Re:haskell for the masses? sure, but only...
on
OCaml For the Masses
·
· Score: 1
It depends. I'm not talking about simple read/write to console, or even redirected equivalent. Yeah, that they could all do. But many languages seem to feel that "stream" is a sufficient concept. It is, indeed, a very flexible concept that can deal with many situations. But it doesn't handle others at all well. (And read-write stream is a violation of the idea that made streams a preferable abstraction over files. Other reasons have since been added, but the restriction to readable OR writable was a big part of the original concept.)
When I just checked yesterday OCaML, or possibly it was Haskell, had the ability to write single bytes, or built in types (short, long, etc.). But there was no indication that one could build things up in buffers and write that. The other could do that, but it required building things up in a mutable byte string, byte by byte, and then converting it to a immutable byte string. And I STILL have no idea how much disk space writing a buffer containing n bytes would require. Apparently some type information gets written at the same time (but I'm not even really certain of that). This makes it difficult to plan a random access IO file. And one comment I ran across made it seem that every time I tried to access the length of the byte string, it would traverse the entire string, counting bytes. This *can't* be correct, but that's what it looked like.
Re:haskell for the masses? sure, but only...
on
OCaml For the Masses
·
· Score: 1
No. Haskell has single element arrays that contain references to one item. When you "change" that item what you are doing is replacing the old reference with a new one. I believe items holding the old reference continue to hold it. But the items are just as immutable as Java strings. You can't change them, you can only replace them.
Re:haskell for the masses? sure, but only...
on
OCaml For the Masses
·
· Score: 3, Insightful
You misunderstand thing. General purpose languages, such as the C or Algol family (which includes, among others Java, Python, and Ruby) do, indeed, include concepts from functional languages. But they don't RESTRICT you to those concepts. It's the insistence that everything be unchanging that is the weakness in the purely functional languages. The Scheme family is the only group of that to have any measurable use. (Note that Lisp is NOT a functional language, even though the concept originated via a subset of Lisp.)
There's much to be said for having the ability to have invariant elements in functions. There's very little in having the requirement that everything be that way. Note that many existing languages that have invariant strings ALSO have mutable strings. Java to name just the most popular example. So by all means allow functions to declare their values invariant. This allows lots of different kinds of optimization. But don't insist that all functions do so. And definitely don't insist that everything be invariant.
Re:haskell for the masses? sure, but only...
on
OCaml For the Masses
·
· Score: 1
That definitely depends on what you're doing. Every time I've looked at a functional language I've bounced. Usually over random access read-write files. IIRC, even doing ordinary IO in Haskell (or possibly it was OCaML) required dropping into C. (Monads have no state, or some such.)
If I were to choose a functional language it would be one of the Scheme family...unless there's something that I really don't understand about Haskell or OCaML. But I have a hard time mapping Scheme variables into byte layouts in a random access file. Still, it is, in principle, doable. Not sure it's doable even in principle in Haskell or OCaML.
FWIW, just yesterday I was looking at Oz (Mozart) again. (Another strange languge, I think it's functional, but their emphasis is on constraint programming.) I bounced on their lack of file IO capability. (This time all I could find documented was text streams that were either input or output, not both.) Well, Oz has an excuse. It's design was finalized over a decade ago, and it's doesn't appear intended for other than academic use.
As far as I'm concerned, though, you can send your monads back to Leibnitz. He got some mileage out of them, but they sure don't appear worth the trouble to me. Except for a severely limited range of problems...and even then there's usually a better way to do things. Final variables, e.g. (I'd say immutable, but I'd mean the way D (Digital Mars D) uses them, which I don't really understand, and which most people won't even have been exposed to. Basically you assign the variable a value at creation time, and it can never change. But immutability can't be cast away, and creation time isn't compile time, so it's not the same as const.or, I think, final.)
The difference is that a tax on pollution can be enforced fairly. Cap and trade, on more than a local scale, can't. Whether you want to call that a weakness in Cap and trade depends on your goals.
He didn't say mislead. That's something you added. He said they will study what gets funded. This is obviously correct, as to study something requires money.
It's also true, however, that there are many individual scientists that have shaded their studies, or neglected to publish studies that came to the "wrong" answer. Elsevier published an entire medical journal that was staffed, edited, and written for by the employees of one drug company. Many of those employees were scientists. All the articles reported favorably on the products of the company. O, and they were reviewed by reviewers that were beholden in some way to that company. (Often as employees.) They *may* all have been honest, I don't know.
Note that this would not have been disreputable (not the parts that I know) if it had come out of the PR agency of that company. But it was ostensibly an independent journal, and the only company's name on it was Elsevier.
Well, that's what they publicly claim. Believe it if you want to. But to do that you've got to close your eyes to their actions.
I think I'll wait another decade or so before trusting MS again. Then I'll see what their track record is. Their current one is decidedly rancid, based on their actions as reported in the news media, rather than just their PR announcements. (Yes, the news media does include their PR announcements. And only occasionally identifies them as such. But one can also find reports by unaffiliated individuals. Like the companies that attempted to attend the committee meeting in Portugal on whether or not to accept the MS proposed word processor document standard.)
Just because something is large and threatening doesn't mean it's unnatural.
FWIW, I *do* believe that anthropogenic climate warming is well established. But it sure wasn't *this* kind of argument that convinced me. By this argument you could say that the Tsunami that recently hit Japan wasn't natural...*IF* you accepted the form of reasoning as valid. Which it isn't.
Most times it seems like they should just wait to find out who gets the Nobel Peace Prize, and then give it to them. After Kissinger got it, it sort of lost all credibility.
You are presuming that if a debt collector is after you, you actually owe someone something. That's not the way I would bet. In my experience those pursued by debt collectors are misidentified. My wife, e.g., was pursued by multiple debt collectors because someone with the same name died in a hospital without paying his bills. They lived in different cities, the other person was male, she isn't. Etc. But there was no way to stop the calls. I'd personally be in favor of requiring that all attempts at debt collection be made in person, with bonded neutral witnesses present. This may be a bit extreme, but after they'd attacked her for over a year my wife ended up in the hospital and is now listed as permanently disabled. It's not entirely their fault, as she had a weak heart to begin with, but she didn't have any trouble with it until they began attacking her.
Hell, just roast them all over a slow fire and be done with it.
There are many things that "spontaneously combust". The classic example is a pile of oily rags in the cellar, but you could also cite Stallman's laptop. Spontaneously doesn't mean without reason, it means without external reason. (And it's not even quite that limited, as, for example, the pile of oily rags needs to be reasonably warm to start with. And oxygen needs to be in the atmosphere. Etc.)
There are several reported cases of humans "spontaneously combusting". They may not all be fabrications or misunderstandings. Just like batteries, we contain excess energy enough to destroy us. It's not clear that it ever releases spontaneously, but there's no proof that it never happens. It is, however, certainly unusual. But then so is a laptop's battery breaking into fire when it's not plugged in, and nobody's in the room. That, however, we know can happen. People doing the same thing is something that I consider still dubious, but not, in principle, unreasonable.
If I recall correctly, having a different fastest particle wouldn't affect the logic behind relativity. It *would* affect the boundary conditions. So if the neutrino can travel slightly faster than light, then our current detected limits are where they are because we are using electromagnetic widgets to accelerate them. But the actual limits would be somewhere slightly (not much!) higher.
So don't make too much out of this, until you start figuring out how to accelerate things with the weak force.
Most of the theory would remain untouched. The meaning of c would change slightly. It would refer to the speed of a neutrino rather than of a photon. Recalibrating the theory to experimental results would be lots of work, but not really significant. All current experimental results would remain good, but their meanings would alter very slightly.
This doesn't give us a faster than light anything....except neutrinos. Neutrinos would be the new clock establishing synchronicity measures. But neutrinos are so close to photons in speed that this wouldn't have much effect on any predictions. And they interact so poorly with ordinary matter that they wouldn't be a practical clock.
OTOH!!! I Am Not A Physicist. I'm a programmer who was interested in physics in high school and college (and to a much lesser extent since then). So if I'm wrong, please let me know.
Debian definitely has a root account, and the last time I checked, so did Ubuntu. Ubuntu *does* want you to do everything through sudo, which to me seems LESS secure, but they do have a root account. All you need to do to "activate" it is to assign it a password. (Well, this was back around hoary hedgehog or whatever it was called.)
I used Ubuntu for awhile, but found it gave me no advantage that I was using over Debian. And I ended up going back to Debian over package availability. (Not a concern for most users, or even most developers.)
FWIW, I am appalled by Unity, but I don't like the Gnome3 desktop much either. I'll likely switch to something else...and I haven't yet chosen which "something". Those interfaces look like they were designed for mobile phones. Perhaps they are good in that area, but I'm primarily a desktop user.
SUSE has, essentially, not been heard of since it was split off Novell. I don't know what the status of that distribution is. At one point I would have declined to recommend it for the same reason I wouldn't recommend Mandrake: Too much pixie dust in the administration. I don't, however, know it's current status.
Yes, GUIs are quite useful. I'm NOT a neophyte, and I practically always use a GUI. But I like to be able to understand what's happening underneath. So the CLI commands need to continue to work, and changes shouldn't happen without my asking them to. SUSE and Mandrake BOTH violated this. The only distro I'm current with is Debian, but I would have no hesitation in recommending Debian stable for someone getting started. (That wasn't the case a few years ago, but it is now.) If you want to play flash or DVDs or something, you'll probably need to add things from non-free that I've avoided, so I can't really talk about the support there.
Why have evolved the ability to put the 'group' ahead of ourselves.
We haven't. We are, however, dependent on the social matrix for our survival, so we tend to support that. And during much of our evolutionary history most of our social matrix consisted of close relatives, so they had a large number of genes in common with us. Thus genes supporting the social matrix gained survival probability even at the cost of their current body. But this is an iffy proposition, so we don't have any strong tendency in that direction, and would rather sacrifice someone else.
People who take care of the species are more likely to survive because the specious will take care of them.
Correct. That's saying what I've been asserting. But it's the individual genes being selected for evolution, not the species, the group, or the body. And remember, no foresight is involved. No intention. Just probability mechanics.
There are some people who think there is a weak effect on evolving towards group selection that can't be explained by gene selection. I haven't seen anything that convinces me, and I don't get the feeling that most evolutionary genetics specialists feel that the evidence is strong enough to put any faith in. Certainly ALMOST all evolutionary history is consistent with a combination of probability and gene-level selection. Enough so that I'd examine any claims to the contrary with quite a skeptical eye. OTOH, I am not a professional in that field, I merely base my opinions on their explanations.
Actually, evolution only selects for things that are good for the individual. The species can go hang. This is the reason for many problems. N.B., however, that the individual in question is the "gene". This often promotes kin-group altruism. But it also quite often translates into the individual body, as a gene has no guaranteed way of recognizing another body as being host to the same gene.
Caution should be used in reading the previous text. If read incautiously they could be taken as implying intent on the part of the gene. This is not present. Genes don't have intents, merely effects. Like a byte of assembled code. (It's more like assembled or compiled code than it is like source code.) Of course, part of this hangs on your definition of "What is a gene?", and that's a question without a hard and fast answer. It's easy to answer "What is a codon?", but "What is a gene?" is a lot more difficult. The very concept of gene seems to drag along purpose as a part of it's meaning, and the same codon can be part of more than one gene, depending on which reactions that it is participating in at the moment...or even on which other processes are happening. The RNA produced from the DNA often has parts deleted during it's conversion into protein, and which parts are deleted can change the effect dramatically. Messy! Worse, much worse, than spaghetti code.
I *am* in favor of the government running every monopoly. And unless the screeners are hired by the airlines rather than by the airports this is a monopoly. A very badly run and intrusive one, but still a monopoly.
I have not been impressed by the private management of public utilities. The post office, e.g., ran much better and at not much higher cost as a government agency.
I think you're wrong. But short-term trading *is* a problem. It could probably be handled with a per/transaction tax of a percentage of the trade...say twice what the average trader makes. And have it adjusted according to a linear equation so that if you hold the stock for longer than 5 years there's no tax, and if you hold it for a minute or less there's a 100% tax. (Obviously I'm not planning on extending the equation to more than 5 years. That should just be free. But it might be worth extending it to below a minute, so that if you sell after holding the stock for less than a minute you owe more than the transaction price in taxes. Possibly all the way to zero, though I doubt that anyone would be so foolish as to engage in HFT under that tax plan.)
We disagree. I may refuse to install flash on my system, but Microsoft is evil. I will grant that this doesn't imply that everything they do is evil, and I don't know enough about this case to make a judgment. There are sufficient other cases, however, that I generally presume that if MS supports something, it's something that should be avoided. Indeed, considerable evidence to the contrary is required before I will even seriously consider it (though I'll often stay undecided...I usually don't need to decide).
Apple, now, Apple is more ambiguous. They will (did in the past under Jobs) often support something merely because it's good technology. This does not appear to be the case with MS.
I'm sorry, but they MUST expect trolls. If they don't come from the UK, they'll come from elsewhere.
N.B.: It's not only commemorative pages that end up needing to defend against such people. It's everyone. But certainly those who have been in any way prominent are more subject to attack, and they MUST expect that.
This is not a justification for such an attack, and while I think the sentence is too harsh, only by about 50%. Maybe 60%. And it won't do any good, as there are so many inconsiderate (sociopathic?) idiots that removing one doesn't noticeably reduce the problem. And it's not a deterrent because, as I said, they are in certain ways idiots. You can't really think that they are motivated by rational considerations.
Maybe I do think the sentence was too harsh. Not in principle, but because it won't do any good. Possibly he should be publicly whipped instead. Something that will leave scars in places that restrict what clothing he must wear to hide it, so he'll remember, and also something quite painful. And embarrassing. Not something that's expensive, or crippling. And maybe a week or two in jail.
Once upon a time the stocks were used for people like this, but I think that currently there would need to be heavy guards to ensure that he wasn't permanently injured while helpless. (They didn't used to worry about that, but to me it seems an appropriate consideration.)
So the reason I consider the sentence too harsh is that it's too expensive to carry out. And it won't do any good. We need something quick, cheap, and effective (on the one being punished). And we can't expect it to affect deter others. That effect, if it exists, is minor and will not serve as a justification.
There often (usually?) aren't any alternatives. And it's usually impossible to be certain that they're as evil as you might guess they are. The evidence is hidden, or sufficiently obscured that those who want to believe otherwise easily can, and with reasonable justification.
Slashdot Mirror
It depends. I'm not talking about simple read/write to console, or even redirected equivalent. Yeah, that they could all do. But many languages seem to feel that "stream" is a sufficient concept. It is, indeed, a very flexible concept that can deal with many situations. But it doesn't handle others at all well. (And read-write stream is a violation of the idea that made streams a preferable abstraction over files. Other reasons have since been added, but the restriction to readable OR writable was a big part of the original concept.)
When I just checked yesterday OCaML, or possibly it was Haskell, had the ability to write single bytes, or built in types (short, long, etc.). But there was no indication that one could build things up in buffers and write that. The other could do that, but it required building things up in a mutable byte string, byte by byte, and then converting it to a immutable byte string. And I STILL have no idea how much disk space writing a buffer containing n bytes would require. Apparently some type information gets written at the same time (but I'm not even really certain of that). This makes it difficult to plan a random access IO file. And one comment I ran across made it seem that every time I tried to access the length of the byte string, it would traverse the entire string, counting bytes. This *can't* be correct, but that's what it looked like.
No. Haskell has single element arrays that contain references to one item. When you "change" that item what you are doing is replacing the old reference with a new one. I believe items holding the old reference continue to hold it. But the items are just as immutable as Java strings. You can't change them, you can only replace them.
You misunderstand thing. General purpose languages, such as the C or Algol family (which includes, among others Java, Python, and Ruby) do, indeed, include concepts from functional languages. But they don't RESTRICT you to those concepts. It's the insistence that everything be unchanging that is the weakness in the purely functional languages. The Scheme family is the only group of that to have any measurable use. (Note that Lisp is NOT a functional language, even though the concept originated via a subset of Lisp.)
There's much to be said for having the ability to have invariant elements in functions. There's very little in having the requirement that everything be that way. Note that many existing languages that have invariant strings ALSO have mutable strings. Java to name just the most popular example. So by all means allow functions to declare their values invariant. This allows lots of different kinds of optimization. But don't insist that all functions do so. And definitely don't insist that everything be invariant.
That definitely depends on what you're doing. Every time I've looked at a functional language I've bounced. Usually over random access read-write files. IIRC, even doing ordinary IO in Haskell (or possibly it was OCaML) required dropping into C. (Monads have no state, or some such.)
If I were to choose a functional language it would be one of the Scheme family...unless there's something that I really don't understand about Haskell or OCaML. But I have a hard time mapping Scheme variables into byte layouts in a random access file. Still, it is, in principle, doable. Not sure it's doable even in principle in Haskell or OCaML.
FWIW, just yesterday I was looking at Oz (Mozart) again. (Another strange languge, I think it's functional, but their emphasis is on constraint programming.) I bounced on their lack of file IO capability. (This time all I could find documented was text streams that were either input or output, not both.) Well, Oz has an excuse. It's design was finalized over a decade ago, and it's doesn't appear intended for other than academic use.
As far as I'm concerned, though, you can send your monads back to Leibnitz. He got some mileage out of them, but they sure don't appear worth the trouble to me. Except for a severely limited range of problems...and even then there's usually a better way to do things. Final variables, e.g. (I'd say immutable, but I'd mean the way D (Digital Mars D) uses them, which I don't really understand, and which most people won't even have been exposed to. Basically you assign the variable a value at creation time, and it can never change. But immutability can't be cast away, and creation time isn't compile time, so it's not the same as const.or, I think, final.)
The difference is that a tax on pollution can be enforced fairly. Cap and trade, on more than a local scale, can't. Whether you want to call that a weakness in Cap and trade depends on your goals.
He didn't say mislead. That's something you added. He said they will study what gets funded. This is obviously correct, as to study something requires money.
It's also true, however, that there are many individual scientists that have shaded their studies, or neglected to publish studies that came to the "wrong" answer. Elsevier published an entire medical journal that was staffed, edited, and written for by the employees of one drug company. Many of those employees were scientists. All the articles reported favorably on the products of the company. O, and they were reviewed by reviewers that were beholden in some way to that company. (Often as employees.) They *may* all have been honest, I don't know.
Note that this would not have been disreputable (not the parts that I know) if it had come out of the PR agency of that company. But it was ostensibly an independent journal, and the only company's name on it was Elsevier.
Funny, I call it bait and switch.
Well, that's what they publicly claim. Believe it if you want to. But to do that you've got to close your eyes to their actions.
I think I'll wait another decade or so before trusting MS again. Then I'll see what their track record is. Their current one is decidedly rancid, based on their actions as reported in the news media, rather than just their PR announcements. (Yes, the news media does include their PR announcements. And only occasionally identifies them as such. But one can also find reports by unaffiliated individuals. Like the companies that attempted to attend the committee meeting in Portugal on whether or not to accept the MS proposed word processor document standard.)
Just because something is large and threatening doesn't mean it's unnatural.
FWIW, I *do* believe that anthropogenic climate warming is well established. But it sure wasn't *this* kind of argument that convinced me. By this argument you could say that the Tsunami that recently hit Japan wasn't natural...*IF* you accepted the form of reasoning as valid. Which it isn't.
Most times it seems like they should just wait to find out who gets the Nobel Peace Prize, and then give it to them. After Kissinger got it, it sort of lost all credibility.
You are presuming that if a debt collector is after you, you actually owe someone something. That's not the way I would bet. In my experience those pursued by debt collectors are misidentified. My wife, e.g., was pursued by multiple debt collectors because someone with the same name died in a hospital without paying his bills. They lived in different cities, the other person was male, she isn't. Etc. But there was no way to stop the calls. I'd personally be in favor of requiring that all attempts at debt collection be made in person, with bonded neutral witnesses present. This may be a bit extreme, but after they'd attacked her for over a year my wife ended up in the hospital and is now listed as permanently disabled. It's not entirely their fault, as she had a weak heart to begin with, but she didn't have any trouble with it until they began attacking her.
Hell, just roast them all over a slow fire and be done with it.
Do you really think the Democrats are any better on THIS issue? There are issues on which they are better, but this isn't one of them.
There are many things that "spontaneously combust". The classic example is a pile of oily rags in the cellar, but you could also cite Stallman's laptop. Spontaneously doesn't mean without reason, it means without external reason. (And it's not even quite that limited, as, for example, the pile of oily rags needs to be reasonably warm to start with. And oxygen needs to be in the atmosphere. Etc.)
There are several reported cases of humans "spontaneously combusting". They may not all be fabrications or misunderstandings. Just like batteries, we contain excess energy enough to destroy us. It's not clear that it ever releases spontaneously, but there's no proof that it never happens. It is, however, certainly unusual. But then so is a laptop's battery breaking into fire when it's not plugged in, and nobody's in the room. That, however, we know can happen. People doing the same thing is something that I consider still dubious, but not, in principle, unreasonable.
If I recall correctly, having a different fastest particle wouldn't affect the logic behind relativity. It *would* affect the boundary conditions. So if the neutrino can travel slightly faster than light, then our current detected limits are where they are because we are using electromagnetic widgets to accelerate them. But the actual limits would be somewhere slightly (not much!) higher.
So don't make too much out of this, until you start figuring out how to accelerate things with the weak force.
Most of the theory would remain untouched. The meaning of c would change slightly. It would refer to the speed of a neutrino rather than of a photon. Recalibrating the theory to experimental results would be lots of work, but not really significant. All current experimental results would remain good, but their meanings would alter very slightly.
This doesn't give us a faster than light anything....except neutrinos. Neutrinos would be the new clock establishing synchronicity measures. But neutrinos are so close to photons in speed that this wouldn't have much effect on any predictions. And they interact so poorly with ordinary matter that they wouldn't be a practical clock.
OTOH!!! I Am Not A Physicist. I'm a programmer who was interested in physics in high school and college (and to a much lesser extent since then). So if I'm wrong, please let me know.
Debian definitely has a root account, and the last time I checked, so did Ubuntu. Ubuntu *does* want you to do everything through sudo, which to me seems LESS secure, but they do have a root account. All you need to do to "activate" it is to assign it a password. (Well, this was back around hoary hedgehog or whatever it was called.)
I used Ubuntu for awhile, but found it gave me no advantage that I was using over Debian. And I ended up going back to Debian over package availability. (Not a concern for most users, or even most developers.)
FWIW, I am appalled by Unity, but I don't like the Gnome3 desktop much either. I'll likely switch to something else...and I haven't yet chosen which "something". Those interfaces look like they were designed for mobile phones. Perhaps they are good in that area, but I'm primarily a desktop user.
SUSE has, essentially, not been heard of since it was split off Novell. I don't know what the status of that distribution is. At one point I would have declined to recommend it for the same reason I wouldn't recommend Mandrake: Too much pixie dust in the administration. I don't, however, know it's current status.
Yes, GUIs are quite useful. I'm NOT a neophyte, and I practically always use a GUI. But I like to be able to understand what's happening underneath. So the CLI commands need to continue to work, and changes shouldn't happen without my asking them to. SUSE and Mandrake BOTH violated this. The only distro I'm current with is Debian, but I would have no hesitation in recommending Debian stable for someone getting started. (That wasn't the case a few years ago, but it is now.) If you want to play flash or DVDs or something, you'll probably need to add things from non-free that I've avoided, so I can't really talk about the support there.
Why have evolved the ability to put the 'group' ahead of ourselves.
We haven't. We are, however, dependent on the social matrix for our survival, so we tend to support that. And during much of our evolutionary history most of our social matrix consisted of close relatives, so they had a large number of genes in common with us. Thus genes supporting the social matrix gained survival probability even at the cost of their current body. But this is an iffy proposition, so we don't have any strong tendency in that direction, and would rather sacrifice someone else.
People who take care of the species are more likely to survive because the specious will take care of them.
Correct. That's saying what I've been asserting. But it's the individual genes being selected for evolution, not the species, the group, or the body. And remember, no foresight is involved. No intention. Just probability mechanics.
There are some people who think there is a weak effect on evolving towards group selection that can't be explained by gene selection. I haven't seen anything that convinces me, and I don't get the feeling that most evolutionary genetics specialists feel that the evidence is strong enough to put any faith in. Certainly ALMOST all evolutionary history is consistent with a combination of probability and gene-level selection. Enough so that I'd examine any claims to the contrary with quite a skeptical eye. OTOH, I am not a professional in that field, I merely base my opinions on their explanations.
Actually, evolution only selects for things that are good for the individual. The species can go hang. This is the reason for many problems.
N.B., however, that the individual in question is the "gene". This often promotes kin-group altruism. But it also quite often translates into the individual body, as a gene has no guaranteed way of recognizing another body as being host to the same gene.
Caution should be used in reading the previous text. If read incautiously they could be taken as implying intent on the part of the gene. This is not present. Genes don't have intents, merely effects. Like a byte of assembled code. (It's more like assembled or compiled code than it is like source code.) Of course, part of this hangs on your definition of "What is a gene?", and that's a question without a hard and fast answer. It's easy to answer "What is a codon?", but "What is a gene?" is a lot more difficult. The very concept of gene seems to drag along purpose as a part of it's meaning, and the same codon can be part of more than one gene, depending on which reactions that it is participating in at the moment...or even on which other processes are happening. The RNA produced from the DNA often has parts deleted during it's conversion into protein, and which parts are deleted can change the effect dramatically. Messy! Worse, much worse, than spaghetti code.
I *am* in favor of the government running every monopoly. And unless the screeners are hired by the airlines rather than by the airports this is a monopoly. A very badly run and intrusive one, but still a monopoly.
I have not been impressed by the private management of public utilities. The post office, e.g., ran much better and at not much higher cost as a government agency.
I think you're wrong. But short-term trading *is* a problem. It could probably be handled with a per/transaction tax of a percentage of the trade...say twice what the average trader makes. And have it adjusted according to a linear equation so that if you hold the stock for longer than 5 years there's no tax, and if you hold it for a minute or less there's a 100% tax. (Obviously I'm not planning on extending the equation to more than 5 years. That should just be free. But it might be worth extending it to below a minute, so that if you sell after holding the stock for less than a minute you owe more than the transaction price in taxes. Possibly all the way to zero, though I doubt that anyone would be so foolish as to engage in HFT under that tax plan.)
We've got one that does car alarms. And trucks backing up.
What about "Here, kitty kitty!".
We disagree. I may refuse to install flash on my system, but Microsoft is evil. I will grant that this doesn't imply that everything they do is evil, and I don't know enough about this case to make a judgment. There are sufficient other cases, however, that I generally presume that if MS supports something, it's something that should be avoided. Indeed, considerable evidence to the contrary is required before I will even seriously consider it (though I'll often stay undecided...I usually don't need to decide).
Apple, now, Apple is more ambiguous. They will (did in the past under Jobs) often support something merely because it's good technology. This does not appear to be the case with MS.
I'm sorry, but they MUST expect trolls. If they don't come from the UK, they'll come from elsewhere.
N.B.: It's not only commemorative pages that end up needing to defend against such people. It's everyone. But certainly those who have been in any way prominent are more subject to attack, and they MUST expect that.
This is not a justification for such an attack, and while I think the sentence is too harsh, only by about 50%. Maybe 60%. And it won't do any good, as there are so many inconsiderate (sociopathic?) idiots that removing one doesn't noticeably reduce the problem. And it's not a deterrent because, as I said, they are in certain ways idiots. You can't really think that they are motivated by rational considerations.
Maybe I do think the sentence was too harsh. Not in principle, but because it won't do any good. Possibly he should be publicly whipped instead. Something that will leave scars in places that restrict what clothing he must wear to hide it, so he'll remember, and also something quite painful. And embarrassing. Not something that's expensive, or crippling. And maybe a week or two in jail.
Once upon a time the stocks were used for people like this, but I think that currently there would need to be heavy guards to ensure that he wasn't permanently injured while helpless. (They didn't used to worry about that, but to me it seems an appropriate consideration.)
So the reason I consider the sentence too harsh is that it's too expensive to carry out. And it won't do any good. We need something quick, cheap, and effective (on the one being punished). And we can't expect it to affect deter others. That effect, if it exists, is minor and will not serve as a justification.
There often (usually?) aren't any alternatives. And it's usually impossible to be certain that they're as evil as you might guess they are. The evidence is hidden, or sufficiently obscured that those who want to believe otherwise easily can, and with reasonable justification.