My thoughts exactly. I wonder, will the next article about relativity reference "some German physics guy"? Or, for that matter, should we be on the lookout for articles about an operating system software codes invented by a Finlandish computer guy?
If, on the other hand, you are more concerned with the facts than with PhD flame-fests, you can try it for yourself and see. I did this a few years back (but if you don't believe me, try it for yourself) and came to the realization that you can't just extend Fourier analysis to imaginary periods (i.e. exponentials), which is in effect what they are doing. Why? Because in series for any finite set of random data, there will be some term with a positive exponent and it will quickly grow to dominate the expansion when you project into the future. In other words, such a series does not converge as time goes to infinity.
I believe that this is the sort of question begging "a priori" factor that they mention in the links everyone keeps citing--if you know, in your heart of hearts, that global warming is real, well of course you have no objection to a model in which exponential run away is built in. And this may be fine if you are trying to quantify something that has already been established, but it is worthless for answering the question "is this exponential growth or not."
But even though they may have limited utility in such cases, I personally, have no faith whatsoever in such models, regardless of whether they are applied to the weather, the stock market, or the number of bunnies on an imaginary island. Why? Because it's too easy to make them say what you want them to, and the people who choose them are only human. I bailed out of the stock market in early 2000 and I've never lost a nights sleep over global warming.
I'm working outside the US at the moment, and my offical address is (in the local language) something along the lines of "In this district of this town, take the road to the Monistary, turn left when you see a sign saying such and such, and look for so and so; it's building #1."
In truth, lie detectors are based on recordable behaviors of humans when exposed to stimulus, namely the sweating or increased breathing. Psycholgists love to talk about that, it's an observable phenomenon.
The problem with "lie detector" tests comes when the result is inconclusive. Does this indicate a lie, or the truth? Neither.
No, the problem is that eating chilli can cause the same symptoms, as can the posibility of sex or failure of an air conditioner. Just because fire trucks have the "recordable behaviour" of showing up at fires, we can not conclude that the fire station is perpetually in flames.
Biased, perhaps, but only in the sense that a textbook that presents the facts is "biased" towards the truth. "Lie detectors" as about as reliable as phrenology or handwriting analysis--e.g. not at all.
--MarkusQ
Information content of slashdot posts
on
Of Ants and Robots
·
· Score: 1
I'd wager that the information content of a typical ant-trail is greater than that of the typical slashdot post. It may not seem like it to us, because we don't place as much value on things like "go left around this rock" as we do on things like "I soviet Russia the rocks go left around you!" but from an information-theory standpoint the ants have us beaten hands down.
yeh but thats leaving a marker, like tieing a string to a tree, how u believe that equates to communication able to support the organization of ants to perform tasks seemingly together is beyond me
Isn't posting on/. "leaving a marker"? Would you claim that it isn't possible to communicate here? I would claim that it is, even in the face of noise (trolls, etc.), errors in the messages ("yeh", "u", etc.) and so forth. I think it's a pretty good analogy for what social insects do, and thus a reasonable refutation of your point.
Well, I know you're not her, but you do have the basic plot about right.
Still, all things considered, I wouldn't change a thing--especially considering that since for a while I was getting basically the same treatment from her as from my employeer, there was a certain symetry to the situation...
Then call and ask specific questions - especially of engineers, who usually don't get calls like that and are more than happy to answer questins about their work. Trade shows are great as well.
Many years ago I dated a woman I met at a trade show while doing exactly that. I was in a micro-crowd at a product demo trying to ask my questions but this "dumb secretary" type had his undivided attention. She was asking really ditzy questions, and he was trying to impress her by giving these long winded, overly complex answers. After I resigned myself to being ignored, I began to realize that she was getting all the information I wanted.
Then I looked at her badge.
It said "Jane Bond" (BTW, Hi, if you're out there.)
After the demo I followed her and started making wise cracks. She tried to play dumb when I confronted her about being an industrial spy, but I pointed out that she had obviously gotten my jokes (sexual inuendo that depends on knowing what a wire-or is is pretty obscure). We wound up going to lunch, and then dinner...
The real problem with new fangled almost-magic batteries is going to be the liability from ISD (Instantanious Spontanious Discharge). I've seen a video of a test with a lab rat, and all I can say is you'll want to keep these things out of reach of rodents! (And that I'm glad I wasn't there to watch it in person!)
That said, I'm fairly sure that voting for this law would, in and of itself, constitute a violation of the bill in question.
The idea is nice, but (as the poster pointed out, there are deep logical problems with passing it as a law). It would have to be reworked as a constitutional amendment and tweeked cairfully if it was to be effective without bizzare unintended consequences.
And if you feel that they are a bunch of buttmunchers who have more allegiance to the oil industry than to the american people, then vote them out.
Tried that. Turns out they have other allegiances than just the oil companies. Like the companies that make the machines that count the votes (optical scan, lever, as well as touch screen).
Otherwise, you have to understand that there's the possibility that there's more to governmental policy than they choose to let you know.
I'm absolutely sure you are correct. That's a large part of why we tried to vote them out. If we spent half what we spend on war on things like science and education at home and abroad, stoped selling (or giving) weapons to every third world crackpot that promises to bolster or "foreign policy," and started following our own laws instead of "bending" them for every corporate player that bought the right ears, we wouldn't need all the paranoid military hardware up there in the first place.
Well they certainly do encourage discussion. I think you could post a question about rubarb pie to Ask Slashdot at start a spirited discussion about Google and how to use it. What I wonder is, is there and question that you could "Ask Slashdot" that would encourage a discussion about something other than Google?
*laugh* Thanks. I of course hope I'm correct as well.
My main advice to anyone thinking of getting into the market (which has worked well for me) is: don't be shy or lag about your reading, and always do the math. I'm not talking about the sort of chartist voodoo that some people advocate, mut more the Warren Buffet style thinking out all the details. Ask your self, what would happen to me if the fed raises/lowers the interest rates? What if sector X crashes, or booms? What would these things do to my position, and what will I do about it? And look at the financials of the companies you bet on (or against). Do they make sense? Do you smell a ton of BS? Do you see how they are making their money--or failing to? Try to get multiple sources for any key data, and don't trust anyone to be absolutely honest. Debug your strategy, just like it was a pice of code.
There are two distinct issues to consider, and they are often directly opposed to each other. The sorts of things you want to make it easy for first time users to grok are often the very things you need to get rid of if you want experienced users to be fast & reduce errors.
Want raw speed and accuracy? Do it in text mode, make everything accessible from the keyboard (or even better, the twenty or so keys of the "ten-key" pad). It will take operators a while to get up to speed, but when they do even the average operators will be two to five times as fast as the very best operators of a mouse and kitchen-sink GUI. And their error rate will be much lower.
On the other hand, be prepared to fight off attempts (every year or so, in my experience) to replace it with a "modern" interface that looks pretty even as it kills productivity.
Mostly from people who consistantly do worse in the market than I do (I've made money every year since the mid-90's). The "logic" is just wrong, since your broker won't actually allow you to have "unlimited" losses--the most you can lose, long or short, is what you have exposed to the market, either by investing or borrowing (you need to have margin to short). You can lose everything by being too exposed long, you can lose everything being too exposed short. But your best bet (IMHO) is to not get too exposed (only invest what you can aford to lose) and have a diversified mix of long & short positions. If you make good choices, you make money when the market goes up, and when it goes down.
Speaking of which, was anyone here lucky enough to short SCOX back when they were way up at 10-20 price? If there's one thing sweeter than watching a crappy company fall, it's profiting from the fall of a crappy company.
I agree 100%. I shorted at 15+ and am loving every minute of this. (I should, by my math, have bailed when it hit $3 the first time and moved on to greener pastures. But I figured it was worth it for the entertainment value--they already paid off my mortgage for me, so now it's just sport.)
"TransHab" was his. NASA was (or would have been) the customer, not the creator, of TransHab. They didn't "layout the ground work". It's like saying "That segway thing isn't so innovative--I read that the post office that almost bought a bunch of them."
How would you feel about a Chemistry class that gave assignments that could only be done with DuPont-brand proprietary reagents? Or an Astronomy class that asked you to record the spectrum of an object only detectable if you used Corning (tm) lenses? Or a Math class where the only way to get the "right" answer was to use a certified HP calculator?
That's not education, that's captive audience marketing.
I think this article/posting is filled with anti-GMO FUD.
I think that's kind of sad for a site like/., which (at least in theory) should be a haven for people who take a positive aproach to life, instead of chicken-little ludites.
And I think that if you really didn't want to know what I think, you shouldn't have asked.
It works great for me. Just have to do a caffeine check before making major changes (and remember to stop the cron job plus test in a sandbox).
Some handy tips:
Use pid files to keep new instances from starting up if a job goes long.
"-j" can be your friend, but (like a real human friend) it can also get you into a heap of trouble if you aren't careful.
Running the make in a permanent loop and just touching things with cron can be a handy trick, especially if you need to let users (or external processes) tweek the process.
Slashdot Mirror
Having seen the work of all three, and met one of them, I'd have to say that he was significantly smarter than Albert or Linus.
--MarkusQ
My thoughts exactly. I wonder, will the next article about relativity reference "some German physics guy"? Or, for that matter, should we be on the lookout for articles about an operating system software codes invented by a Finlandish computer guy?
--MarkusQ
If, on the other hand, you are more concerned with the facts than with PhD flame-fests, you can try it for yourself and see. I did this a few years back (but if you don't believe me, try it for yourself) and came to the realization that you can't just extend Fourier analysis to imaginary periods (i.e. exponentials), which is in effect what they are doing. Why? Because in series for any finite set of random data, there will be some term with a positive exponent and it will quickly grow to dominate the expansion when you project into the future. In other words, such a series does not converge as time goes to infinity.
I believe that this is the sort of question begging "a priori" factor that they mention in the links everyone keeps citing--if you know, in your heart of hearts, that global warming is real, well of course you have no objection to a model in which exponential run away is built in. And this may be fine if you are trying to quantify something that has already been established, but it is worthless for answering the question "is this exponential growth or not."
But even though they may have limited utility in such cases, I personally, have no faith whatsoever in such models, regardless of whether they are applied to the weather, the stock market, or the number of bunnies on an imaginary island. Why? Because it's too easy to make them say what you want them to, and the people who choose them are only human. I bailed out of the stock market in early 2000 and I've never lost a nights sleep over global warming.
--MarkusQ
I'm working outside the US at the moment, and my offical address is (in the local language) something along the lines of "In this district of this town, take the road to the Monistary, turn left when you see a sign saying such and such, and look for so and so; it's building #1."
--MarkusQ
No, the problem is that eating chilli can cause the same symptoms, as can the posibility of sex or failure of an air conditioner. Just because fire trucks have the "recordable behaviour" of showing up at fires, we can not conclude that the fire station is perpetually in flames.
--MarkusQ
Biased, perhaps, but only in the sense that a textbook that presents the facts is "biased" towards the truth. "Lie detectors" as about as reliable as phrenology or handwriting analysis--e.g. not at all.
--MarkusQ
I'd wager that the information content of a typical ant-trail is greater than that of the typical slashdot post. It may not seem like it to us, because we don't place as much value on things like "go left around this rock" as we do on things like "I soviet Russia the rocks go left around you!" but from an information-theory standpoint the ants have us beaten hands down.
--MarkusQ
yeh but thats leaving a marker, like tieing a string to a tree, how u believe that equates to communication able to support the organization of ants to perform tasks seemingly together is beyond me
Isn't posting on /. "leaving a marker"? Would you claim that it isn't possible to communicate here? I would claim that it is, even in the face of noise (trolls, etc.), errors in the messages ("yeh", "u", etc.) and so forth. I think it's a pretty good analogy for what social insects do, and thus a reasonable refutation of your point.
--MarkusQ
Well, I know you're not her, but you do have the basic plot about right.
Still, all things considered, I wouldn't change a thing--especially considering that since for a while I was getting basically the same treatment from her as from my employeer, there was a certain symetry to the situation...
--MarkusQ
Then call and ask specific questions - especially of engineers, who usually don't get calls like that and are more than happy to answer questins about their work. Trade shows are great as well.
Many years ago I dated a woman I met at a trade show while doing exactly that. I was in a micro-crowd at a product demo trying to ask my questions but this "dumb secretary" type had his undivided attention. She was asking really ditzy questions, and he was trying to impress her by giving these long winded, overly complex answers. After I resigned myself to being ignored, I began to realize that she was getting all the information I wanted.
Then I looked at her badge.
It said "Jane Bond" (BTW, Hi, if you're out there.)
After the demo I followed her and started making wise cracks. She tried to play dumb when I confronted her about being an industrial spy, but I pointed out that she had obviously gotten my jokes (sexual inuendo that depends on knowing what a wire-or is is pretty obscure). We wound up going to lunch, and then dinner...
--MarkusQ
The real problem with new fangled almost-magic batteries is going to be the liability from ISD (Instantanious Spontanious Discharge). I've seen a video of a test with a lab rat, and all I can say is you'll want to keep these things out of reach of rodents! (And that I'm glad I wasn't there to watch it in person!)
--MarkusQ
That said, I'm fairly sure that voting for this law would, in and of itself, constitute a violation of the bill in question.
The idea is nice, but (as the poster pointed out, there are deep logical problems with passing it as a law). It would have to be reworked as a constitutional amendment and tweeked cairfully if it was to be effective without bizzare unintended consequences.
--MarkusQ
Tried that. Turns out they have other allegiances than just the oil companies. Like the companies that make the machines that count the votes (optical scan, lever, as well as touch screen). I'm absolutely sure you are correct. That's a large part of why we tried to vote them out. If we spent half what we spend on war on things like science and education at home and abroad, stoped selling (or giving) weapons to every third world crackpot that promises to bolster or "foreign policy," and started following our own laws instead of "bending" them for every corporate player that bought the right ears, we wouldn't need all the paranoid military hardware up there in the first place.
--MarkusQ
Well they certainly do encourage discussion. I think you could post a question about rubarb pie to Ask Slashdot at start a spirited discussion about Google and how to use it. What I wonder is, is there and question that you could "Ask Slashdot" that would encourage a discussion about something other than Google?
Hey, maybe I should Ask Slashdot that!
--MarkusQ
*laugh* Thanks. I of course hope I'm correct as well.
My main advice to anyone thinking of getting into the market (which has worked well for me) is: don't be shy or lag about your reading, and always do the math. I'm not talking about the sort of chartist voodoo that some people advocate, mut more the Warren Buffet style thinking out all the details. Ask your self, what would happen to me if the fed raises/lowers the interest rates? What if sector X crashes, or booms? What would these things do to my position, and what will I do about it? And look at the financials of the companies you bet on (or against). Do they make sense? Do you smell a ton of BS? Do you see how they are making their money--or failing to? Try to get multiple sources for any key data, and don't trust anyone to be absolutely honest. Debug your strategy, just like it was a pice of code.
--MarkusQ
There are two distinct issues to consider, and they are often directly opposed to each other. The sorts of things you want to make it easy for first time users to grok are often the very things you need to get rid of if you want experienced users to be fast & reduce errors.
Want raw speed and accuracy? Do it in text mode, make everything accessible from the keyboard (or even better, the twenty or so keys of the "ten-key" pad). It will take operators a while to get up to speed, but when they do even the average operators will be two to five times as fast as the very best operators of a mouse and kitchen-sink GUI. And their error rate will be much lower.
On the other hand, be prepared to fight off attempts (every year or so, in my experience) to replace it with a "modern" interface that looks pretty even as it kills productivity.
--MarkusQ
Mostly from people who consistantly do worse in the market than I do (I've made money every year since the mid-90's). The "logic" is just wrong, since your broker won't actually allow you to have "unlimited" losses--the most you can lose, long or short, is what you have exposed to the market, either by investing or borrowing (you need to have margin to short). You can lose everything by being too exposed long, you can lose everything being too exposed short. But your best bet (IMHO) is to not get too exposed (only invest what you can aford to lose) and have a diversified mix of long & short positions. If you make good choices, you make money when the market goes up, and when it goes down.
--MarkusQ
I agree 100%. I shorted at 15+ and am loving every minute of this. (I should, by my math, have bailed when it hit $3 the first time and moved on to greener pastures. But I figured it was worth it for the entertainment value--they already paid off my mortgage for me, so now it's just sport.)
--MarkusQ
"TransHab" was his. NASA was (or would have been) the customer, not the creator, of TransHab. They didn't "layout the ground work". It's like saying "That segway thing isn't so innovative--I read that the post office that almost bought a bunch of them."
--MarkusQ
Then he should object to the instructor.
How would you feel about a Chemistry class that gave assignments that could only be done with DuPont-brand proprietary reagents? Or an Astronomy class that asked you to record the spectrum of an object only detectable if you used Corning (tm) lenses? Or a Math class where the only way to get the "right" answer was to use a certified HP calculator?
That's not education, that's captive audience marketing.
--MarkusQ
Point taken. I wouldn't trust some of them to count their own toes...which I'm sure they'll claim they have, being "persons" now and everything.
--MarkusQ
Maybe his sense of humour fell through a one-way hash function some time back, but it's pretty clear from context that he's kidding.
--MarkusQ
Read more and tell me what you think.
I think this article/posting is filled with anti-GMO FUD.
I think that's kind of sad for a site like /., which (at least in theory) should be a haven for people who take a positive aproach to life, instead of chicken-little ludites.
And I think that if you really didn't want to know what I think, you shouldn't have asked.
--MarkusQ
It works great for me. Just have to do a caffeine check before making major changes (and remember to stop the cron job plus test in a sandbox).
Some handy tips:
--MarkusQ
One (me).
And at least two others that I know in meatspace...
--MarkusQ