slow (version 6 may have improved due to incorporation of LINPACK), and absolutely STUPID(!!) about memory issues.
The core routines (linear system solvers, etc.) in matlab were LINPACK (linear systems)
and EISPACK (eigenvalue routines)
from the beginning. Cleve Moler, who
wrote the original matlab, was one of the coauthors of LINPACK. I haven't followed it, but if there has been any change here, I suppose Matlab might have incorporated the more recent LAPACK. Either way Speed of the core computations shouldn't
be a big problem.
It's using Matlab as anything other than
a slick interface to Fortran code that causes
performance problems. Memory in particular is a real killer with Matlab. Unfortunately, it's inherent in the language. Matlab lacks pointers or any ability to pass matrices to functions by reference. The upshot of this is, instead of allowing you to overwrite a matrix, matlab has to allocate new space to store
the results of every computation. Loops aren't
terribly fast either, though many matlab programs
can work around this using the subarray notation. Octave suffers from similar problems.
Fixing these problems would turn Matlab into a real programming language, but it would cut down on its ease of use, readability, and interactivity. If you did this you would essentially get something that looked like Fortran 90 with some really nice numerical libraries. That's not the point of matlab, though I think it's a pretty good argument for Fortran 90 as what you might use for production code after experimenting/prototying in matlab.
Umm, aspirin has similar anti-inflammatory properties to ibuprofen (Advil). I believe that the painkiller in
Aleve is also an anti-inflammatory.
Yep. Naproxyn sodium (Aleve), ibuprofen and
aspirin are all anti-inflamatories and have
similar effects and side effects. Definitely good
first treatment for tendonitis. (Using naproxyn sodium for tendonitis in my shoulder at the moment).
Acetominaphen isn't an anti-inflamatory at all as far as I know.
Could be. Unfortunately I never dug deep enough into the 6502 to figure that out. I was just a teenager at the time farting around with assembly. Those sort of concerns would have gone right over my head. Having to clear carry bits and limitations on indirect addressing, etc. seemed like annoying quirks that couldn't (in my mind) have any reasonable justification. I loved the graphics chip set on my Atari, but I was jealous of fancier instructions on the Z80 and 6809.
With hindsight, I suspect a lot of 6502 quirks were design decisions. It seemed like a fairly tightly optimized processor. It never ran at a very high clock rate compared to the Z80, but comparatively it did pretty well at the lower clock rates. I should probably dig up info on how many clock cycles the Z80 and 6502 took to execute various instructions. My memory might be faulty here and this wouldn't be too hard to check.
I was in the Atari camp, but talk of 6502 assembly can put me on a real nostalgia trip. Did anyone else wonder why there wasn't an add without carry instruction? Nearly 5% of my code must have been clearing the carry bit (CLC). The quirks of this chip were burned on my brain during my teenage years.
I still have my Atari 800, though I haven't run it for years. I can't quite bring myself to get rid of it.
The custom chip set for the Amiga was
designed by Jay Miner who also did the graphics chip set for the Atari 8-bit machines. I don't
think he had come over to Commodore when the C64 was being designed, but I'm not sure and can't turn up anything on a web search. I also don't know if he is still alive. The Atari 800 and the Amiga were great machines in their day. If he is dead, I'll belatedly mourn his passing.
The recent resurgence in public debate about
nuclear power has been really hard to follow. Both
sides seem to disagree on the facts, so I can
only conclude that someone (possibly both sides) is bullshitting.
On the one hand, I recently read an editorial in the Washington Post in which a pro-nuke activist claimed that recent advances in reprocessing technology make it possible to
process old fuel, without producing weapons grade plutonium and leaving behind stuff that has a half-life measured in decades (considerably alleviating concerns about long-term storage).
On the other hand, I've never heard an anti-nuke type address this possibility. From their point of view reprocessing is always a proliferation risk and they never mention the possibility that reprocessing might help with the storage problem.
Then there are completely conflicting accounts on whether nuclear power is economically
reasonable. The anti-nukes claim that reactors have only worked because of subsidies. The pro-nukes claim that this is only because of lack of standardization in the industry and the burden of unreasonable opposition to building new plants.
Figuring out who is bullshitting is starting to look like a major research project to me. I'd love to find a source that didn't bring in a great deal of biased axe-grinding, but I'm pretty sure I haven't seen it yet.
I'll repeat what others have said: This problem has been solved for quite some time; governments just haven't required diesel engines to have proper emmissions controls. What is taking longer to deal with is getting sulfur out of diesel fuel, but even that won't be a problem soon. As others have pointed out diesel engines
can run on just about anything that will burn
with enough compression. This opens the possibility of all sorts of renewable fuel sources.
And diesel engines have one advantage that appeals to me at least: if they are cared for reasonably they last a hell of a long time compared to gasoline engines. Also, diesel
cars used to easier to work on for the home mechanic since they didn't previously have all
the computerized stuff. (This latter point has
probably changed with the new cleaner diesels.)
That is the standard contract. However there are a couple of ways around it. First,
authors are often publishing after they've released a departmental technical report that might differ only in minor ways from the journal
paper. No journal publisher is going to try to
prevent you from releasing an earlier technical
report on the web.
Also, I know people who cross out offensive
sections of the contract or write in the margins
that they reserve the right of web publication.
This generally works.
Finally, even if you sign it, the journals
just don't care that much about the odd article
on someone's homepage. In my field, numerical analysis, I've never heard of anyone getting
into trouble with a publisher over this sort of thing.
FWIW, I've got one of the bastards to sign
right now. I guess I will sign as-is so as not
to annoy my co-author. But it does piss me off.
Maybe that's the year in the movie; I don't remember. I had a copy of the book handy on the bookshelf next to the computer and it's definitely 1997 there. It's near the end where Dave lobotomizes Hal.
Also, the festivities in Urbana weren't in 1992. I would have been there. So they must have been in 1997.
To go off on a tangent, the first time I heard about the Hal shifted one letter = IBM was in 1981. I was a proto-geek of 13 with a brand new Atari 400, an assembler cartridge, a lovely tape drive for storing programs and a small B/W "monitor" on which I could also watch Dukes of Hazard during moments of rest. I was teaching myself 6502 assembly. I tried to start a computer club at school, but a
luddite teacher axed the idea and told me that computers were "dehumanizing" and "HAL+1=IBM. Think about it." Bastard.
Ah well. Even without the club I eventually learned to clear carry bits before starting an addition (what, no add without carry on a 6502!!?) and wrote some really unplayably crappy games. Those were fun days.
I suspect you know, even if you are getting
"corrected" by others, but perhaps others don't.
That would have been Hal's 4th B-day. He was born on Jan. 12 1997 in Urbana Illinois (the host town to the university where I spent 9 years...). There was something commemorating
that event at the time, though I had left for greener pastures and missed it.
Maybe it's still not too late to send a card:
Hal 9000, Production number 3, c/o Hal Plant, Urbana IL 61801.
"Dave, are you still there? Why didn't you remember my birthday?"
That is a good idea. However one legal entity would probably have to hold the copyright. This means you would have to sign over your copyright. To some extent I think this is what the FSF does: they encourage people to sign over their copyright so they can handle legal details. If you like the GPL, signing over copyright to the FSF seems like it might have some potential (but, alas, largely untested in court) advantages.
Regarding another person's response to your post (about whether multiple programs in a compilation can be registered at once): I'm not sure I see a problem. The quoted sections seem not to be targeted toward compilations, but people do manage to defend compilations under copyright law. The trick in complying with these rules seems to be deciding what printed media to provide. An energetic organization could provide 25+25 pages for each program in the compilation. (It's a lot, but not completely out of the question if you're not doing it by yourself on one printer). Copyright would be formally established and stealing a single program from a registered copyrighted compilation in which the same organization held copyright on each program could never be justified as fair use.
Hmmm... On the other hand, I have a Windows '98 "compilation" CD. I wonder if MS printed out the required pages from each program.
There is a long tradition of this sort of thing. My PhD project involved wasting lots of
time on Usenet. The only major problem with this was that it was not particularly original: most
other grad students were doing similar things.
The graduates of most CS departments are
not going to have the background for programming
for scientific applications. That's just a fact.
Math and numerical analysis courses used to be a
part of the CS curriculum long
ago, but they no longer are. Many CS departments don't like supporting this sort of research or courses and many academics
interested in scientific computing are now in
math or engineering departments.
Where does this leave you? You need to look around and find people with different (i.e. not straight CS) backgrounds. Many CS, math
and engineering departments are collaborating
on creating degree programs in scientific computing. This is often for graduate degrees, but not in all cases. Many math departments have an emphasis on scientific computing and computational mathematics. Many engineers become
quite good programmers either in the course of
getting a degree or on the job.
If you look around you will find people. I can assure you there are a lot of people interested in this sort of thing who are annoyed that most of the jobs seem to be in business related programming.
Not to point out the obvious, but
choosing not to prosecute people for copying
a page of a textbook etc. is something the
government can do without exercising any authority
whatsoever. You may disagree with fair use, but it's hardly an aggressive extension of government authority.
Your car example is not analogous. Someone
whose intellectual property rights are violated
is not denied the use of that property. So I would obviously be very unhappy if you took my car. However I am not ungenerous. I won't
let you have my car, but I will grant you
the right to quote selected portions of this post
if you wish to post a reply. Ain't fair use grand?
I wasn't able to get back on to the page to
check, but I don't remember a great level
of detail on the computations that are being
done. Non-linear control problems associated with
robotics aren't generally
thought of as the easiest thing in the world.
There's a lot of math behind the stabilization
of non-linear systems. It's possible that they
are using a fairly computationally intensive method. This would not be like slapping together a PID controller and tuning
the parameters until it can walk without falling
over.
Whether they really need 700MHz processors
in each leg, I can't say. But it's not a possibility
you could dismiss without knowing a fair bit
about non-linear control methods and what method
was used in particular in this case. In any
event these processors are going to be doing
a hell of a lot more than "telling" the legs
to move up and down regularly.
And it's actually readable if you have
a really solid math background. (This isn't
intended to be a vacuous comment: a lot of
mathematical papers are only readable by
specialists and not even by other mathematicians
working in other areas.)
IMO the really amazing gem in this paper
is the noisy coding theorem. Shannon proved
that if you have a communication channel
that introduces errors (i.e. a noisy channel)
then that channel always has a channel capacity.
If you send data at a rate lower than the
channel capacity then you can prove that there
are coding schemes that result in a average
probability of a bit error being as small as
you like. (However smaller error rates mean
more complicated codes). Unfortunately Shannon's
proof was nonconstructive. He essentially showed
that if you transmit the data in large blocks
and encode those blocks randomly you get
a probability of a bit error that goes to
zero as the block size increases. These codes
do not require ever increasing redundancy to
achieve this.
Unfortunately decoding
a random encoding applied to a huge block of
data is not computationally practical. A major
thread in information and coding theory has
been trying to find tractable codes that behave
somewhat like Shannon's codes. It's a hard
problem, but your modem works better because
of it.
Personally I like this result because
it was so counter-intuitive: who else would
have thought it might be possible, even
theoretically, to construct codes that make the
error rate approach zero without introducing
ever more redundancy into the code?
Shannon was a giant in engineering,
computer science and mathematics.
Next, the part back in the states where Hannibal is having a phone-off with Clarice at the fare. He always evades her view, even going so far as to touch her and still not get seen.
Some spoilers, although I don't think they
give enough detail to ruin anyone's enjoyment
of the movie:
I watched this bit (and the rest of the movie) in the theater
in union station (the basement floor). It was a bit surreal watching the place you are in on film. It got a good laugh from the audience.
For what it's worth, it wasn't as good as the first film, but I do think Moore was well cast and
did a good job and Hopkins was wonderful as usual. Unfortunately, it didn't have the psychological edge of the Silence of the Lambs. And the funny bits didn't mesh with the rest of the movie (I'm pretty sick, but even I couldn't
laugh at dinner table conversation over a plate
of brains at the end). OTOH, it did make me feel ill; I like gruesome horror films and I rarely find anything that either scares me or makes me feel like I have
to look away. This film at least pushed me to my limits on gore and it had some good acting.
I enjoyed it, but they really should have worked more on the screenplay and done something creepier instead of just going for something that was supposed to be disgusting and funny. Does every film have to have laughs in it? What's wrong
with just trying to scare us?
More like 18 million. And, Vegemite aside, a fair bit of good stuff comes from Australia. They also have a tradition of exporting their best and brightest to work in other countries (funding for universities really sucks under the current
Australian government---after living there for a few years, I just moved back because it's easier to pursue a career in the USA). If we could get figures for ex-pat Aussies working on projects
I suspect that everyone would be quite impressed for what Australians do with only 18 million people.
I respectfully disagree. Linux has only in the last year or two gotten to the point where I would start to recommend it to reasonably average Windows users. And I would still only recommend it to those who are frustrated with Windows' "quirks" (i.e. crashes) and who have fairly focused and specialized computing needs.
IMO the battle for the desktop hasn't even really begun yet.
I think there are a few forces that might well push Linux on to more desktops.
1. Linux is free. Windows is starting to look like a hefty
chunk of the price of a new computer. With profit margins where they are, equipment manufacturers have to be interested.
2. Linux is Free (i.e. Open Source). This is
perhaps a disappointment to companies who want to close the source and sell their own version. However it's a big advantage to hardware companies: they get a huge set of free programmers and they can tune the software in-house for their needs.
3. Microsoft might be broken up. If so, expect to see Word for Linux. Expect to see legal complications for MS if they don't do it.
Word running under KDE would allow Linux to provide the basic needs of 90% of non-gaming Windows users. And games aren't of any interest to a lot of people.
4. MS has stepped on a lot of toes to get where they are. This is in part why so many companies are interested in Linux. They would like to get MS out of the picture.
5. It's all about apps. A lot of programmers like Linux. If you can get programmers on board,
apps will follow; and users won't be far behind.
6. Average users don't like windows that much. My non-techie former girlfriend would curse Windows regularly. When she saw that I didn't have the same problems she expressed considerable interest in Linux (without prompting from me)---until I told her there wasn't Word for Linux.
None of these is a strong enough point to show that Linux _will_ become a viable average user desktop. Some of the points are even arguable; I tried to throw in everything I could think of. But I think it's indisputable that there are a lot of reasons why a lot of suits and non-geeks would like to see Linux succeed. And the battle has yet to be fought.
Indeed. And as such, I believe he comes in line after Speaker of the House for the presidency if everyone else dies. After 50
years his presidential bid would finally be
vindicated! Anyone else here seen Kind Hearts
and Coronets with Alec Guinness? Great movie!
Guinness played all the noble victims of an
aggresive and murderous social climber---bump
off all the relatives closer to the title than
you are and
assume it yourself.
Movie remake pitch: Picture Strom
killing Shrub Bush while Shrub is in a fox hunting outfit and then killing Dick Cheney in drag, both victims played by hmmm.... I like
Kevin Spacey for this. Strom could be played
by Bob Hope or the exhumed corpse of George Burns. It'd make a great remake with a twist ending: he could get caught but then pardon himself in the end.
Sorry for the dubious quality of this post. It's late and I'm still jet
lagged from flying from Sydney to Washington DC
a few days back. I promise to produce a more
coherent screenplay later.
Slashdot Mirror
The online rate is rather unreasonable. I don't know why they do that since the compact dead tree version (with magnifying glass) is more reasonably priced (at least for what you get). If memory serves it can be had for under $300.
Be sure to be specific when Fortran bashing. Fortran 77 should justifiably elicit shudders and shouts, but Fortran 90 and later are very nice for scientific programming. In fact, modern Fortran doesn't look a whole lot like Fortran 77.
If I had moderator points right now, I'd bump this up. Usenet reached a low for a few years there, but it has gotten better as more people have moved to the web.
Slashdot can be fun, but web based forums just don't work as well for extended discussions. Unless you're really fast, your chances of having several meaningful back-and-forth posts with someone here are negligible. By the time you've made a post and someone has responded to it, it's yesterdays Slashdot and you might as well forget about posting a followup. For what it does Usenet is still the best.
The core routines (linear system solvers, etc.) in matlab were LINPACK (linear systems) and EISPACK (eigenvalue routines) from the beginning. Cleve Moler, who wrote the original matlab, was one of the coauthors of LINPACK. I haven't followed it, but if there has been any change here, I suppose Matlab might have incorporated the more recent LAPACK. Either way Speed of the core computations shouldn't be a big problem.
It's using Matlab as anything other than a slick interface to Fortran code that causes performance problems. Memory in particular is a real killer with Matlab. Unfortunately, it's inherent in the language. Matlab lacks pointers or any ability to pass matrices to functions by reference. The upshot of this is, instead of allowing you to overwrite a matrix, matlab has to allocate new space to store the results of every computation. Loops aren't terribly fast either, though many matlab programs can work around this using the subarray notation. Octave suffers from similar problems.
Fixing these problems would turn Matlab into a real programming language, but it would cut down on its ease of use, readability, and interactivity. If you did this you would essentially get something that looked like Fortran 90 with some really nice numerical libraries. That's not the point of matlab, though I think it's a pretty good argument for Fortran 90 as what you might use for production code after experimenting/prototying in matlab.
Yep. Naproxyn sodium (Aleve), ibuprofen and aspirin are all anti-inflamatories and have similar effects and side effects. Definitely good first treatment for tendonitis. (Using naproxyn sodium for tendonitis in my shoulder at the moment).
Acetominaphen isn't an anti-inflamatory at all as far as I know.
Could be. Unfortunately I never dug deep enough into the 6502 to figure that out. I was just a teenager at the time farting around with assembly. Those sort of concerns would have gone right over my head. Having to clear carry bits and limitations on indirect addressing, etc. seemed like annoying quirks that couldn't (in my mind) have any reasonable justification. I loved the graphics chip set on my Atari, but I was jealous of fancier instructions on the Z80 and 6809.
With hindsight, I suspect a lot of 6502 quirks were design decisions. It seemed like a fairly tightly optimized processor. It never ran at a very high clock rate compared to the Z80, but comparatively it did pretty well at the lower clock rates. I should probably dig up info on how many clock cycles the Z80 and 6502 took to execute various instructions. My memory might be faulty here and this wouldn't be too hard to check.
I was in the Atari camp, but talk of 6502 assembly can put me on a real nostalgia trip. Did anyone else wonder why there wasn't an add without carry instruction? Nearly 5% of my code must have been clearing the carry bit (CLC). The quirks of this chip were burned on my brain during my teenage years.
I still have my Atari 800, though I haven't run it for years. I can't quite bring myself to get rid of it.
The custom chip set for the Amiga was designed by Jay Miner who also did the graphics chip set for the Atari 8-bit machines. I don't think he had come over to Commodore when the C64 was being designed, but I'm not sure and can't turn up anything on a web search. I also don't know if he is still alive. The Atari 800 and the Amiga were great machines in their day. If he is dead, I'll belatedly mourn his passing.
The recent resurgence in public debate about nuclear power has been really hard to follow. Both sides seem to disagree on the facts, so I can only conclude that someone (possibly both sides) is bullshitting.
On the one hand, I recently read an editorial in the Washington Post in which a pro-nuke activist claimed that recent advances in reprocessing technology make it possible to process old fuel, without producing weapons grade plutonium and leaving behind stuff that has a half-life measured in decades (considerably alleviating concerns about long-term storage).
On the other hand, I've never heard an anti-nuke type address this possibility. From their point of view reprocessing is always a proliferation risk and they never mention the possibility that reprocessing might help with the storage problem.
Then there are completely conflicting accounts on whether nuclear power is economically reasonable. The anti-nukes claim that reactors have only worked because of subsidies. The pro-nukes claim that this is only because of lack of standardization in the industry and the burden of unreasonable opposition to building new plants.
Figuring out who is bullshitting is starting to look like a major research project to me. I'd love to find a source that didn't bring in a great deal of biased axe-grinding, but I'm pretty sure I haven't seen it yet.
I'll repeat what others have said: This problem has been solved for quite some time; governments just haven't required diesel engines to have proper emmissions controls. What is taking longer to deal with is getting sulfur out of diesel fuel, but even that won't be a problem soon. As others have pointed out diesel engines can run on just about anything that will burn with enough compression. This opens the possibility of all sorts of renewable fuel sources.
And diesel engines have one advantage that appeals to me at least: if they are cared for reasonably they last a hell of a long time compared to gasoline engines. Also, diesel cars used to easier to work on for the home mechanic since they didn't previously have all the computerized stuff. (This latter point has probably changed with the new cleaner diesels.)
That is the standard contract. However there are a couple of ways around it. First, authors are often publishing after they've released a departmental technical report that might differ only in minor ways from the journal paper. No journal publisher is going to try to prevent you from releasing an earlier technical report on the web.
Also, I know people who cross out offensive sections of the contract or write in the margins that they reserve the right of web publication. This generally works.
Finally, even if you sign it, the journals just don't care that much about the odd article on someone's homepage. In my field, numerical analysis, I've never heard of anyone getting into trouble with a publisher over this sort of thing.
FWIW, I've got one of the bastards to sign right now. I guess I will sign as-is so as not to annoy my co-author. But it does piss me off.
Told by Feynman about Russell.
Or, as reputed to have occurred in an interchange between Bertrand Russell and a little old lady after a lecture by Russell on planetary motion:
"What you have told us is rubbish. The world is really a flat plate supported on the back of a giant tortoise".
Russell gave a superior smile before asking what the turtle was standing on.
"You're very clever young man, very clever," replied the old woman, "but it's turtles all the way down."
Maybe that's the year in the movie; I don't remember. I had a copy of the book handy on the bookshelf next to the computer and it's definitely 1997 there. It's near the end where Dave lobotomizes Hal.
Also, the festivities in Urbana weren't in 1992. I would have been there. So they must have been in 1997.
To go off on a tangent, the first time I heard about the Hal shifted one letter = IBM was in 1981. I was a proto-geek of 13 with a brand new Atari 400, an assembler cartridge, a lovely tape drive for storing programs and a small B/W "monitor" on which I could also watch Dukes of Hazard during moments of rest. I was teaching myself 6502 assembly. I tried to start a computer club at school, but a luddite teacher axed the idea and told me that computers were "dehumanizing" and "HAL+1=IBM. Think about it." Bastard.
Ah well. Even without the club I eventually learned to clear carry bits before starting an addition (what, no add without carry on a 6502!!?) and wrote some really unplayably crappy games. Those were fun days.
I suspect you know, even if you are getting "corrected" by others, but perhaps others don't. That would have been Hal's 4th B-day. He was born on Jan. 12 1997 in Urbana Illinois (the host town to the university where I spent 9 years...). There was something commemorating that event at the time, though I had left for greener pastures and missed it.
Maybe it's still not too late to send a card: Hal 9000, Production number 3, c/o Hal Plant, Urbana IL 61801.
"Dave, are you still there? Why didn't you remember my birthday?"
That is a good idea. However one legal entity would probably have to hold the copyright. This means you would have to sign over your copyright. To some extent I think this is what the FSF does: they encourage people to sign over their copyright so they can handle legal details. If you like the GPL, signing over copyright to the FSF seems like it might have some potential (but, alas, largely untested in court) advantages.
Regarding another person's response to your post (about whether multiple programs in a compilation can be registered at once): I'm not sure I see a problem. The quoted sections seem not to be targeted toward compilations, but people do manage to defend compilations under copyright law. The trick in complying with these rules seems to be deciding what printed media to provide. An energetic organization could provide 25+25 pages for each program in the compilation. (It's a lot, but not completely out of the question if you're not doing it by yourself on one printer). Copyright would be formally established and stealing a single program from a registered copyrighted compilation in which the same organization held copyright on each program could never be justified as fair use.
Hmmm... On the other hand, I have a Windows '98 "compilation" CD. I wonder if MS printed out the required pages from each program.
There is a long tradition of this sort of thing. My PhD project involved wasting lots of time on Usenet. The only major problem with this was that it was not particularly original: most other grad students were doing similar things.
The graduates of most CS departments are not going to have the background for programming for scientific applications. That's just a fact. Math and numerical analysis courses used to be a part of the CS curriculum long ago, but they no longer are. Many CS departments don't like supporting this sort of research or courses and many academics interested in scientific computing are now in math or engineering departments.
Where does this leave you? You need to look around and find people with different (i.e. not straight CS) backgrounds. Many CS, math and engineering departments are collaborating on creating degree programs in scientific computing. This is often for graduate degrees, but not in all cases. Many math departments have an emphasis on scientific computing and computational mathematics. Many engineers become quite good programmers either in the course of getting a degree or on the job.
If you look around you will find people. I can assure you there are a lot of people interested in this sort of thing who are annoyed that most of the jobs seem to be in business related programming.
Not to point out the obvious, but choosing not to prosecute people for copying a page of a textbook etc. is something the government can do without exercising any authority whatsoever. You may disagree with fair use, but it's hardly an aggressive extension of government authority.
Your car example is not analogous. Someone whose intellectual property rights are violated is not denied the use of that property. So I would obviously be very unhappy if you took my car. However I am not ungenerous. I won't let you have my car, but I will grant you the right to quote selected portions of this post if you wish to post a reply. Ain't fair use grand?
I wasn't able to get back on to the page to check, but I don't remember a great level of detail on the computations that are being done. Non-linear control problems associated with robotics aren't generally thought of as the easiest thing in the world. There's a lot of math behind the stabilization of non-linear systems. It's possible that they are using a fairly computationally intensive method. This would not be like slapping together a PID controller and tuning the parameters until it can walk without falling over.
Whether they really need 700MHz processors in each leg, I can't say. But it's not a possibility you could dismiss without knowing a fair bit about non-linear control methods and what method was used in particular in this case. In any event these processors are going to be doing a hell of a lot more than "telling" the legs to move up and down regularly.
And it's actually readable if you have a really solid math background. (This isn't intended to be a vacuous comment: a lot of mathematical papers are only readable by specialists and not even by other mathematicians working in other areas.)
IMO the really amazing gem in this paper is the noisy coding theorem. Shannon proved that if you have a communication channel that introduces errors (i.e. a noisy channel) then that channel always has a channel capacity. If you send data at a rate lower than the channel capacity then you can prove that there are coding schemes that result in a average probability of a bit error being as small as you like. (However smaller error rates mean more complicated codes). Unfortunately Shannon's proof was nonconstructive. He essentially showed that if you transmit the data in large blocks and encode those blocks randomly you get a probability of a bit error that goes to zero as the block size increases. These codes do not require ever increasing redundancy to achieve this.
Unfortunately decoding a random encoding applied to a huge block of data is not computationally practical. A major thread in information and coding theory has been trying to find tractable codes that behave somewhat like Shannon's codes. It's a hard problem, but your modem works better because of it.
Personally I like this result because it was so counter-intuitive: who else would have thought it might be possible, even theoretically, to construct codes that make the error rate approach zero without introducing ever more redundancy into the code?
Shannon was a giant in engineering, computer science and mathematics.
Some spoilers, although I don't think they give enough detail to ruin anyone's enjoyment of the movie:
I watched this bit (and the rest of the movie) in the theater in union station (the basement floor). It was a bit surreal watching the place you are in on film. It got a good laugh from the audience.
For what it's worth, it wasn't as good as the first film, but I do think Moore was well cast and did a good job and Hopkins was wonderful as usual. Unfortunately, it didn't have the psychological edge of the Silence of the Lambs. And the funny bits didn't mesh with the rest of the movie (I'm pretty sick, but even I couldn't laugh at dinner table conversation over a plate of brains at the end). OTOH, it did make me feel ill; I like gruesome horror films and I rarely find anything that either scares me or makes me feel like I have to look away. This film at least pushed me to my limits on gore and it had some good acting. I enjoyed it, but they really should have worked more on the screenplay and done something creepier instead of just going for something that was supposed to be disgusting and funny. Does every film have to have laughs in it? What's wrong with just trying to scare us?
More like 18 million. And, Vegemite aside, a fair bit of good stuff comes from Australia. They also have a tradition of exporting their best and brightest to work in other countries (funding for universities really sucks under the current Australian government---after living there for a few years, I just moved back because it's easier to pursue a career in the USA). If we could get figures for ex-pat Aussies working on projects I suspect that everyone would be quite impressed for what Australians do with only 18 million people.
I respectfully disagree. Linux has only in the last year or two gotten to the point where I would start to recommend it to reasonably average Windows users. And I would still only recommend it to those who are frustrated with Windows' "quirks" (i.e. crashes) and who have fairly focused and specialized computing needs. IMO the battle for the desktop hasn't even really begun yet.
I think there are a few forces that might well push Linux on to more desktops.
1. Linux is free. Windows is starting to look like a hefty chunk of the price of a new computer. With profit margins where they are, equipment manufacturers have to be interested.
2. Linux is Free (i.e. Open Source). This is perhaps a disappointment to companies who want to close the source and sell their own version. However it's a big advantage to hardware companies: they get a huge set of free programmers and they can tune the software in-house for their needs.
3. Microsoft might be broken up. If so, expect to see Word for Linux. Expect to see legal complications for MS if they don't do it. Word running under KDE would allow Linux to provide the basic needs of 90% of non-gaming Windows users. And games aren't of any interest to a lot of people.
4. MS has stepped on a lot of toes to get where they are. This is in part why so many companies are interested in Linux. They would like to get MS out of the picture.
5. It's all about apps. A lot of programmers like Linux. If you can get programmers on board, apps will follow; and users won't be far behind.
6. Average users don't like windows that much. My non-techie former girlfriend would curse Windows regularly. When she saw that I didn't have the same problems she expressed considerable interest in Linux (without prompting from me)---until I told her there wasn't Word for Linux.
None of these is a strong enough point to show that Linux _will_ become a viable average user desktop. Some of the points are even arguable; I tried to throw in everything I could think of. But I think it's indisputable that there are a lot of reasons why a lot of suits and non-geeks would like to see Linux succeed. And the battle has yet to be fought.
Indeed. And as such, I believe he comes in line after Speaker of the House for the presidency if everyone else dies. After 50 years his presidential bid would finally be vindicated! Anyone else here seen Kind Hearts and Coronets with Alec Guinness? Great movie! Guinness played all the noble victims of an aggresive and murderous social climber---bump off all the relatives closer to the title than you are and assume it yourself.
Movie remake pitch: Picture Strom killing Shrub Bush while Shrub is in a fox hunting outfit and then killing Dick Cheney in drag, both victims played by hmmm.... I like Kevin Spacey for this. Strom could be played by Bob Hope or the exhumed corpse of George Burns. It'd make a great remake with a twist ending: he could get caught but then pardon himself in the end.
Sorry for the dubious quality of this post. It's late and I'm still jet lagged from flying from Sydney to Washington DC a few days back. I promise to produce a more coherent screenplay later.