> Your hardware analogy is more appropriate to using code libraries rather than > implementing everything from scratch, which no-one is arguing against here, > as far as I can tell.
Well, be fair, that's because it wasn't the subject. Some libraries are harder to figure out than writing from scratch... for example, if you need to draw graphs you can write your own [in OpenGL, X, DirectX, awt, whatever] or you can find a graph-drawing library.. which might be appropriate, or might end up being more work to figure out than just doing it yourself.
The larger the project the more likely it is to be worth implementing things yourself rather than relying on a library. If it takes 2 weeks to write it yourself v. 2 days to figure out how to use a library, you use the library right ? Then months down the line you hit a bug or a missing feature that turns out to be part of the library, and you dont have the source, or even if you do, you dont understand it properly and its harder to change than just writing your own from scratch with your in house coding practices etc.
Are you out of your mind ? Just how much time do you expect students to spend working on this course ?
Ant is a painful necessity for professional java programmers. Its only useful or necessary with large projects. This an intro course, they'll have quite enough to worry about learning the basics of OO, java syntax and the java language. How many fricking classes do you need to deal with in java to read from a file: [File ?.. oh, InputStream.. oh FileInputStream, great, no wait.. maybe FileReader but then better use BufferedReader, oh wait, I need to deal with IOException too, but what should I do with that, use throws or rethrow RuntimeException] ? Seriously, they've got enough to worry about getting to grips with the language. Learning bloody ant is too much.
> We didn't attack Iraq, we attacked its government. > There is a huge difference.
No, there really isn't. In no war that I can think of did the aggressors not claim they were attacking the regime rather that the country or its people. eg Germany claimed, and even somewhat believed, that when they invaded Russia they would be greated as liberators for freeing the Russian people from Stalin's yoke. [btw, fuck Godwin]
> FYI, you forget a couple of things: the term "illegal" > has no meaning outside the context of a nation, > and "war crimes" is supposed to describe actual crimes > committed during wartime, not simply an unjust war itself. > Tone down the rhetoric.
Part of a nations law depends on the treaties that it has signed up for. A signed treaty is considered by the constitution to be legally binding so breaking it is illegal.
> "war crimes" is supposed to describe actual crimes > committed during wartime, not simply an unjust war itself.
Wrong again, launching a war of agression is the number one war crime from which all others flow.
Bring it in yourself dammit. Jeez, no wonder Italy has no software industry to speak of if that's the attitude. If there are all these qualified creative under utilized engineers sitting around looking for something more interesting to do that just sceams *opportunity*. If one the other hand you have a bunch of people waiting for someone to come and offer them a more interesting job, carry on waiting.
As you say the PC was invented to return the power to the user. The idea was it would be simple enough for people to administer their own machines. In small informal environments it even works like that.
> The fact is that few companies sell IT as a final product
True, but if having a working computer system is necessary for people to do their jobs, then IT is contributing to the end product as much as anyone else. Reading between the lines, I can see you have encountered IT people with bad attitudes, but your own is a bit suspect: I get a hint of "we're the ones doing the useful work, IT is there to serve us"... truth is, if you need IT then you're working together. Attitudes don't come from nowhere: if the IT department acts like dicks, chances are they feel stressed and put-upon and it shows. Alternatively, you have some assholes working in the IT department, but asshole distribution is normally pretty uniform.
B52s were used extensively during "shock and awe" campaign to destroy a bunch of targets in Iraq. The title of the campaign might give you a little clue as to what the intent was, ie to terrorise Iraq into submission. B52s are not precision tools and the vast majority of the bombs dropped were not "smart".. they were just bombs. You don't need to explicity "target" civilians in order to kill them.
Secondly, why is it only civilian death that counts ? Does the fact that someone was conscripted into the Iraqi army mean they deserve to die ? In particular, does it mean its OK to kill them in their own country by dropping bombs from planes ?
At Edinburgh we started with Pascal, and moved onto assembler later which was the wrong way round IMHO. The most illuminating moment I can remember was learning to design basic logic circuits with nand gates - it got rid of the "and then something magic happens" feeling of unease I had had beforehand. Fundamentals of circuit design followed by assembler and then some structured language [like C] would have been the right way to start.
Starting with VB or even java is very wrong - there's far too much to learn that has nothing to do with the heart of the matter and is just a matter of language specifics, environement, libraries and tools. It doesn't matter how proficient you get in OO languages and fancy IDEs, you'll never have the proper depth of understanding until you understand the basics of assembler.
Computers are all the same - some of them have more memory than others and some of them are speedier than others but they are functionally the same. Given enough time and memory, any computer can simulate any other computer. Intelligence and speed are separate concepts. Imagine we had a chat with some aliens who lived 10 light years away. There answers would be no less intelligent even though we had to wait 20 years to receive a reply.
So, intelligence is no more likely to emerge all of a sudden from the latest supercomputer than from that rusty old pentium you've been using as a firewall. Either human brains are not Turing machines, or the hardware is sufficient to host super intelligence already. Either way, the software has got a long way to go. Kurzweil believes we'll be there in 30 years, but his arguments about software advances [and brain reverse engineering] were somewhat unconvincing.
> when the monitored conversations occured between foriegn, self-proclaimed enemies of the United States who are engaged in armed conflict with us, and people inside the United States?
Who says ? The administration certainly likes to imply those were the only conversations listened to, but Gonzales went out of his way to avoid confirming this.
> I question your patriotism.
And I question yours. If being an American means anything, it means respect for the constitution. Trying to justify the efforts of a president to remove the protections in the constitution brands you as a traitor to the republic.
> In computers, most often, you'll find that each gene directly influences one particular aspect of the solution.
Give me a little credit - I wasnt doing this for fun. You're right, the mapping of a solution description to a format that actually has a chance of being solved via GAs is the most difficult and critical part of the problem. You can try to encode the solution directly by encoding the design parameters directly into the genome, and this is the kind of half assed solution described in some clueless books, but you will never solve remotely tricky problems like that. In situations where the solution is less obvious one needs to encode the problem as a custom designed set of instructions for building solutions to the problem and then use those as the starting point.
> Natural selection does not create complexity
Then what does create the complexity ? DNA is closer to a set of instructions for building an organism than it is the actual organism, but so what ? Natural selection is supposedly the mechanism by which the design problems actually solved
> Sopwith to jet fighter ain't that difficult
Then show me how to solve *that* design problem with a GA and I'll believe that life can be solved with a GA.
> PS: If you really think that a genetic algorithm with a fixed, hand-defined fitness function provides an acceptable picture of biological evolution "in the wild", then I'm sorry but NO, you DON'T understand artificial evolution "better than all but a handful of people on this planet."
What the fuck are you talking about ? I never suggested anything of the sort. What I said was "life experience is equivalent to the evaluation function" ie the richness of life experience is not supposed to be a part of the design process according to evolutionary theory. Eg, it doesnt a snake won a battle with a wombat because there is no mechansim to use that experience to improve the design of the next generation. The only thing that is supposed to count is the fact that surviving the battle made it more likely to produce offspring.
In industry we call people with PhDs "junior". I eat PhDs for breakfast;)
Seriously though, it should occur to you that I wasnt in a position to do this kind of work and get paid [very nicely] for it entirely through chance. Sure, there are thousands of people with PhDs in GA research, I know some too. I stand by what I said though - only the very most capable and genuninely useful of those thousands of people will ever get a full time paid position using GAs in real world situations.
Yes, I do see the difference, and you're correct to a large extent. However, at some stage a bunch of very difficult design problems were solved.. and the theory is that this occured via natural selection. Now, it's certainly plausible that this happened, but its not the slam dunk the likes of Dawkins would have us believe. My own experience in this matter suggests to me that we are so far missing something fundamental in our understanding.
Rapid evolutionary responses can certainly occur, even in 3 generations, but this is not solving a design problem - they just show that if the good design already exists in the population it will rapidly take over if conditions necessitate that. I've read Dawkins.. he's an interesting writer but he doesn't understand probability.
Depends what you mean by evolutionary theory. Creatures evolve, sure, you won't find many science "believers" disagreeing with you there. What's a lot less obvious is that evolution occurs solely due to natural selection/sexual selection, crossover and mutation or any other explanation that "evolutionary theory" has so far come up with.
I used to think natural selection was a sufficient explanation for life. What cured me was I worked fulltime for two years using genetic algorithms to solve a hard problem [it was about efficiently scheduling orders through a large metal rolling factory to optimize throughput]. I had several previous years experience in all kinds of optimization problems - [I had used GA's before in toy situations, but also had extensive experience with simulated annealing and numerous types of numerical optimization problems].
I was successful in this project, ie I eventually produced a program that produced better schedules than the (intelligent) fulltime, experienced humans who used to produce the weeks schedule using their own knowledge and experience along with various supporting computer tools. The heart of the solution was a GA coupled with various tricks used whereever beneficial. It helped that I had an Origin 2000 at my disposal. GAs really do work, and not wishing to sound arrogant, I understand that process and what they're capable of (from a computational potential perspective) better than all but a handful of people on this planet.
But... as a result of all this, I don't understand evolution anymore. It doesn't add up. From an optimization perspective, the problems I was solving where something like 10^100 times less difficult than the problems solved by evolution. Genetic algorithms run into a bit of a wall beyond a certain point.
Now, I'll grant you that I didnt have several hundred million years to work on the problem. However, I did experiment with population sizes in the 100'000s and I did evolve them for 10,000s of generations. Those are small numbers when considering evolution of bacteria, but they're pretty realistic numbers when considering evolution from an Ape to a human (or some equally fit alternative). Remember, according to evolutionary theory, the life events of an individual are irrelevent.All that matters is whether it produces offspring (also some evolutionary advantage to protecting close relatives, but even that can be simulated quite easily), and supposedly the mechanism is crossover and mutation, ie in a GA, life experience is equivalent to the evaluation function.
Based on my experience, I dont believe you could solve a problem such as the development of an equivalent to an F16 starting with a sopwith camel using GAs unless you used [at least] tens of millions of individuals over millions of generations. It seems to me that evolutionary problems are solved *much* faster than one has any right to expect. Apes evolved into modern humans with populations in the hundred thousands over tens of thousands of generations. I know that humans were just the design we ended up with, not the goal, but.. think in terms of the creation of a creature as much more "fit" [from evolutionary perspsective] as a human to an ape.
So, anyway.. I dont believe in "evolutionary theory" as a sufficent explanation anymore. I don't believe in intelligent design either, but I dont blame people who do
You're right there. People think they understand it, when really they don't. Once people understand the idea of how a randomly directed process can produce results that are far from random they click and go wow.. I guess it could really work. I was at that stage for many years.
What cured me was I worked fulltime for two years using genetic algorithms to solve a hard problem [it was about efficiently scheduling orders through a large metal rolling factory to optimize throughput]. I had several previous years experience in all kinds of optimization problems - [I had used GA's before in toy situations, but also had extensive experience with simulated annealing and numerous types of numerical optimization problems].
I was successful in this project, ie I eventually produced a program that produced better schedules than the (intelligent) fulltime, experienced humans who used to produce the weeks schedule using their own knowledge and experience along with various supporting computer tools. The heart of the solution was a GA coupled with various tricks used whereever beneficial. It helped that I had an Origin 2000 at my disposal. GAs really do work, and not wishing to sound arrogant, I understand that process and what they're capable of (from a computational potential perspective) better than all but a handful of people on this planet.
But... as a result of all this, I don't understand evolution anymore. It doesn't add up. From an optimization perspective, the problems I was solving where something like 10^100 times less difficult than the problems solved by evolution.
Now, I'll grant you that I didnt have several hundred million years to work on the problem. However, I did experiment with population sizes in the 100'000s and I did evolve them for 10,000s of generations. Those are small numbers when considering evolution of bacteria, but they're pretty realistic numbers when considering evolution from an Ape to a human (or some equally fit alternative). Remember, according to evolutionary theory, the life events of an individual are irrelevent.All that matters is whether it produces offspring (also some evolutionary advantage to protecting close relatives, but even that can be simulated quite easily), and supposedly the mechanism is crossover and mutation, ie in a GA, life experience is equivalent to the evaluation function.
Based on my experience, I dont believe you could solve a problem such as the development of an equivalent to an F16 starting with a sopwith camel using GAs unless you used [at least] tens of millions of individuals over millions of generations. It seems to me that evolutionary problems are solved *much* faster than one has any right to expect. Apes evolved into modern humans with populations in the hundred thousands over tens of thousands of generations. I know that humans were just the design we ended up with, not the goal, but.. think in terms of the creation of a creature as much more "fit" [from evolutionary perspsective] as a human to an ape.
So, anyway.. I dont believe in evolution as any kind of explanation anymore. I don't believe in intelligent design either, but I dont blame people who do.
Learn some statistics. 2000 is a respectable number, its how the sample is selected which makes the difference. One can accurately access the views of a billiion people (ie within 1 % accurary) with a sample of 2000 as long as the sample is [a] truly random in the sense that everyone has equal chance of being picked, and [b] people have no particular reason to lie. [a] is the hard one.
1. Since there will always be terrorism, there will always be a war on terrorism, so none of these measures are temporary "wartime" measures. They are permanent.
2. The wiretapping argument is not about foreign intelligence, it is about domestic intelligence
3. FISA warrants can be obtained retroactively so arguing there may be no time is idiotic
4. Will you be as in favour as unfettered presidential power once a democrat is in power or are you just a partisan apologist hack with no principles ?
> if the bit of creationism undermines the entire scientific method?
If a theory is so fragile it loses credibility upon the slightest hint of another possibility maybe it deserves it.
If the evidence for evolution is covered in any reasonable way, it will survive encounters with a bit of creationism. The idea that all evolution came about through survival of the fittest or sexual selection though is considerably less obvious.
Slashdot Mirror
> Your hardware analogy is more appropriate to using code libraries rather than
> implementing everything from scratch, which no-one is arguing against here,
> as far as I can tell.
Well, be fair, that's because it wasn't the subject. Some libraries are
harder to figure out than writing from scratch... for example, if you need
to draw graphs you can write your own [in OpenGL, X, DirectX, awt, whatever]
or you can find a graph-drawing library.. which might be appropriate, or
might end up being more work to figure out than just doing it yourself.
The larger the project the more likely it is to be worth implementing things
yourself rather than relying on a library. If it takes 2 weeks to write
it yourself v. 2 days to figure out how to use a library, you use the library right ? Then months down the line you hit a bug or a missing feature that turns out to be part of the library, and you dont have the source, or even if
you do, you dont understand it properly and its harder to change than just
writing your own from scratch with your in house coding practices etc.
So anyway, even this subject isnt cut and dried.
Are you out of your mind ? Just how much time do you expect students to spend working on this course ?
.. oh FileInputStream, great, no wait.. maybe FileReader but then better use BufferedReader, oh wait, I need to deal with IOException too, but what should I do with that, use throws or rethrow RuntimeException] ? Seriously, they've got enough to worry about getting to grips with the language. Learning bloody ant is too much.
Ant is a painful necessity for professional java programmers. Its only useful or necessary with large projects. This an intro course, they'll have quite enough to worry about learning the basics of OO, java syntax and the java language. How many fricking classes do you need to deal
with in java to read from a file: [File ?.. oh, InputStream
> We didn't attack Iraq, we attacked its government.
> There is a huge difference.
No, there really isn't. In no war that I can think of did
the aggressors not claim they were attacking the regime
rather that the country or its people. eg Germany claimed,
and even somewhat believed, that when they invaded Russia
they would be greated as liberators for freeing the Russian
people from Stalin's yoke. [btw, fuck Godwin]
> FYI, you forget a couple of things: the term "illegal"
> has no meaning outside the context of a nation,
> and "war crimes" is supposed to describe actual crimes
> committed during wartime, not simply an unjust war itself.
> Tone down the rhetoric.
Part of a nations law depends on the treaties that it has signed up for.
A signed treaty is considered by the constitution to be legally binding
so breaking it is illegal.
> "war crimes" is supposed to describe actual crimes
> committed during wartime, not simply an unjust war itself.
Wrong again, launching a war of agression is the number one war
crime from which all others flow.
http://en.wikipedia.org/wiki/List_of_war_crimes
What is it, opposite day ?
The Last Samurai [aka Dances with wolves II: Japan] shouldnt even
have been a talkie. High res is moving in the wrong direction.
Impact me harder baby
> Bring in the innovation!
Bring it in yourself dammit. Jeez, no wonder Italy
has no software industry to speak of if that's the attitude.
If there are all these qualified creative under utilized engineers
sitting around looking for something more interesting to do
that just sceams *opportunity*. If one the other hand
you have a bunch of people waiting for someone to come
and offer them a more interesting job, carry on waiting.
Heh, interesting to see a different perspective.
As you say the PC was invented to return the power to the user. The idea was it would be simple enough for people to administer their own machines. In small informal environments it even works like that.
> The fact is that few companies sell IT as a final product
True, but if having a working computer system is necessary for people to do their jobs, then IT is contributing to the end product as much as anyone else. Reading between the lines, I can see you have encountered IT people with bad attitudes, but your own is a bit suspect: I get a hint of "we're the ones doing the useful work, IT is there to serve us"... truth is, if you need IT then you're working together. Attitudes don't come from nowhere: if the IT department acts like dicks, chances are they feel stressed and put-upon and it shows. Alternatively, you have some assholes working in the IT department, but asshole distribution is normally pretty uniform.
B52s were used extensively during "shock and awe" campaign to destroy a bunch of targets in Iraq. The title of the campaign might give you a little clue as to what the intent was, ie to terrorise Iraq into submission. B52s are not precision tools and the vast majority of the bombs dropped were not "smart".. they were just bombs. You don't need to explicity "target" civilians in order to kill them.
Secondly, why is it only civilian death that counts ?
Does the fact that someone was conscripted into the Iraqi army mean they deserve to die ? In particular, does it mean its OK to kill them in their own country by dropping bombs from planes ?
Finally some sense.
At Edinburgh we started with Pascal, and moved onto assembler later
which was the wrong way round IMHO. The most illuminating moment I can
remember was learning to design basic logic circuits with nand gates - it
got rid of the "and then something magic happens" feeling of unease I had
had beforehand. Fundamentals of circuit design
followed by assembler and then some structured language [like C]
would have been the right way to start.
Starting with VB or even java is very wrong - there's far too much to learn
that has nothing to do with the heart of the matter and is just a matter of
language specifics, environement, libraries and tools. It doesn't matter
how proficient you get in OO languages and fancy IDEs, you'll never
have the proper depth of understanding until you
understand the basics of assembler.
Computers are all the same - some of them have more memory than others and some of them are speedier than others but they are functionally the same. Given enough time and memory, any computer can simulate any other computer. Intelligence and speed are separate concepts. Imagine we had a chat with some aliens who lived 10 light years away. There answers would be no less intelligent even though we had to wait 20 years to receive a reply.
So, intelligence is no more likely to emerge all of a sudden from the latest supercomputer than from that rusty old pentium you've been using as a firewall. Either human brains are not Turing machines, or the hardware is sufficient to host super intelligence already. Either way, the software has got a long way to go. Kurzweil believes we'll be there in 30 years, but his arguments about software advances [and brain reverse engineering] were somewhat unconvincing.
> The USA never had a chapter of history as dark as that
You might find a few native Americans who disagree with you on that.
> they make good belts.
Really ? I know crocadiles make good belts, but I didnt know you could, er, cut out the middle man.
> when the monitored conversations occured between foriegn, self-proclaimed enemies of the United States who are engaged in armed conflict with us, and people inside the United States?
Who says ? The administration certainly likes to imply those were the only conversations listened to, but Gonzales went out of his way to avoid confirming this.
> I question your patriotism.
And I question yours. If being an American means anything, it means respect for the constitution. Trying to justify the efforts of a president to remove the protections in the constitution brands you as a traitor to the republic.
> In computers, most often, you'll find that each gene directly influences one particular aspect of the solution.
Give me a little credit - I wasnt doing this for fun. You're right, the mapping of a solution description to a format that actually has a chance of being solved via GAs is the most difficult and critical part of the problem. You can try to encode the solution directly by encoding the design parameters directly into the genome, and this is the kind of half assed solution described in some clueless books, but you will never solve remotely tricky problems like that. In situations where the solution is less obvious one needs to encode the problem as a custom designed set of instructions for building solutions to the problem and then use those as the starting point.
> Natural selection does not create complexity
Then what does create the complexity ? DNA is closer to a set of instructions for building an organism than it is the actual organism, but so what ? Natural selection is supposedly the mechanism by which the design problems actually solved
> Sopwith to jet fighter ain't that difficult
Then show me how to solve *that* design problem with a GA and I'll believe that life can be solved with a GA.
> PS: If you really think that a genetic algorithm with a fixed, hand-defined fitness function provides an acceptable picture of biological evolution "in the wild", then I'm sorry but NO, you DON'T understand artificial evolution "better than all but a handful of people on this planet."
What the fuck are you talking about ? I never suggested anything of the sort.
What I said was "life experience is equivalent to the evaluation function" ie the richness of life experience is not supposed to be a part of the design process according to evolutionary theory. Eg, it doesnt a snake won a battle with a wombat because there is no mechansim to use that experience to improve the design of the next generation. The only thing that is supposed to count is the fact that surviving the battle made it more likely to produce offspring.
I know, I know.. sorry, but the ape is 3 letters and ape-like hominid is kinda long winded.
In industry we call people with PhDs "junior". I eat PhDs for breakfast ;)
Seriously though, it should occur to you that I wasnt in a position to do this kind of work and get paid [very nicely] for it entirely through chance. Sure, there are thousands of people with PhDs in GA research, I know some too. I stand by what I said though - only the very most capable and genuninely useful of those thousands of people will ever get a full time paid position using GAs in real world situations.
Yes, I do see the difference, and you're correct to a large extent. However, at some stage a bunch of very difficult design problems were solved.. and the theory is that this occured via natural selection. Now, it's certainly plausible that this happened, but its not the slam dunk the likes of Dawkins would have us believe. My own experience in this matter suggests to me that we are so far missing something fundamental in our understanding.
Rapid evolutionary responses can certainly occur, even in 3 generations, but this is not solving a design problem - they just show that if the good design already exists in the population it will rapidly take over if conditions necessitate that. I've read Dawkins .. he's an interesting writer but he doesn't understand probability.
Depends what you mean by evolutionary theory. Creatures evolve, sure, you won't find many science "believers" disagreeing with you there. What's a lot less obvious is that evolution occurs solely due to natural selection/sexual selection, crossover and mutation or any other explanation that "evolutionary theory" has so far come up with.
I used to think natural selection was a sufficient explanation for life. What cured me was I worked fulltime for two years using genetic algorithms to solve a hard problem [it was about efficiently scheduling orders through a large metal rolling factory to optimize throughput]. I had several previous years experience in all kinds of optimization problems - [I had used GA's before in toy situations, but also had extensive experience with simulated annealing and numerous types of numerical optimization problems].
I was successful in this project, ie I eventually produced a program that produced better schedules than the (intelligent) fulltime, experienced humans who used to produce the weeks schedule using their own knowledge and experience along with various supporting computer tools. The heart of the solution was a GA coupled with various tricks used whereever beneficial. It helped that I had an Origin 2000 at my disposal. GAs really do work, and not wishing to sound arrogant, I understand that process and what they're capable of (from a computational potential perspective) better than all but a handful of people on this planet.
But... as a result of all this, I don't understand evolution anymore. It doesn't add up. From an optimization perspective, the problems I was solving where something like 10^100 times less difficult than the problems solved by evolution. Genetic algorithms run into a bit of a wall beyond a certain point.
Now, I'll grant you that I didnt have several hundred million years to work on the problem. However, I did experiment with population sizes in the 100'000s and I did evolve them for 10,000s of generations. Those are small numbers when considering evolution of bacteria, but they're pretty realistic numbers when considering evolution from an Ape to a human (or some equally fit alternative). Remember, according to evolutionary theory, the life events of an individual are irrelevent.All that matters is whether it produces offspring (also some evolutionary advantage to protecting close relatives, but even that can be simulated quite easily), and supposedly the mechanism is crossover and mutation, ie in a GA, life experience is equivalent to the evaluation function.
Based on my experience, I dont believe you could solve a problem such as the development of an equivalent to an F16 starting with a sopwith camel using GAs unless you used [at least] tens of millions of individuals over millions of generations. It seems to me that evolutionary problems are solved *much* faster than one has any right to expect. Apes evolved into modern humans with populations in the hundred thousands over tens of thousands of generations. I know that humans were just the design we ended up with, not the goal,
but.. think in terms of the creation of a creature as much more "fit" [from evolutionary perspsective] as a human to an ape.
So, anyway.. I dont believe in "evolutionary theory" as a sufficent explanation anymore. I don't believe in intelligent design either, but I dont blame people who do
You're right there. People think they understand it, when really they don't. Once people understand the idea of how a randomly directed process can produce results that are far from random they click and go wow.. I guess it could really work. I was at that stage for many years.
What cured me was I worked fulltime for two years using genetic algorithms to solve a hard problem [it was about efficiently scheduling orders through a large metal rolling factory to optimize throughput]. I had several previous years experience in all kinds of optimization problems - [I had used GA's before in toy situations, but also had extensive experience with simulated annealing and numerous types of numerical optimization problems].
I was successful in this project, ie I eventually produced a program that produced better schedules than the (intelligent) fulltime, experienced humans who used to produce the weeks schedule using their own knowledge and experience along with various supporting computer tools. The heart of the solution was a GA coupled with various tricks used whereever beneficial. It helped that I had an Origin 2000 at my disposal. GAs really do work, and not wishing to sound arrogant, I understand that process and what they're capable of (from a computational potential perspective) better than all but a handful of people on this planet.
But... as a result of all this, I don't understand evolution anymore. It doesn't add up. From an optimization perspective, the problems I was solving where something like 10^100 times less difficult than the problems solved by evolution.
Now, I'll grant you that I didnt have several hundred million years to work on the problem. However, I did experiment with population sizes in the 100'000s and I did evolve them for 10,000s of generations. Those are small numbers when considering evolution of bacteria, but they're pretty realistic numbers when considering evolution from an Ape to a human (or some equally fit alternative). Remember, according to evolutionary theory, the life events of an individual are irrelevent.All that matters is whether it produces offspring (also some evolutionary advantage to protecting close relatives, but even that can be simulated quite easily), and supposedly the mechanism is crossover and mutation, ie in a GA, life experience is equivalent to the evaluation function.
Based on my experience, I dont believe you could solve a problem such as the development of an equivalent to an F16 starting with a sopwith camel using GAs unless you used [at least] tens of millions of individuals over millions of generations. It seems to me that evolutionary problems are solved *much* faster than one has any right to expect. Apes evolved into modern humans with populations in the hundred thousands over tens of thousands of generations. I know that humans were just the design we ended up with, not the goal,
but.. think in terms of the creation of a creature as much more "fit" [from evolutionary perspsective] as a human to an ape.
So, anyway.. I dont believe in evolution as any kind of explanation anymore. I don't believe in intelligent design either, but I dont blame people who do.
Learn some statistics. 2000 is a respectable number, its how the sample is selected which makes the difference. One can accurately access the views of a billiion people (ie within 1 % accurary) with a sample of 2000 as long as the sample is [a] truly random in the sense that everyone has equal chance of being picked, and [b] people have no particular reason to lie. [a] is the hard one.
1. Since there will always be terrorism, there will always be a war on terrorism, so none of these measures are temporary "wartime" measures. They are permanent.
2. The wiretapping argument is not about foreign intelligence, it is about domestic intelligence
3. FISA warrants can be obtained retroactively so arguing there may be no time is idiotic
4. Will you be as in favour as unfettered presidential power once a democrat is in power or are you just a partisan apologist hack with no principles ?
5. Ok, I admit it, 4 wasn't really a question
> Could it possibly be because the software has also improved by orders of magnitude as well ?
Nah, that's not it
well, one change is we have to add BS to get past the filters
> if the bit of creationism undermines the entire scientific method?
If a theory is so fragile it loses credibility upon the slightest hint of another possibility maybe it deserves it.
If the evidence for evolution is covered in any reasonable way, it will survive encounters with a bit of creationism. The idea that all evolution came about through survival of the fittest or sexual selection though is considerably less obvious.