Boy do you need to go back to school. Edsger wrote more and better stuff in his lifetime than anyone here on Slashdot. Did you ever get directions from Google Maps or Mapquest? Thank Edsger -- his shortest path algorithm is what they all use, and by the way, he wrote that before you were born, most of you.
You know the semaphores used in the multi-cpu Linux kernels? Yep, you owe Edsger for them, too. And programming languages like C, Pascal, etc.? He helped write the first Alogol compiler, the great-grand-daddy of them all, once again before most of you were born.
Just because he eschewed the run-break-fix approach so beloved of the folks who are spewing billions of lines of error-laden code into the world today, doesn't mean he hadn't forgotten more about writing code than most folks here have ever learned. And yes, he advocated developing code formally, and he liked to do it with pen and paper.
So learn about who you're making snide comments about, and show some respect.
When people are still using any algorithm you came up with 30 years from now, you will have the right to say something about Edsger Dijkstra.
I remember years ago at a conference overhearing two gals from Yahoo discussing how some of their male coworkers had developed an image search algorithm that tried to find pictures of nipples. The punch line was "of course, they were trying to find them, not filter them..."
Couldn't you do a captcha where the first presentation has no cats? The user has to hit the refresh once or twice before seeing a cat, and then pick it; if they pick any of the non-cats,
you call them a 'bot...
If you look at the bottom of the linked page, there are now links to two PDF files which contain a lot of the sort of details removed from the main web page...
Well, let's see... Monopolies have been illegal and regulated since the archetypes -- U.S. Steel, Standard Oil, etc. which grew without bounds
without government-granted monopolies. Fortunately the economically clueless folks in the country haven't yet dismantled the anti-trust laws entirely.
The Securities and Exchange commission has to approve big mergers, thus some
semblance of competition is maintained.
The point is that were you to remove that structure, big competing companies would merge into monopolies, to the detriment of the economy, and history has demonstrated this, yeilding a legal structure to prevent that "bug" in the capitalist model.
In the 1930's, lots of banking regulations were put into place in response to
the collapses of the Great Depression. A few years ago, we had major banking
deregulation. Now we have the current mortgage crisis, because we let loose
one of the bugs in the capitalist system -- people will take loans that they can pay in the short term but not in the long term, if you let them.
That doesn't mean capitalism is wrong or broken, it just means that capitalism only works with some amount of regulation; and some folks seem to forget that
from time to time. Can there be too much regulation? Sure! But there can also be too little.
My opinion on this, on the other hand, is that folks who have studied economics have learned that unbridled (unregulated?) capitalism leads inexorably to monopolies using unfair business practices to stomp out competition, and to
people doing things like buying stock on margin that they can't afford to actually buy (leading to the Great Depression), or to people buying houses with variable rate mortgages they can't afford to pay (leading to the current credit mess).
Therefore, the capitalist system needs regulation to keep it out of those corners where it doesn't work anymore.
This is why when we get people in power who don't realize this, and start turning off more and more of government's regulatory facilities, (rather than making them less oenerous but keeping them effective) we end up with problems in our economy.
I disagree with your premise. Programming languages do not really remove whole categories of errors.
To take the oft-cited example of array bounds,
just because your programming language throws an exception when you attempt to address an array out of bounds doesn't mean the condition is avoided, it just means an exception gets thrown rather than the program dumping core. And it doesn't mean the exception thus generated is handled in a way that makes the behavior of the code correct.
Or take type-checking. Sure, in a strictly typed language I can't directly make the
mistake of saying: int_variable = float_variable
when I ought not to, because the smart compiler will warn me that I'm throwing away precision, or that it's
a type clash, or whatever. But the foolish programmer simply "fixes" the code to say: int_variable = trunc(float_variable)
And voila, the compiler is happy. The same semantic error is being made, but the
compiler has been told that's what you really meant to do, and is therefore quiet
about it.
Rather, the probability of programmer error has much more to do with whether the programmer has thought through the boundary conditions etc. rigorously instead of declaring it working the second it compiles and runs a single test case.
Oh, ye of lesser cynicism. I also, long ago, used to believe that language features could improve software reliability. Nowadays the idea just makes me cackle -- in actuality the universe just invents better idiots.
We do liquid helium for superconducting magnets, that need to be that cold, but computers we just cool with water pumped to/from a nearby pond. Of course then you have to deal with the zebra mussels...
The whole point of Turing's thesis was that the Turing machine was a constructable physical device. Sure, one reasons about them mathematically, and can describe them mathematically, but the whole point was that they could be physically built.
Secondly, lots of folks on this thread are confusing Simple Turing machines which implement some specific algorithm or other with the Universal Turing Machine, which can read a description of a Simple Turing machine (i.e. a program) and then emulate its execution.
I didn't mean to imply there was any lack of technical know-how. The problem is that the US NRC won't approve anything other than the same-old TMI-style design...
It doesn't help that our country insists on rebuilding the same old, flawed design for nuclear power plants, rather than any one of a dozen or so better designs that are out there which are far safer. The system we're using was designed with a separate system of breeder reactors in mind, to reprocess waste into fuel, which have never been built, and which (in the initial plan) involved schlepping nuclear waste all over the country.
Inherently safer designs like pebble bed reactors and molten salt reactors are not
being used, rather the same old Three Mile Island design is proposed for new plants.
Now of course, there are people who are against any sort of nuclear power, regardless. But I think that's largely because the past "Nuclear Power is perfectly safe" propaganda has made them untrusting of any statements about nuclear safety and/or dangers.
I grew up 13 miles downriver from Three Mile Island. So I know a lot of people with an axe to grind about nuclear safety; and most of them are not really "Greenies". Many of them still believe they haven't been told the whole truth about the accident there, much the way folks in the wider US population of a given age don't neccesarily believe they've been told the whole truth about the Kennedy assassination... So I think to win those folks over, you need a demonstrably safer design, and you need to really explain the details.
The gentle poster describes one case, where code taken from a BSD licensed product is a small portion of another product, which is propiretary. The other possibility is where the code from the BSD licensed product is the entirety of the proprietary product -- that is, someone is just plain selling the BSD licensed product in its entirety as a proprietary package. This repackaging hides the fact that the source code is available, and makes the users of the package who obtain it from the proprietary vendor lose their freedom to modify the source code and get better/different behavior from the software.
This is the sort of case that the BSD folks either aren't concerned about, or aren't thinking about. It isn't a matter of whether the "project" or the "code" is free, it is a matter of whether the users of the project or code are free.
The other aspect this discussion overlooks is the payment-in-kind model; a person
who gives away their hard work in software development using the GPL is expecting
other people who modify that work to "pay in kind" and give their code changes to the original author in "payment" for the code he gave them. So by giving their code away, they get code changes, bug fixes, etc. back. Sort of a GPL enforcement of the Christian/Buddhist "Give and ye shall receive" ideal.
Folks who modify BSD licensed software are under no compunction to send modifications, improvements, or bugfixes back to the original author. I suppose you could say the BSD folks are relying on actual philosophical principles to be repaid for what they give away, rather than license clauses...
If they haven't read them, then definitely any Anne McCaffery, Ray Bradbury, Madeline L'Engle, any Asimov you might have, Connie Willis, The Stainless Steel Rat series by Harrison, Laumer's Retief books, Clifford Simak's books, Zelazny's younger stuff (Madwand, etc.) , Barba Hambly's assorted series, the Roswell High books, Andre Norton's stuff, L.E. Modessitt Jr.'s assorted series...
If they're a little older... -- Ursula LeGuin, Octavia Butler, Philip K. Dick, Zelazny's later stuff, Vonnegut, Niven, Frank Herbert, Definitely Neal Stephenson...
And the computer related classics -- The Adolescence of P1 by Ryan, The Shockwave Rider by Brunner, Neuromancer by Gibson...
And of course fantasy, Tolkein, the Shanarra books, etc.
And then, only after they've read all of the usual fantasy stuff, then they can truly appreciate Pratchett...
Slashdot Mirror
Boy do you need to go back to school. Edsger wrote more and better stuff in his lifetime than anyone here on Slashdot. Did you ever get directions from Google Maps or Mapquest? Thank Edsger -- his shortest path algorithm is what they all use, and by the way, he wrote that before you were born, most of you. You know the semaphores used in the multi-cpu Linux kernels? Yep, you owe Edsger for them, too. And programming languages like C, Pascal, etc.? He helped write the first Alogol compiler, the great-grand-daddy of them all, once again before most of you were born.
Just because he eschewed the run-break-fix approach so beloved of the folks who are spewing billions of lines of error-laden code into the world today, doesn't mean he hadn't forgotten more about writing code than most folks here have ever learned. And yes, he advocated developing code formally, and he liked to do it with pen and paper.
So learn about who you're making snide comments about, and show some respect. When people are still using any algorithm you came up with 30 years from now, you will have the right to say something about Edsger Dijkstra.
You could make a frame of reference centered around any random religious shrine in Jerusalem, or around your neighborhood Taco Bell ...
I remember years ago at a conference overhearing two gals from Yahoo discussing how some of their male coworkers had developed an image search algorithm that tried to find pictures of nipples. The punch line was "of course, they were trying to find them, not filter them..."
The time we move to in the winter *is* the right time.
We should just stay there all year. I used to live in Indiana, most of which, up until a few years ago, did exactly that. It works fine.
They only changed it so as not to be always confusing people from other places.
And especially not "my junk is in my fanny-pack".
All in favor, tag the article!
Couldn't you do a captcha where the first presentation has no cats? The user has to hit the refresh once or twice before seeing a cat, and then pick it; if they pick any of the non-cats, you call them a 'bot...
Looks like it has an <div id=":qq"> around it...
Problem solved.
If you look at the bottom of the linked page, there are now links to two PDF files which contain a lot of the sort of details removed from the main web page...
With that kind of thinking, I begin to understand our current U.S. President...
The point is that were you to remove that structure, big competing companies would merge into monopolies, to the detriment of the economy, and history has demonstrated this, yeilding a legal structure to prevent that "bug" in the capitalist model.
In the 1930's, lots of banking regulations were put into place in response to the collapses of the Great Depression. A few years ago, we had major banking deregulation. Now we have the current mortgage crisis, because we let loose one of the bugs in the capitalist system -- people will take loans that they can pay in the short term but not in the long term, if you let them.
That doesn't mean capitalism is wrong or broken, it just means that capitalism only works with some amount of regulation; and some folks seem to forget that from time to time. Can there be too much regulation? Sure! But there can also be too little.
Therefore, the capitalist system needs regulation to keep it out of those corners where it doesn't work anymore.
This is why when we get people in power who don't realize this, and start turning off more and more of government's regulatory facilities, (rather than making them less oenerous but keeping them effective) we end up with problems in our economy.
To take the oft-cited example of array bounds, just because your programming language throws an exception when you attempt to address an array out of bounds doesn't mean the condition is avoided, it just means an exception gets thrown rather than the program dumping core. And it doesn't mean the exception thus generated is handled in a way that makes the behavior of the code correct.
Or take type-checking. Sure, in a strictly typed language I can't directly make the mistake of saying:
int_variable = float_variable
when I ought not to, because the smart compiler will warn me that I'm throwing away precision, or that it's a type clash, or whatever. But the foolish programmer simply "fixes" the code to say:
int_variable = trunc(float_variable)
And voila, the compiler is happy. The same semantic error is being made, but the compiler has been told that's what you really meant to do, and is therefore quiet about it.
Rather, the probability of programmer error has much more to do with whether the programmer has thought through the boundary conditions etc. rigorously instead of declaring it working the second it compiles and runs a single test case.
Oh, ye of lesser cynicism. I also, long ago, used to believe that language features could improve software reliability. Nowadays the idea just makes me cackle -- in actuality the universe just invents better idiots.
It's the National Fire Protection Association who writes things like the national electrical code. It's all about avoiding things that have caused fires.
We do liquid helium for superconducting magnets, that need to be that cold, but computers we just cool with water pumped to/from a nearby pond. Of course then you have to deal with the zebra mussels...
I originally found this one via Google, but I don't recall what I was searching for. I suspect it may be what you're looking for though.
Secondly, lots of folks on this thread are confusing Simple Turing machines which implement some specific algorithm or other with the Universal Turing Machine, which can read a description of a Simple Turing machine (i.e. a program) and then emulate its execution.
So computer programs are both equivalent to a physical machine, and an abstract mathematical proof, at the same time.
That is, you can argue it either way, and they're both right.
That said, I maintain that you should have to acutally build the Turing machine to get the patent :-)
I didn't mean to imply there was any lack of technical know-how. The problem is that the US NRC won't approve anything other than the same-old TMI-style design...
Inherently safer designs like pebble bed reactors and molten salt reactors are not being used, rather the same old Three Mile Island design is proposed for new plants.
Now of course, there are people who are against any sort of nuclear power, regardless. But I think that's largely because the past "Nuclear Power is perfectly safe" propaganda has made them untrusting of any statements about nuclear safety and/or dangers.
I grew up 13 miles downriver from Three Mile Island. So I know a lot of people with an axe to grind about nuclear safety; and most of them are not really "Greenies". Many of them still believe they haven't been told the whole truth about the accident there, much the way folks in the wider US population of a given age don't neccesarily believe they've been told the whole truth about the Kennedy assassination... So I think to win those folks over, you need a demonstrably safer design, and you need to really explain the details.
This is the sort of case that the BSD folks either aren't concerned about, or aren't thinking about. It isn't a matter of whether the "project" or the "code" is free, it is a matter of whether the users of the project or code are free.
The other aspect this discussion overlooks is the payment-in-kind model; a person who gives away their hard work in software development using the GPL is expecting other people who modify that work to "pay in kind" and give their code changes to the original author in "payment" for the code he gave them. So by giving their code away, they get code changes, bug fixes, etc. back. Sort of a GPL enforcement of the Christian/Buddhist "Give and ye shall receive" ideal. Folks who modify BSD licensed software are under no compunction to send modifications, improvements, or bugfixes back to the original author. I suppose you could say the BSD folks are relying on actual philosophical principles to be repaid for what they give away, rather than license clauses...
Yes! I had almost forgotten...
He has some new books pointed squarely at younger readers, though... Hat Full of Sky and The Wee Free Men...
If they're a little older... -- Ursula LeGuin, Octavia Butler, Philip K. Dick, Zelazny's later stuff, Vonnegut, Niven, Frank Herbert, Definitely Neal Stephenson...
And the computer related classics -- The Adolescence of P1 by Ryan, The Shockwave Rider by Brunner, Neuromancer by Gibson...
And of course fantasy, Tolkein, the Shanarra books, etc. And then, only after they've read all of the usual fantasy stuff, then they can truly appreciate Pratchett...