Come on. Just googling for the result is not enough. The fact that the top result for that search is an anti-evolution website does not mean you've gone anywhere towards appropriate research.
My question is, have there been any experiments done with induced mutation, combined with natural selection, to establish benchmarks for the end-to-end evolutionary process under controlled conditions?
Sort of. There have been uses of evolution with unnatural selection to create life that did not exist before. The critter was even patented. And everyone shuddered at the implication.
First, you do not need radiation for mutations to occur. You already have an internal copy error rate of about 1 in 500,000 base pairs. And radiation has far more effects than simply screwing with DNA. Irradiating with an unusually concentrated source will do more than just mutate DNA.
Next, consider complexity. Suppose you have 1 coin, but you aren't sure if it is fair or not. How many tosses will it take you to determine this? Now, say you have 5 coins, you aren't sure if they are all fair or not, and you have to flip them all simultaneously, how many tosses does it take to determine if the coins are biased?
In living organisms there are 6 copying possibilities at each location: Correct base pair copy, mutation to one of 3 possible different bases (substitution), removal of base (deletion), adding a new base (insertion). These do not all occur with equal probability (indeed, a correct copy occurs 99.9+% of the time).
The latter 5 possibilities all open up the possibility for change. Some of these changes will occur between coding regions, in which case they do nothing. Of those that occur in a DNA coding region, there are several more possibilities: silent mutation (a substitution that does not change the amino acid being coded for), a change in amino acid being coded for, a reading from shift (an insertion or deletion that does occur as a multiple of 3).
Each one of those possibilities also has possible outcomes: No change (protein being coded for still works), partially functioning protein (anywhere from almost 100% to almost 0%), non-functioning protein (protein can no longer fold due to structurally change), differently functioning protein (protein now acts in a way slightly or drastically different than it used to).
Now consider that in a simple bacteria of say 3 million bases, there are 3 million places for any one mutation to occur. In a rat, the number of bases is over 100x more numerous.
Lastly, what do you mean by "good" mutation? Is resistance to antibiotics "good"? If a bacteria spontaneously develops antibiotic resistance, but is in a petri dish that exposes it to a lot of UV radiation, and it dies from UV exposure, has it developed a "good" mutation? That particular mutation would be "good" if the bacteria were inside a host and protected from UV, and that host were likely to take antibiotics. If that host were not likely to take antibiotics, it would be a useless mutation.
Now I'm not saying that the data you're after doesn't exist, but you are severely trivializing the amount of work it would take to obtain this data, and quantify "good" vs. "bad" mutations. And in the end, once we had it, what good would it do us?
IIRC, Grokster lost their case because they advertised themselves as a great way to get movies and music for free. Essentially marketing themselves as a conduit for copyright infringement. So much so that people were confused and actually believed they were downloading the stuff legally. (Don't start on how could someone think that, I won't argue the qualities of human ignorance.)
There are plenty of good uses for P2P. Copyright infringement, while popular, is not a "good" use.
It provides a plausible explanation for the origin of species, but has no predictive power at all.
Oh that's just nonsense.
Before the rise of antibiotic resistant bacteria their existence was predicted by evolution. Researchers knew if a single bacteria, through random mutation, developed a resistance to an antibiotic, it would have an obvious survival advantage and spread more rapidly. In several countries, if you contract a disease from a local prostitute, it's almost gauranteed to be a super-resistant strain because some genius government there thought they would be clever and give these women antibiotics as a prophylactic measure. Worked for a little while, then that damned evolution thing kicked in.
That's why HIV carriers are on a drug cocktail. It's far less likely the virus is going to develop an immunity to all the different drugs at once. If you were to give the drugs one at a time, however, evolution predicts the rise of an HIV virus that could resist them all.
Basically, the underlying philosophy of Free/Open Source Software is that it is alright to copy someone else's work, as long as you make your changes public.
Kind of a negative phrasing of it, but yes. That's the idea.
This inevitably encourages people to take the 'path of least resistance', completing their project in less time but only contributing a fraction of the total work.
Yes, we usually refer to it as "standing on the shoulders of giants." But this is in no way limited to FOSS. When I worked at a large IT company the chief software architect's mantra was "it is cheaper to buy than to build!" He wasn't real particular about FOSS, his point was if we could buy something off the shelf, we should always do that instead of making it ourselves. Putting something in the development pipeline was a slow, expensive process: Gather requirements, write up the use cases, design the software, review, do the actual coding, do code reviews, write up a test plan, send it through QA.
This laxity, or laziness, is soon seen as the norm.
This laxity or laziness most major corporations would call "efficient." Seriously, software development is expensive.
Lo and behold, when someone at a F/OSS software-heavy campus like MIT starts a project and finds that no-one has done the heavy lifting for him already, he still sees it as his right to get full reward with fractional work.
I don't follow exactly what you're saying. If someone uses Linux + Apache + MySQL + PHP to make a piece of weblog software, do they deserve no credit because LAMP provides for 99% of all the code?
It really is too bad - given all the real advantages of F/OSS, like the Open Source - but I think F/OSS should be curtailed, or even counter-legalized, in order to stop the decline in American work ethic and scientific rigour.
This is a fantastic jump of logic. From not having to pay money for a software license to falsifying data? Wow! How does paying money for a software license somehow make scientists more honest?
This sort of black mark generally means that the person will likely have to leave academia and research altogether. In research your integrity is everything. If you lie once, nobody knows if you won't lie again. Peer reviewed Journals will generally refuse your papers without reading them. No research body would risk your name going on one of their papers which would cause it to get red flagged for automatic refusal. It's a very grave situation which can't just be dismissed by "I made a mistake." The guy went through enough school to get a Ph.D., he knew what would happen if he got caught.
The function just needed to determine the type of triangle that would be formed given the length of its sides. The function took three integers and returned the type of triangle formed. A decidedly different problem from integer triangles link you posted.
Although my example was somewhat extreme, I have seen a lot of code where the original author thought it would be cool to create a pseudo-language through clever macros. The situation got a lot worse with C++ when they started mixing all that with operator overloading. As a very simple main stream example, consider:
A piece of code that appears in nearly all MFC applications. The code appears outside of a function or class definition. What does it do? What can it do? Why doesn't it look like C?
I'm not sure that x 10 is any more efficient than x = 9. Both involve a subtraction, setting status flags, ANDing the status flags and branching on the result.
I strongly suspect the idiom has more to do with pointers & arrays. c[5] is the same as *(c+5) and is the same number you'd use to access the 6th element of the array in assembly language. If you have an array of 10 chars, you'd access them as arrayAddress+0 - arrayAddress+9 in assembly, and since C and assembly have been tight throughout C's existence, C uses the assembly convention.
There is a professor at my school who teaches a class at the tail end of the CS curriculum that all CS majors must take. The problem is the guy who teaches it seems to have given up understanding of all scholarly pursuits of CS some 20 years ago. The class consists of him showing us PPT presentations and giving us articles from "experts" in such a way as to pre-suppose that we are not qualified to engage in a scholarly discussion on the merits or demerits of something he hands out. I've become so irritated with this teaching style I don't care that I annoy the prof. Upper division major courses *should* be a forum for critical thinking and exploration. So anyway...
A couple weeks ago we went through an "identify all the test cases" exercise with a simple problem: a function which classifies 3 side lengths with whichever type of triangle they could form. It was basically looking to see if you could identify boundary conditions and negative test cases.
My suggestion to test lengths of MAX_INT or (MAX_INT/2)+1 was pooh-poohed because the infallible software testing guru who wrote the exercise didn't find such a condition a necessary test.
However the obvious negative test case - three side lengths that could not form a triangle did need to be tested. Someone asks "well how do you determine that those sides could not form a triangle?" The Prof shoots back "If the sum of any two sides is less than the value of a third side." So I call out "you sure you don't want to test MAX_INT?" The prof paused a couple moments, then ignored me and continued on. The whole class let out a nervous laugh.
Apparently testing gurus don't always remember that integers overflow, and that fact shakes the foundation of some Prof's world.
Bullshit. Voting is enough. The power to remove an incumbent from office is sufficient. That's the reason why very vocal minority religious groups have co-opted the US government's domestic policy to horrible effect. That's also the reason why domestic social policies tend to favor the elderly. And lastly, that's the reason why the threat to the USA's math/science/engineering dominance will go unchallenged: too many of those who vote are either fundamentalist Christian's who distrust science, or are elderly and not concerned with problems 15-20 years in the future.
H1-B's are not supposed to be imported for the lower-tech jobs unless there is a scarcity of native talent for those lower-tech jobs. H1-B's are expected to be as trained, or more so, than their American counterparts. That's what the Benedict Arnold CEO's keep telling us - the supply of US tech talent is lacking. They have to import highly qualified people to continue innovating.
If what you say is true - H1-B's are less skilled - then the author's argument is strengthened.
Most people in the tech market today tend to feel that H1-B's are really being used to decrease salary pressure. If you're qualified but demand $80,000/yr, well they can just hire an H1-B at $60,000/yr. That's not what H1-B's are for.
Paragraph B is referring to people making copies of the seal. If you put the seal on a t-shirt, paragraph B covers you.
By your reading of paragraph B, if a reporter took a picture of the president at a press conference and then a newspaper ran the picture, they'd be violating paragraph B. I don't think that's a reasonable assessment.
Well, if we are sufficiently bored, we might read the first sentence of the article:
When you build a PC, you're often making a series of delicate trade-offs.
I'm sure some of you are hard core enough to be building your own computers from assorted Solaris bits and pieces... But I think the article was clearly aimed at people building their own PC from commodity hardware.
Factor in the time you spend dorking around with the miniitx and it gets worse.
I don't know about other Mini-ITX modders out there, but to me if you factor in the time I spend dorking around with the mini-itx, it gets better. For me, at least, there was a positive feedback loop. I enjoyed mucking with the board. If I had the time and money right now, I'd do it again because of the recreational value I got from it. I still like to point out that my home server has a peak power draw of just under 25W - and that only occurs when the unit is powered on and the hard drives have to suck extra juice to spin up.
a company that was full of innovative ideas and actually seemed to want to please its customers.
What? Are we thinking the same TSR? The TSR that put itself into bankruptcy by alienating itself from its customers - threatening any who dared post a module they made themselves with legal action citing the module as a derivative product of their IP?
OMFG, didn't you pay attention?! The Gods made the clockwork owl! c'mon we know that the Greek Pantheon of Gods is more sophisticated than we are today.
Wait! What if this computer were actually made by the Gods?? Maybe we now have irrefutable proof of their existence! Take THAT Greek God-Biatches, not so omniscient now are you? ARE YOU?!
Since that's basically a probability question, I thought I'd follow up with my own favorite probability mind bender:
Your married-with-two-kids co-worker invites you over to dinner. When you arrive a son of the coworker answers the door. What is the probability that the other child is a girl?
Followup:
The co-workers oldest child, a son, answers the door. What is the probability that the other child is a girl?
Most who have gone through a formal stats class have seen this one before, but it is always fun to try and wrap your head around it the first time.
When you picked your first door you had a 1/3 chance of picking the winning door, and a 2/3 chance of picking a losing door. When Monty revealed a booby-prize behind one of the other doors, you still had a 1/3 chance of picking the winning door and a 2/3 chance of picking a losing door. It's just that now you know if you picked the losing door, you know which of the other doors is the winner.
I guess another way of looking at it is: you had a probability of success of 0.33 when you first picked a door. How is knowing information about another door going to suddenly make your door MORE likely to be a success?
Based on the experiment, it's possible that the "Death Ray" existed, but they do point out that their experiment is rough, and also doesn't prove that it was used.
Which is particularly interesting because that's just what the MythBusters said... Only the MythBusters did a better job recreating the conditions the weapon would have been used under. I think everyone is just bitter because the MythBusters didn't ignite their ship (again, due to the improved realism of their experiment).
Slashdot Mirror
Come on. Just googling for the result is not enough. The fact that the top result for that search is an anti-evolution website does not mean you've gone anywhere towards appropriate research.
My question is, have there been any experiments done with induced mutation, combined with natural selection, to establish benchmarks for the end-to-end evolutionary process under controlled conditions?
Sort of. There have been uses of evolution with unnatural selection to create life that did not exist before. The critter was even patented. And everyone shuddered at the implication.
First, you do not need radiation for mutations to occur. You already have an internal copy error rate of about 1 in 500,000 base pairs. And radiation has far more effects than simply screwing with DNA. Irradiating with an unusually concentrated source will do more than just mutate DNA.
Next, consider complexity. Suppose you have 1 coin, but you aren't sure if it is fair or not. How many tosses will it take you to determine this? Now, say you have 5 coins, you aren't sure if they are all fair or not, and you have to flip them all simultaneously, how many tosses does it take to determine if the coins are biased?
In living organisms there are 6 copying possibilities at each location: Correct base pair copy, mutation to one of 3 possible different bases (substitution), removal of base (deletion), adding a new base (insertion). These do not all occur with equal probability (indeed, a correct copy occurs 99.9+% of the time).
The latter 5 possibilities all open up the possibility for change. Some of these changes will occur between coding regions, in which case they do nothing. Of those that occur in a DNA coding region, there are several more possibilities: silent mutation (a substitution that does not change the amino acid being coded for), a change in amino acid being coded for, a reading from shift (an insertion or deletion that does occur as a multiple of 3).
Each one of those possibilities also has possible outcomes: No change (protein being coded for still works), partially functioning protein (anywhere from almost 100% to almost 0%), non-functioning protein (protein can no longer fold due to structurally change), differently functioning protein (protein now acts in a way slightly or drastically different than it used to).
Now consider that in a simple bacteria of say 3 million bases, there are 3 million places for any one mutation to occur. In a rat, the number of bases is over 100x more numerous.
Lastly, what do you mean by "good" mutation? Is resistance to antibiotics "good"? If a bacteria spontaneously develops antibiotic resistance, but is in a petri dish that exposes it to a lot of UV radiation, and it dies from UV exposure, has it developed a "good" mutation? That particular mutation would be "good" if the bacteria were inside a host and protected from UV, and that host were likely to take antibiotics. If that host were not likely to take antibiotics, it would be a useless mutation.
Now I'm not saying that the data you're after doesn't exist, but you are severely trivializing the amount of work it would take to obtain this data, and quantify "good" vs. "bad" mutations. And in the end, once we had it, what good would it do us?
IIRC, Grokster lost their case because they advertised themselves as a great way to get movies and music for free. Essentially marketing themselves as a conduit for copyright infringement. So much so that people were confused and actually believed they were downloading the stuff legally. (Don't start on how could someone think that, I won't argue the qualities of human ignorance.)
There are plenty of good uses for P2P. Copyright infringement, while popular, is not a "good" use.
It provides a plausible explanation for the origin of species, but has no predictive power at all.
Oh that's just nonsense.
Before the rise of antibiotic resistant bacteria their existence was predicted by evolution. Researchers knew if a single bacteria, through random mutation, developed a resistance to an antibiotic, it would have an obvious survival advantage and spread more rapidly. In several countries, if you contract a disease from a local prostitute, it's almost gauranteed to be a super-resistant strain because some genius government there thought they would be clever and give these women antibiotics as a prophylactic measure. Worked for a little while, then that damned evolution thing kicked in.
That's why HIV carriers are on a drug cocktail. It's far less likely the virus is going to develop an immunity to all the different drugs at once. If you were to give the drugs one at a time, however, evolution predicts the rise of an HIV virus that could resist them all.
Fair enough, I didn't read the subject as part of the comment. But um, there is a less angst-ridden way to mention that.
Basically, the underlying philosophy of Free/Open Source Software is that it is alright to copy someone else's work, as long as you make your changes public.
Kind of a negative phrasing of it, but yes. That's the idea.
This inevitably encourages people to take the 'path of least resistance', completing their project in less time but only contributing a fraction of the total work.
Yes, we usually refer to it as "standing on the shoulders of giants." But this is in no way limited to FOSS. When I worked at a large IT company the chief software architect's mantra was "it is cheaper to buy than to build!" He wasn't real particular about FOSS, his point was if we could buy something off the shelf, we should always do that instead of making it ourselves. Putting something in the development pipeline was a slow, expensive process: Gather requirements, write up the use cases, design the software, review, do the actual coding, do code reviews, write up a test plan, send it through QA.
This laxity, or laziness, is soon seen as the norm.
This laxity or laziness most major corporations would call "efficient." Seriously, software development is expensive.
Lo and behold, when someone at a F/OSS software-heavy campus like MIT starts a project and finds that no-one has done the heavy lifting for him already, he still sees it as his right to get full reward with fractional work.
I don't follow exactly what you're saying. If someone uses Linux + Apache + MySQL + PHP to make a piece of weblog software, do they deserve no credit because LAMP provides for 99% of all the code?
It really is too bad - given all the real advantages of F/OSS, like the Open Source - but I think F/OSS should be curtailed, or even counter-legalized, in order to stop the decline in American work ethic and scientific rigour.
This is a fantastic jump of logic. From not having to pay money for a software license to falsifying data? Wow! How does paying money for a software license somehow make scientists more honest?
This sort of black mark generally means that the person will likely have to leave academia and research altogether. In research your integrity is everything. If you lie once, nobody knows if you won't lie again. Peer reviewed Journals will generally refuse your papers without reading them. No research body would risk your name going on one of their papers which would cause it to get red flagged for automatic refusal. It's a very grave situation which can't just be dismissed by "I made a mistake." The guy went through enough school to get a Ph.D., he knew what would happen if he got caught.
The function just needed to determine the type of triangle that would be formed given the length of its sides. The function took three integers and returned the type of triangle formed. A decidedly different problem from integer triangles link you posted.
Although my example was somewhat extreme, I have seen a lot of code where the original author thought it would be cool to create a pseudo-language through clever macros. The situation got a lot worse with C++ when they started mixing all that with operator overloading. As a very simple main stream example, consider:
BEGIN_MESSAGE_MAP( MFC_Window, CFrameWnd)
ON_WM_LBUTTONDOWN()
ON_WM_LBUTTONUP()
END_MESSAGE_MAP()
A piece of code that appears in nearly all MFC applications. The code appears outside of a function or class definition. What does it do? What can it do? Why doesn't it look like C?
I'm not sure that x 10 is any more efficient than x = 9. Both involve a subtraction, setting status flags, ANDing the status flags and branching on the result.
I strongly suspect the idiom has more to do with pointers & arrays. c[5] is the same as *(c+5) and is the same number you'd use to access the 6th element of the array in assembly language. If you have an array of 10 chars, you'd access them as arrayAddress+0 - arrayAddress+9 in assembly, and since C and assembly have been tight throughout C's existence, C uses the assembly convention.
As long as we're telling stupid prof stories...
There is a professor at my school who teaches a class at the tail end of the CS curriculum that all CS majors must take. The problem is the guy who teaches it seems to have given up understanding of all scholarly pursuits of CS some 20 years ago. The class consists of him showing us PPT presentations and giving us articles from "experts" in such a way as to pre-suppose that we are not qualified to engage in a scholarly discussion on the merits or demerits of something he hands out. I've become so irritated with this teaching style I don't care that I annoy the prof. Upper division major courses *should* be a forum for critical thinking and exploration. So anyway...
A couple weeks ago we went through an "identify all the test cases" exercise with a simple problem: a function which classifies 3 side lengths with whichever type of triangle they could form. It was basically looking to see if you could identify boundary conditions and negative test cases.
My suggestion to test lengths of MAX_INT or (MAX_INT/2)+1 was pooh-poohed because the infallible software testing guru who wrote the exercise didn't find such a condition a necessary test.
However the obvious negative test case - three side lengths that could not form a triangle did need to be tested. Someone asks "well how do you determine that those sides could not form a triangle?" The Prof shoots back "If the sum of any two sides is less than the value of a third side." So I call out "you sure you don't want to test MAX_INT?" The prof paused a couple moments, then ignored me and continued on. The whole class let out a nervous laugh.
Apparently testing gurus don't always remember that integers overflow, and that fact shakes the foundation of some Prof's world.
Bullshit. Voting is enough. The power to remove an incumbent from office is sufficient. That's the reason why very vocal minority religious groups have co-opted the US government's domestic policy to horrible effect. That's also the reason why domestic social policies tend to favor the elderly. And lastly, that's the reason why the threat to the USA's math/science/engineering dominance will go unchallenged: too many of those who vote are either fundamentalist Christian's who distrust science, or are elderly and not concerned with problems 15-20 years in the future.
H1-B's are not supposed to be imported for the lower-tech jobs unless there is a scarcity of native talent for those lower-tech jobs. H1-B's are expected to be as trained, or more so, than their American counterparts. That's what the Benedict Arnold CEO's keep telling us - the supply of US tech talent is lacking. They have to import highly qualified people to continue innovating.
If what you say is true - H1-B's are less skilled - then the author's argument is strengthened.
Most people in the tech market today tend to feel that H1-B's are really being used to decrease salary pressure. If you're qualified but demand $80,000/yr, well they can just hire an H1-B at $60,000/yr. That's not what H1-B's are for.
Paragraph B is referring to people making copies of the seal. If you put the seal on a t-shirt, paragraph B covers you.
By your reading of paragraph B, if a reporter took a picture of the president at a press conference and then a newspaper ran the picture, they'd be violating paragraph B. I don't think that's a reasonable assessment.
Well, if we are sufficiently bored, we might read the first sentence of the article:
When you build a PC, you're often making a series of delicate trade-offs.
I'm sure some of you are hard core enough to be building your own computers from assorted Solaris bits and pieces... But I think the article was clearly aimed at people building their own PC from commodity hardware.
Wow, your server and mine seem to have a lot in common.
I should charge royalties on that comment.
Factor in the time you spend dorking around with the miniitx and it gets worse.
I don't know about other Mini-ITX modders out there, but to me if you factor in the time I spend dorking around with the mini-itx, it gets better. For me, at least, there was a positive feedback loop. I enjoyed mucking with the board. If I had the time and money right now, I'd do it again because of the recreational value I got from it. I still like to point out that my home server has a peak power draw of just under 25W - and that only occurs when the unit is powered on and the hard drives have to suck extra juice to spin up.
a company that was full of innovative ideas and actually seemed to want to please its customers.
What? Are we thinking the same TSR? The TSR that put itself into bankruptcy by alienating itself from its customers - threatening any who dared post a module they made themselves with legal action citing the module as a derivative product of their IP?
Is this how one pleases one's customers?
I'd kill for a good, trustworthy carpenter.
WWJD?!
OMFG, didn't you pay attention?! The Gods made the clockwork owl! c'mon we know that the Greek Pantheon of Gods is more sophisticated than we are today.
Wait! What if this computer were actually made by the Gods?? Maybe we now have irrefutable proof of their existence! Take THAT Greek God-Biatches, not so omniscient now are you? ARE YOU?!
Unfortunately no. The two questions have different answers.
Since that's basically a probability question, I thought I'd follow up with my own favorite probability mind bender:
Your married-with-two-kids co-worker invites you over to dinner. When you arrive a son of the coworker answers the door. What is the probability that the other child is a girl?
Followup:
The co-workers oldest child, a son, answers the door. What is the probability that the other child is a girl?
Most who have gone through a formal stats class have seen this one before, but it is always fun to try and wrap your head around it the first time.
When you picked your first door you had a 1/3 chance of picking the winning door, and a 2/3 chance of picking a losing door. When Monty revealed a booby-prize behind one of the other doors, you still had a 1/3 chance of picking the winning door and a 2/3 chance of picking a losing door. It's just that now you know if you picked the losing door, you know which of the other doors is the winner.
I guess another way of looking at it is: you had a probability of success of 0.33 when you first picked a door. How is knowing information about another door going to suddenly make your door MORE likely to be a success?
Based on the experiment, it's possible that the "Death Ray" existed, but they do point out that their experiment is rough, and also doesn't prove that it was used.
Which is particularly interesting because that's just what the MythBusters said... Only the MythBusters did a better job recreating the conditions the weapon would have been used under. I think everyone is just bitter because the MythBusters didn't ignite their ship (again, due to the improved realism of their experiment).