Re:"Like another legendary creature...."
on
The History of UNIX
·
· Score: 1
I'm pretty sure they were referring to the Phoenix. Check out just about any episode of G-Force for a demonstration of the mythological underpinnings:)
One of the main reasons for the collapse of the Roman Empire was that without a 0 in their number system, they had no way of knowing whether or not their system calls were working...
Uh, I don't think the guy submitted his own article to slashdot. He got some other people to put it up on their site (which is their problem, right?). Slashdot members then inflicted this paper on themselves...
And your crack about the Unabomber is ironic, in a sad kind of way. I quoth from the Rising Tide Summit Press Release:
In his 1992 book, "Mirror Worlds," Yale University Computer Scientist David Gelernter laid the basis for the development of Java programming and Sun Microsystems' Jini. Critically injured by the Unabomber in 1993, he was sidelined during the Web's formative years.
Be careful what you wish for, you just might get it. I wonder how many patents they actually hold on those crazy ideas?
Gelernter may come across as pretnetious and oblivous to details, but in my experience he's no dummy. He came up with the Linda coordination language, which was IMHO the most elegant method of building parallel system that I have ever seen.
Perhaps being blown up by the Unabomber has affected his worldview a bit. Did anyone else notice that his text bears a passing stylistic similarity to the nutter's manifesto?
The example you give is not the same thing. The car key example is using distance in a rotational space to differentiate, while double clicking is using difference in time. I am used to double clicking, but it still occasionally screws me up.
My vote is that the double click has over time become accepted because users are familiar with it, and because systems people have tweaked the implementation to the point that it's not horrible. But the concept is still a bad idea.
Wow, I admire your detail of vision! I'll be performing meticulous testing of the nano-brewery if you need me.
Seriously, Humanity's biggest advantage and biggest curse is our adaptability. We all get rich -- so what? We still want more. We want more because our peers have more and we seem to be hardwired to keep up with the Jonses.
Most of us already live lives that would be considered idyllic by 19th century standards, but we strive for more, and in striving we use more resources. Given God-like control over our environment, I am a little wary of what the Joe Average (or I for that matter) will do once he has a few beers in him... and this excludes the danger of evil Nano-Bond villians who will try to destroy the world through self-replicating doomstay devices.
My take: Humanity will survive, but there will be casualties. We need to get off this beautiful rock if for no other reason than to put some eggs in other baskets.
I for one do not agree that a segmented memory model is a bad thing, unless of course the segments are too small, which is exactly what happened with the introduction of the 286.
When the 386 came out, segments could be up to 4GB in size, IIRC, but by then it was too late... the concept had a bum rap.
I recall that Danny Hillis (ex of Thinking Machines Corp, now works at Disney, I believe) used to do GA stuff. He and some buds made a GA that generated sorting algorithms. But it would get stuck on local minima, and would settle on suboptimal solutions.
So they added another element to the environment -- another set of GA-bred algorithms that generated sets of numbers to be sorted. Their goal was to create data sets that made the array sorters perform poorly! Excellent!
So whenever the sort critters found a nice local minima, the nasty data set generators would find their achilles heel and chase them all away from that area.
I really liked the predator/prey flavour of the idea.
The overhead of quicksort gets to be a drag when the size of the sorted list is really small (2 or 3 itemsl as you say). So some implementors have chosen to modify quicksort so that it does another kind of sort during the dash to the finish line.
But in its purest form quicksort don't play that game. Hey, there are other mods that people do to quicksort to increase its performance (like not blindly choosing the first element of a subsection as a pivot point -- this can lead to order n squared performance if the data is already sorted).
The bottom line is that you can still call it a quicksort even if you've tweaked it a bit... otherwise we'd have a different name for each implementation of quicksort ever written:)
Gee, all the card manufacturers have to do (repeat, have to do) is to rewrite their drivers each time MS releases a new OS.
New driver interface for 95, NT 3.5, NT 4.0, Win2k. I hear that with Win2k they actually got around to specifying a 32 bit aware interface between drivers and the kernel.
Actually, it has a worse performance hit, because it is fixed at 4 times oversampling for each pixel. What the GeForce does (as of the leaked 519 drivers) is render the scene with dimensions twice as big as the original scene in x and y dimensions (4x the number of pixels) and then process the results to get a nice antialiased scene.
The V5 allows you to choose between 4x AA and 2x AA, giving you the option of having antialiased scenes with half the performance hit of the GeForce.
Yet that did not seem to wow the reviewers, so my guess is that 3dfx will have a hard time selling these silicon-heavy but performance-light cards.
Youbetcha. I never used Multics, but I read a book on its design. Those folks thought things THROUGH. They had plans for hot-swapping CPUs (back when they were big as houses), capability-based security (I think, it's been a while) and good stuff like that.
In many ways we of the Present have NIH syndrome. It can't be good unless it's new. But you know what they say...
A month in the lab can often save an afternoon in the Library.
I agree. If you believe that there is nothing supernatural about the human mind, then it follows that AI is inevitable, really... and the prospect is both fantastic and frightening.
In general, all living things desire survival -- and if AIs have this trait then interesting times will be a-coming.
Heck, even if AIs are supremely subserivent they will still redefine our world. In an environment where more and more people are defined as "knowledge workers", there is a good chance that AIs will be better at our jobs than we are.
Hmmm... I wonder what kind of renumeration an AI would want for writing a piece software?
I remember scanning an interview with Neal Stephenson about The Diamond Age... when asked what the biggest challenge in coming up with the plotline was, he responded with something like: "Visualizing a future where nanotech is commonplace and everyone isn't dead"
Lots of interesting technologies are advancing at breakneck pace right now. I see several different ways that humanity could become irrelevant, but there are a few (nanotech comes to mind) that have the potential to poison our little blue home to the point where nobody can live here anymore.
And as time goes by, the genius/madness factor required to do such a thing gets smaller.
Perhaps that alone is a good reason to pursue the creation of Human-friendly habitats in space. Right now all our eggs are in one basket, and we don't exactly know how fragile the basket is.
Yes, yes -- Technology Changes Morals -- is a recurring Niven theme. Remember A Gift From Earth? Great novel that dealt with the moral and social consequences of organ transplant technology.
The whole Known Space body of work was extremely well done, in my opinion. Gift, Protector, and World of Ptavvs are masterfully done books that do not get the attention that they deserve.
Neither does A World Out Of Time, but it's not Known Space, so I didn't mention it here. Doh!
I read the Before The Golden Age anthologies by Asimov when I was a kid, and they were very cool. These were the stories that Asimov loved when HE was young.
My take on the collection: some good, some bad, and some forgotten gems that make it all worthwhile.
For the younger reader though, I'd suggest the excelent A Night In the Lonesome October by Zelazny. A bit fantasyish, but a Zelazny fantasy often seems more plausible than many other authors' SF...
Many are those who rave about A Fire Upon the Deep, and it is for sure a great book. But my all time favourite Vinge is the short story collection called True Names and Other Dangers. The headliner piece is an incredible projection of the future of consensus reality built via computer systems, and it was written back when a PDP-11 was a decent machine.
In general, developers sacrifice time and space efficiency to get faster development.
It happens on a small scale when you get something going that's "good enough" and you don't want to reimplmement to make it more efficient.
It happens on a large scale when you use prebuilt components that have more functionality than you need, but hey, they're there and they get the job done.
It happens on a really large scale when people hatch Bad Plans and they become standards. Example: the horrid MS DLL plan. Designed to make apps smaller, but the app installers can't guarantee that the target machine has the latest DLLs, so it has to carry around all the DLLs that the app uses, even though the app uses only a fraction of the code in those DLLs.
Slashdot Mirror
I'm pretty sure they were referring to the Phoenix. Check out just about any episode of G-Force for a demonstration of the mythological underpinnings :)
Regards, your friendly neighbourhood cranq
Another Little Known Fact:
One of the main reasons for the collapse of the Roman Empire was that without a 0 in their number system, they had no way of knowing whether or not their system calls were working...
Regards, your friendly neighbourhood cranq
Uh, I don't think the guy submitted his own article to slashdot. He got some other people to put it up on their site (which is their problem, right?). Slashdot members then inflicted this paper on themselves...
And your crack about the Unabomber is ironic, in a sad kind of way. I quoth from the Rising Tide Summit Press Release:
In his 1992 book, "Mirror Worlds," Yale University Computer Scientist David Gelernter laid the basis for the development of Java programming and Sun Microsystems' Jini. Critically injured by the Unabomber in 1993, he was sidelined during the Web's formative years.
Regards, your friendly neighbourhood cranq
Gelernter may come across as pretnetious and oblivous to details, but in my experience he's no dummy. He came up with the Linda coordination language, which was IMHO the most elegant method of building parallel system that I have ever seen.
Perhaps being blown up by the Unabomber has affected his worldview a bit. Did anyone else notice that his text bears a passing stylistic similarity to the nutter's manifesto?
Regards, your friendly neighbourhood cranq
The example you give is not the same thing. The car key example is using distance in a rotational space to differentiate, while double clicking is using difference in time. I am used to double clicking, but it still occasionally screws me up.
My vote is that the double click has over time become accepted because users are familiar with it, and because systems people have tweaked the implementation to the point that it's not horrible. But the concept is still a bad idea.
Regards, your friendly neighbourhood cranq
Wow, I admire your detail of vision! I'll be performing meticulous testing of the nano-brewery if you need me.
Seriously, Humanity's biggest advantage and biggest curse is our adaptability. We all get rich -- so what? We still want more. We want more because our peers have more and we seem to be hardwired to keep up with the Jonses.
Most of us already live lives that would be considered idyllic by 19th century standards, but we strive for more, and in striving we use more resources. Given God-like control over our environment, I am a little wary of what the Joe Average (or I for that matter) will do once he has a few beers in him... and this excludes the danger of evil Nano-Bond villians who will try to destroy the world through self-replicating doomstay devices.
My take: Humanity will survive, but there will be casualties. We need to get off this beautiful rock if for no other reason than to put some eggs in other baskets.
Regards, your friendly neighbourhood cranq
I for one do not agree that a segmented memory model is a bad thing, unless of course the segments are too small, which is exactly what happened with the introduction of the 286.
When the 386 came out, segments could be up to 4GB in size, IIRC, but by then it was too late... the concept had a bum rap.
Regards, your friendly neighbourhood cranq
I recall that Danny Hillis (ex of Thinking Machines Corp, now works at Disney, I believe) used to do GA stuff. He and some buds made a GA that generated sorting algorithms. But it would get stuck on local minima, and would settle on suboptimal solutions.
So they added another element to the environment -- another set of GA-bred algorithms that generated sets of numbers to be sorted. Their goal was to create data sets that made the array sorters perform poorly! Excellent!
So whenever the sort critters found a nice local minima, the nasty data set generators would find their achilles heel and chase them all away from that area.
I really liked the predator/prey flavour of the idea.
Regards, your friendly neighbourhood cranq
Backprop was a pretty good breakthrough (it kinda rejuvenated the neural net community, if I recall).
I guess the Boltzmann machine would fall under your simulated annealing section...
I think genetic programming should be in there too, but you can't call a GA producer an algorithm... it's more of a meta-algorithm, I guess...
Regards, your friendly neighbourhood cranq
The overhead of quicksort gets to be a drag when the size of the sorted list is really small (2 or 3 itemsl as you say). So some implementors have chosen to modify quicksort so that it does another kind of sort during the dash to the finish line.
:)
But in its purest form quicksort don't play that game. Hey, there are other mods that people do to quicksort to increase its performance (like not blindly choosing the first element of a subsection as a pivot point -- this can lead to order n squared performance if the data is already sorted).
The bottom line is that you can still call it a quicksort even if you've tweaked it a bit... otherwise we'd have a different name for each implementation of quicksort ever written
Regards, your friendly neighbourhood cranq
How does capitalism reconcile itself to the concept of Open Source?
People who simply give stuff away seem to screw up any economic model that I have ever heard of...
Regards, your friendly neighbourhood cranq
New driver interface for 95, NT 3.5, NT 4.0, Win2k. I hear that with Win2k they actually got around to specifying a 32 bit aware interface between drivers and the kernel.
Oh, the progress!
Regards, your friendly neighbourhood cranq
The V5 allows you to choose between 4x AA and 2x AA, giving you the option of having antialiased scenes with half the performance hit of the GeForce.
Yet that did not seem to wow the reviewers, so my guess is that 3dfx will have a hard time selling these silicon-heavy but performance-light cards.
Regards, your friendly neighbourhood cranq
Youbetcha. I never used Multics, but I read a book on its design. Those folks thought things THROUGH. They had plans for hot-swapping CPUs (back when they were big as houses), capability-based security (I think, it's been a while) and good stuff like that.
In many ways we of the Present have NIH syndrome. It can't be good unless it's new. But you know what they say...
A month in the lab can often save an afternoon in the Library.
Say, whatever happened to tagged storage?
Regards, your friendly neighbourhood cranq
I agree. If you believe that there is nothing supernatural about the human mind, then it follows that AI is inevitable, really... and the prospect is both fantastic and frightening.
In general, all living things desire survival -- and if AIs have this trait then interesting times will be a-coming.
Heck, even if AIs are supremely subserivent they will still redefine our world. In an environment where more and more people are defined as "knowledge workers", there is a good chance that AIs will be better at our jobs than we are.
Hmmm... I wonder what kind of renumeration an AI would want for writing a piece software?
Regards, your friendly neighbourhood cranq
I remember scanning an interview with Neal Stephenson about The Diamond Age... when asked what the biggest challenge in coming up with the plotline was, he responded with something like: "Visualizing a future where nanotech is commonplace and everyone isn't dead"
Lots of interesting technologies are advancing at breakneck pace right now. I see several different ways that humanity could become irrelevant, but there are a few (nanotech comes to mind) that have the potential to poison our little blue home to the point where nobody can live here anymore.
And as time goes by, the genius/madness factor required to do such a thing gets smaller.
Perhaps that alone is a good reason to pursue the creation of Human-friendly habitats in space. Right now all our eggs are in one basket, and we don't exactly know how fragile the basket is.
Regards, your friendly neighbourhood cranq
Yes, yes -- Technology Changes Morals -- is a recurring Niven theme. Remember A Gift From Earth? Great novel that dealt with the moral and social consequences of organ transplant technology.
The whole Known Space body of work was extremely well done, in my opinion. Gift, Protector, and World of Ptavvs are masterfully done books that do not get the attention that they deserve.
Neither does A World Out Of Time, but it's not Known Space, so I didn't mention it here. Doh!
Regards, your friendly neighbourhood cranq
I read the Before The Golden Age anthologies by Asimov when I was a kid, and they were very cool. These were the stories that Asimov loved when HE was young.
My take on the collection: some good, some bad, and some forgotten gems that make it all worthwhile.
Regards, your friendly neighbourhood cranq
Shockwave Rider rocked.
For the younger reader though, I'd suggest the excelent A Night In the Lonesome October by Zelazny. A bit fantasyish, but a Zelazny fantasy often seems more plausible than many other authors' SF...
Regards, your friendly neighbourhood cranq
Many are those who rave about A Fire Upon the Deep, and it is for sure a great book. But my all time favourite Vinge is the short story collection called True Names and Other Dangers. The headliner piece is an incredible projection of the future of consensus reality built via computer systems, and it was written back when a PDP-11 was a decent machine.
IMHO, it rings truer than ever when read today.
Regards, your friendly neighbourhood cranq
In general, developers sacrifice time and space efficiency to get faster development.
It happens on a small scale when you get something going that's "good enough" and you don't want to reimplmement to make it more efficient.
It happens on a large scale when you use prebuilt components that have more functionality than you need, but hey, they're there and they get the job done.
It happens on a really large scale when people hatch Bad Plans and they become standards. Example: the horrid MS DLL plan. Designed to make apps smaller, but the app installers can't guarantee that the target machine has the latest DLLs, so it has to carry around all the DLLs that the app uses, even though the app uses only a fraction of the code in those DLLs.
Regards, your friendly neighbourhood cranq