In the Soviet communist era there was a joke current in eastern Europe about the Poles breeding a million-ruble dog and shopping it around to brother socialist states. A functionary returns to Poland from his trip to Bulgaria, triumphant. "Did you get the million?" his commissar asked? "Not exactly," came the answer, "but I did make the sale! Look: two half-million ruble cats!"
Interesting quote, which reminds me, oddly, of Steve Wolfram.
In his magnum opus "A New Kind of Science", which I have glanced at in the bookstore and have not the slightest intention of reading anytime soon, Wolfram has spent a great deal of time and money working out simple rules that give rise to complex and interesting behavior.
The important point to notice in the present context as well as in Wolfram's is that while surprisingly simple rules can in some cases give rise to complex intelligent behavior, and indeed it's endlessly interesting when they do, most of them don't.
I presume Mr. Hock is well aware of this. I suggest, though, that this caveat is something for the reader to keep in mind before storing this quote among their Guiding Principles.
Recall that Microsoft renames technical platforms at an alarming rate. This happens so frequently that I conclude it is part of their business model, though I've never understood the advantage of it. Maybe it's a way of locking technical professionals in or out (I'm definitely an outie) by making it unproductively hard to keep up with MS stuff and the rest of the world at the same time. Anyway, is it possible that Hailstorm is alive and well, but about to be renamed to some other meteorological phenomenon, maybe PartlyCloudyChanceofFlurries or something?
MOD THIS PERSON UP, HUH? I share the frustration regarding science threads on/. ; the very attitude of frustration gives me some confidence that this person is for real.
While his comments are informative and reassuring, the underlying moral and legal questions remain. Can research be dangerous by releasing enormously destabilizing technologies? Obviously, yeah. Is the right approach for society to default to tolerating something until it's proven dangerous or to default to suppressing something until it's proven benign? Well, the former is much easier and much more in line with our moral and political traditions, so that's what we're almost certainly gonna do. Is it enough? I'm among those who thinks not.
Our correspondent above (MOD HIM UP, DAMMIT) says light regulation is quite adequate for nanotech in the foreseeable future. This doesn't put me at ease regarding gene hacking, AI, etc., though it reassures me about nanotech specifically.
Correlation does not refute causation
on
Baked Alaska
·
· Score: 3, Insightful
I realize that there's no examination to be qualified as a Slashdot reader, but it amazes me that people with the wit to read Slashdot make such ridiculous arguments as people inevitably do on this subject.
Correlation is not causation, but there's a mechanism, a prediction, a verification of the prediction, and a complete lack of any alternative plausible hypotheses at this point.
Just because we understand the physiology of hangovers, and you drank like a fish last night, and you have a terrible headache just like the last six times you overdid it doesn't mean that your headache is a hangover. After all, correlation is not causation. Still, it might be a good idea to ease up on your drinking anyway.
Anyone who claims the evidence is weak at this point is willfully ignoring the evidence, or selecting *very* carefully from it, or listening to someone else who is doing so.
Things are pretty much on track with the earliest greenhouse predictions from 15 years ago. (Biggest and earliest changes were expected at high northern latitudes. What do you know...)
And it gets dramatically worse from here on. Fossil fuels, in addition to being responsible for a lot of otherwise dangerous global entanglements, are doing damage to the world not only increasingly but acceleratingly. Nothing but ideology and special interests prevent us from escaping our headlong dive toward widespread environmental disruption combined with getting messed up in medieval throwback geopolitics. Losing fossil fuel dependency fast is a big double win, but it's a little inconvenient to some corporations. Hmm.
It's really time people with any brain cells started to look at the
evidence.
It seems to me that the implication is that I need a license to purchase a single transistor, since I can build an ADC given a few dozen of these and a few resistors and some wire.
Unless the law is construed to require licensing possession of any electronics whatsoever, it's not hard to build an ADC. Real pirates aren't finnicky about such licenses and will have no trouble finding a supply of unlicensed op-amps and flip-flops in old equipment.
Parts houses, small-scale electronics board assembly shops and their suppliers, research institutions and small engineering shops, on the other hand, should be completely freaked out by this. If they've been silent, maybe it is because they find the implications of this lunacy so ridiculous as to be unable to give the matter any credibility.
Ultimately, physical reality is unlicensed. Of course the proposed law will fail miserably to "plug" the "hole" a.k.a. "physical reality", but it could do an amazing amount of damage in the process. This is like trying to license every food item everywhere on the grounds that it might contain alcohol or be used to create alcohol, which is justifiably a regulated substance.
If this passes or even gets out of committee the way it's described here, the congress is dramatically less competent than one would hope.
Hmm. If it's a mere order of magnitude, why can't the actual real internet scale up?
Anyway, in which real-world instance does the delay of ftping or FedExing data have sufficient incremental costs that justify even the operating expense of internet 2, which is to generously presume that it actually ends up in substantial use rather than as an expensive empty pipe. (Waxahatchie anyone?)
Here's another way to look at it. If I'm running a research institution researching anything but internet 2 itself, why would I prefer an internet 2 node to the gigantic computing cluster I could afford instead right onsite?
A sensible cost-benefit analysis would have to prove that the cost of generating the bits remotely and then shipping them is substantially less than the cost of generating them locally.
If compute power gets cheaper faster than bandwidth, and since computers are pretty much indifferent to where they are located, I can't see how to formulate a sensible argument in that vein.
Why should I pay more for the pipe than the guy on the other end paid for his computer that I'm borrowing?
The only way this makes sense is if bandwidth is cheap and compute power is scarce. Correct me if I'm wrong, but this seems to me wrong on both counts.
We've already established that FedEx wins on bandwidth, now what about cost per bit.
For me to get time on the FedEx petabit jumbo jet costs what, $10E-10/bit? Now presume that internet 2 will have a hundred nodes, and will cost ten billion (optimistic on both counts) so about a hundred million per node or about ten million per node per year. So one second costs about 3 cents, and I get 0.4 gig for it presuming there is the demand for full utilization.
So where is the scientific reason for spending a hundred times more per bit? If it's a big shipment, I can wait for the plane. If it's a small shipment, I can wait for good old internet 1. If it's interactive, I should fly myself to the computer that's doing the crunching or upload the code to my local platform. I have yet to see a legitimate scientific application for this. Maybe there's a futuristic entertainment angle to internet 2, but should NSF be funding ultra-luxury entertainment?
Internet 2 is a solution looking for a problem.
Or maybe sequelitis. "The first one was a hit, let's hurry up and get another one just like it out."
Bah. I don't pay taxes so people can win pointless expensive races. Show me how this helps anything that is remotely in the public interest.
The assimilation technique to wchich you refer is about weather prediction - it amounts to tuning the initial condiitions so you can get deeper into the specific dynamics.
Climate prediction is not about dynamics,it is about statistics. In other words, it is about identifying the shape of the butterfly, not where the dot happens to be on the butterfly.
So you have just made a much more sophisticated version of the same error that everyone who wants to believe that climate principles are somehow unknowable (ooh, "chaos", so let me keep my SUV) are making.
Tuning the model to generate appropriate statistics is very different than tuning the model to generate very specific dynamics. In the latter case you are limited by chaotic nonlinear dynamics to a few weeks. In the former, you are trying to identify processes that are interacting in complex ways but are fundamentally dissipative and hence predictable in principle.
Not as alarming as the/. blurb made out, but still revealing of the corporate mindset. Apparently AOL/TW/CNN still finds something dubious or alarming about the concept that people would have something to say to each other and use their technology to do it. In the mass media world, everyone who wasn't a member of a tiny content-production elite was expected to be a consumer and only a consumer. To the extent that everyone is now a publisher, this is threatened.
AOL/TW/CNN obviously has risked much to become a major player in the content game. Their discomfort with a world in which anyone is a content producer leaks out here. You'd hope they would find ways to profit from this prospect of freedom, rather than trying to squelch it, but it's not surprising that some folks in that outfit don't get it.
As for me, I'm not anti-big-corporation where big corporations matter. I like airlines and bridge builders and silicon foundries, but I'm not about to set one up in my basement. I don't like Starbucks, because their main value-added is de-localizing what ought to be a lot of small businesses.
If information megacorps want to help me, they'll help me make the most of all the content out there, and they'll help me stay secure even though there's no sensible way to keep bad people out of chat rooms. I don't want to live in a world where people steal my credit card, but even more I don't want to live in a world where significant powers feel free to characterize online chat as subversive.
However as a counter-point I
have to ask if you can prove this is not a regular occurrence? Geology is about thousands and millions of years, and while you can assert that recent
findings point to an anomaly compared to average global temperature patterns over the past million years, how do you know these wild fluctuations aren't
normal?
You claim to like real facts, yet you write as if
you didn't read what I wrote. I said the CO2 excursion was absolutely without precedent.
There are strong reasons to expect that the climate system will be settling down from this huge and sudden input for several thousand years. These strong reasons are simple consequences of elementary systems theory and classical physics.'
What we've seen so far is only slightly abnormal, but we ain't seen nothin' yet.
Again, if this is in agreement with what some idiot believes it doesn't make it false any more than it makes it true.
Learn to listen to the real debate, not its echoes.
Temperature averages over 40 years do not a geological event make
No, but an excursion of CO2 concentrations outside the range of the past million years over the space of a single century is indeed a geological event.
nor can you make the
assertion that burning fossil fuels is causing global warming without having to prove it.
Waiting for "proof" is like waiting until after the fire to purchase property insurance.
Of course, you certainly need to amass a lot of coherent evidence before you make the increasing claims of plausibility, statistical significance, and generally accepted. Within the field of physical climatology, that's all happenned over the last twenty years. Of course, to read the libertarian press, which would find this piece of physics hard to reconcile with their politics, you wouldn't know it.
Unfortunately in the real world physics trumps philosophy every time.
This moves in the direction of microtransactions, so it is good. When people pay content and editorial providers directly, content and editorial provision will improve, and the web will finally live up to its true potential.
Shame about the PayPal thing, though. What's wrong with my MasterCard?
Re:We are in *much* more trouble one way
on
SSSCA Hearing
·
· Score: 2
How many programmers are going to release something for free that might land them in prison for 5 years?
Assuming you're for real and not trying to subvert the position you claim, your last sentence misses the point at best.
I for one don't care how much Hollywood or Nashville charges for their crap. If it's cheap enough, I may watch or listen, and if it isn't I won't. I have no intention of circumventing any copy protection, and while I think copyright law is already excessive, I'm not overwhelmed by sympathy for your breaking the law for such a self-indulgent pursuit.
The reason we are in *much* more trouble one way is that under this proposal *any* technology is *presumed* illegal unless proven otherwise. Who knows when some technologically illiterate authorities will decide that some perfectly innocent activity is felonious? It's not as if there were no precedent for that!
In fact, it's possible to read this proposal as saying that almost any action whatsoever regarding information technology is considered a felony until proven otherwise by getting some sort of a "certification".
In other words, "Sit back and consume. Production of information is for licensed professionals. We'll bring back the old safe days where information was controlled by a cartel, and the idea that any ordinary person had the right to promulgate their opinions was just a nice theory with no practical implications. We'll take care of your entertainment and we'll decide what you need to know and what you don't."
This isn't just about your two thousand MP3s, kid.
please be fair to Intel
on
SSSCA Hearing
·
· Score: 2
The position that DRM should be technically possible to enforce (Intel's position) may be controversial on Slashdot, but it's dramatically less destructive than the position that all hardware must be licensed.
DRM, to whatever extent that it should exist, should seek some sort of license on the host equipment before yielding its goodies. That's one thing.
I should not have to get a license before I plug an IC into a socket or write a line of perl. That's another thing entirely. Obviously, that outcome will be disastrous for many sorts of innovation, including but not limited to open source, and it will also be a real blow to free expression.
The difference is enormous. It's not "either way we are in trouble", it's "one way we don't get to watch lousy movies and listen to bad music without paying the bastards their cut, and the other way we don't get to touch digital technology of any sort without a license". DRM may or may not be a big deal (I think it will eventually backfire and go away, like every copy protection scheme before it). Moving to a world where any technological activity is expressly forbidden without asking for permission is a very big, very bad deal.
"Either way we are in trouble" is silly. It's like saying "hmm, if I hit the brakes hard enough to avoid hitting that jackknifed semi, I'll spill my coffee, so I'm in trouble either way". It may be true, but it's really not a serious question as to which option to prefer.
Actually, if you have the time to compose a message, take the trouble to print it out, fold it into an envelope, and stick a stamp on it.
Email doesn't count for the same reason spam doesn't count. Show them you care enough to spring for an envelope and a stamp. My impression is that legislators take signed postal mail much more seriously than email, and what's more, I think that is a reasonable position for them to take.
Is that a climate prediction, or a seasonal weather prediction? Climate is long-term.
When the north pole faces the sun, I predict it will be warmer in the north than when the south pole faces the sun, all else being relatively equal. This is a climate prediction. It predicts the physical properties of the atmosphere in response to a shift in the earth's axis relative to the sun. It is no different in principle from predicting shifts in properties of the atmosphere due to changes in atmospheric composition.
The details are harder to work out because we have less historical evidence to work with. (Whether or not this is good news is left as an exercise for the reader.) However, the chaotic trajectory of the system is not especially relevant.
For those familiar with the pop literature, we aren't discussing where on the butterfly the system is going to be at a particular moment. We are discussing how we are changing the underlying system, and hence the overall shape of the butterfly, much as it changes naturally every season.
The problem with the IPCC data as fuel to your approach is that the weighted average itself is biased. For one thing, the field has been largely led by
climate modelers, even though the validity of the models is highly questionable.
*sigh* That again. Look, science is exactly identical to the construction and testing of models. I presume you mean high-order fluid-dynamics based models in particular. These are much more highly constrained than some would have you believe.
It just isn't possible to build a fluid dynamical model that reproduces the contemporary climate well that gives you whatever arbitrary future prediction you want. Unfortunately, you'll have to try it for yourself before you can do other than take my word for it.
Anyway, I'll presume you are taliking about that class of model below.
Climate models, like weather models, have a lot of "tweak factors" which
are used to adjust for factors that the model cannot incorporate. This means that models are tweaked to produce a match to history, and then their
forecast is used.
Assuming that were true (it isn't exactly true of any significant models I know of) that would not constitute a bias in a statistical sense. It would not systematically cause the overestimate or underestimate of any quantity.
Selection (publication selection) leads those models to be included as the best
forecasters (fund managers are given more money if they have good track records).
An interesting analogy, but as Tom Peters points out, selection works on available diversity. It's actually much easier to build a mututal fund than a (3D fluid-dynamical atmosphere-ocean) climate model. The behaviors of these funds is much less constrained than the models'. So the analogy doesn't really hold.
This is one reason that the IPCC consensus estimate changes significantly (and not in a convergent direction) from one report to the next.
"Not in a consistent direction" is what one would hope for in converging on a complex truth, isn't it?
In addition, a cost/benefit analysis requires a good analysis of the cost of remediation. In the environmental area, most analysis goes towards the
"benefit" (degrees of avoided warming per century, or in your case, avoided economic losses from pessimistic outcomes). But little focus is given to the
cost (economic impact with trickle-down costs). Since the economic system seems to be as hard to predict as the climate, this means that we need to
take the most pessimistic views of the economic cost of remediation into our cost benefit analysis also!
Absolutely, it's a multidimensional integration across all the uncertainties weighted by the probability distribution. You can ask any competent electrical engineer how to do this if there aren't any economists who can actually contemplate such a thing.
An AC wrote:
This is a very perceptive comment, but it neglects two key aspects which make the economic analysis tricky.
Oh, it's very hard. That doesn't justify not trying! This stuff is important!
First, when doing a proper cost-benefit
analysis of a process which stretches over centuries, the choice of a discount rate can swing the decision either way (i.e. if I choose to make a dollar
today at the cost of ten dollars in a hundred years, even slight changes in the projected interest rates over the next century can completely change
whether it's a wise choice or not)
Yes, unfortunately however sound the discount rate is for individual decisions, it is invariably destructive and immoral for collective decisions that involve items that are fundamntally irreplaceable. Once you get to quibbling about the discount rate, you are quibbling about the rate at which society sacrifices those resources which deliver value on a longer time scale. One of the moral responsibilities of the collective component of the society is to identify such irreplaceable resources and place them out of the reach of decision processes affected by the discount rate.
A stable climate is such a resource. Many other resources are currently valued based on the presumption of a reasonably stable climate. Hint: it's February 24 and the current temperature in Madison WI is 58 F. What's the valuation of nearby winter resorts? What about my old wooden house, which was not designed for a climate where termites are viable?
Second, even if you agree on a discount rate and manage to get some kind of economically realistic cost-weighted
estimate of the damage from global warming, you can't just plug in the numbers and go merrily on your way, because you also have to consider the
relative uncertainties of the two costs. The situation is exactly analogous to a farmer selling futures on his grain harvest. A rational farmer will sell futures
at a price which is slightly less than the time-discounted price he expects to receive for his grain. The reason for this is that by doing so, he is gaining a
tangible benefit -- he is insured against a poor harvest. The mathematics quickly get out of hand, but suffice it to say that given a choice between two
outcomes, one of which has a small uncertainty in the outcome and the other of which has a large uncertainty, it's rational to choose the first, even if its
expected value is slightly smaller than the second (the tradeoff between reducing uncertainty and reducing expected value is the level of risk aversion).
The application this has to global warming is that a risk averse policymaker will actually choose to pollute more than the break-even point between
economic/environmental harms, because the economic harms are 1)more immediate and 2)have a lower uncertainty given the current state of the
science. I've always had a sneaking suspicion that if you wanted to change decision makers' minds, it would be more effective to decrease the
uncertainty in the effects of global warming than to inflate people's expectations of harms. If we can ever predict global warming to something like a
quarter of a degree, the decision would be a no-brainer. With the current 2 degree spread in the consensus view, a foot-dragger can always say
"Yeah, but we might get lucky."
An excellent point, though I think it's closely related. The political career of a politician has a discount rate for the future much as does money. The planet's climate system has a range of time constants ramging from a decade to hundred million years and is being subjected to a hundred-year transient. All we are arguing about from a physics point of view is the size of the impulse. A little grasp of systems theory is enough to show that once the impulse is let loose there is no reversing it. Thinking in terms of a 5 or 10 percent discount rate is ludicrously out of proportion.
How to do economic measurement is much more to the point than how to do climate modeling, where at least the fundamental principles (classical physics mostly, with a sprinkling of chemistry here and there) are at least agreed upon.
A nice book called "The Hidden Order" by somebody Friedman but not Milton, explains the market libertarian argument clearly, rigorously and succinctly, not to mention smugly. It is worth reading and utterly wrong.
The easiest way to disprove a hypothesis containing the word "always" is to find a single counterexample. In this case I find it easy enough. Requiring and funding universal public education removes the freedom of the individual to decide whether it is to their own advantage to attend school. Which societies, then, are better off in the aggregate? Those with universal mandated public education, or those where poor people are allowed the free choice of sending their children to work at an early age?
The fundamental error of the Friedmanesque argument (I think he is related to Milton, IIRC) is the presumption without demonstration that the net change in well-being is the sum of the aggregate individual changes in well being, and that the well-being of the surrounding context is not affected in any way except by direct effects on individuals.
While my children may be worse off by a few pennies by not stitching together sneakers for rich people, in other words, they are better off for being more likely to live in a world with better opportunities.
This is the fundamental intention behind all regulations, whether they are well-designed or horribly botched. Of course, every regulation reduces the universe of choices and hence has an immediate negative impact. Stopping the analysis there is pretty silly, though.
No, and I should have pushed this point harder when I was actually in the climate change business. You have encouraged me to reconsider how to elaborate and defend the argument. Many thanks.
Slashdot Mirror
In the Soviet communist era there was a joke current in eastern Europe about the Poles breeding a million-ruble dog and shopping it around to brother socialist states. A functionary returns to Poland from his trip to Bulgaria, triumphant. "Did you get the million?" his commissar asked? "Not exactly," came the answer, "but I did make the sale! Look: two half-million ruble cats!"
In his magnum opus "A New Kind of Science", which I have glanced at in the bookstore and have not the slightest intention of reading anytime soon, Wolfram has spent a great deal of time and money working out simple rules that give rise to complex and interesting behavior.
The important point to notice in the present context as well as in Wolfram's is that while surprisingly simple rules can in some cases give rise to complex intelligent behavior, and indeed it's endlessly interesting when they do, most of them don't.
I presume Mr. Hock is well aware of this. I suggest, though, that this caveat is something for the reader to keep in mind before storing this quote among their Guiding Principles.
Recall that Microsoft renames technical platforms at an alarming rate. This happens so frequently that I conclude it is part of their business model, though I've never understood the advantage of it. Maybe it's a way of locking technical professionals in or out (I'm definitely an outie) by making it unproductively hard to keep up with MS stuff and the rest of the world at the same time. Anyway, is it possible that Hailstorm is alive and well, but about to be renamed to some other meteorological phenomenon, maybe PartlyCloudyChanceofFlurries or something?
While his comments are informative and reassuring, the underlying moral and legal questions remain. Can research be dangerous by releasing enormously destabilizing technologies? Obviously, yeah. Is the right approach for society to default to tolerating something until it's proven dangerous or to default to suppressing something until it's proven benign? Well, the former is much easier and much more in line with our moral and political traditions, so that's what we're almost certainly gonna do. Is it enough? I'm among those who thinks not.
Our correspondent above (MOD HIM UP, DAMMIT) says light regulation is quite adequate for nanotech in the foreseeable future. This doesn't put me at ease regarding gene hacking, AI, etc., though it reassures me about nanotech specifically.
The original article can be found here
Correlation is not causation, but there's a mechanism, a prediction, a verification of the prediction, and a complete lack of any alternative plausible hypotheses at this point.
Just because we understand the physiology of hangovers, and you drank like a fish last night, and you have a terrible headache just like the last six times you overdid it doesn't mean that your headache is a hangover. After all, correlation is not causation. Still, it might be a good idea to ease up on your drinking anyway.
Anyone who claims the evidence is weak at this point is willfully ignoring the evidence, or selecting *very* carefully from it, or listening to someone else who is doing so.
Things are pretty much on track with the earliest greenhouse predictions from 15 years ago. (Biggest and earliest changes were expected at high northern latitudes. What do you know...)
And it gets dramatically worse from here on. Fossil fuels, in addition to being responsible for a lot of otherwise dangerous global entanglements, are doing damage to the world not only increasingly but acceleratingly. Nothing but ideology and special interests prevent us from escaping our headlong dive toward widespread environmental disruption combined with getting messed up in medieval throwback geopolitics. Losing fossil fuel dependency fast is a big double win, but it's a little inconvenient to some corporations. Hmm.
It's really time people with any brain cells started to look at the evidence.
This isn't just horrifying, it's impossible.
It seems to me that the implication is that I need a license to purchase a single transistor, since I can build an ADC given a few dozen of these and a few resistors and some wire.
Unless the law is construed to require licensing possession of any electronics whatsoever, it's not hard to build an ADC. Real pirates aren't finnicky about such licenses and will have no trouble finding a supply of unlicensed op-amps and flip-flops in old equipment.
Parts houses, small-scale electronics board assembly shops and their suppliers, research institutions and small engineering shops, on the other hand, should be completely freaked out by this. If they've been silent, maybe it is because they find the implications of this lunacy so ridiculous as to be unable to give the matter any credibility.
Ultimately, physical reality is unlicensed. Of course the proposed law will fail miserably to "plug" the "hole" a.k.a. "physical reality", but it could do an amazing amount of damage in the process. This is like trying to license every food item everywhere on the grounds that it might contain alcohol or be used to create alcohol, which is justifiably a regulated substance.
If this passes or even gets out of committee the way it's described here, the congress is dramatically less competent than one would hope.
Hmm. If it's a mere order of magnitude, why can't the actual real internet scale up?
Anyway, in which real-world instance does the delay of ftping or FedExing data have sufficient incremental costs that justify even the operating expense of internet 2, which is to generously presume that it actually ends up in substantial use rather than as an expensive empty pipe. (Waxahatchie anyone?)
Here's another way to look at it. If I'm running a research institution researching anything but internet 2 itself, why would I prefer an internet 2 node to the gigantic computing cluster I could afford instead right onsite?
A sensible cost-benefit analysis would have to prove that the cost of generating the bits remotely and then shipping them is substantially less than the cost of generating them locally.
If compute power gets cheaper faster than bandwidth, and since computers are pretty much indifferent to where they are located, I can't see how to formulate a sensible argument in that vein.
Why should I pay more for the pipe than the guy on the other end paid for his computer that I'm borrowing?
The only way this makes sense is if bandwidth is cheap and compute power is scarce. Correct me if I'm wrong, but this seems to me wrong on both counts.
Actually, I can imagine better uses for the money and talent being thrown at this project for no discernible reason.
We've already established that FedEx wins on bandwidth, now what about cost per bit.
For me to get time on the FedEx petabit jumbo jet costs what, $10E-10/bit? Now presume that internet 2 will have a hundred nodes, and will cost ten billion (optimistic on both counts) so about a hundred million per node or about ten million per node per year. So one second costs about 3 cents, and I get 0.4 gig for it presuming there is the demand for full utilization.
So where is the scientific reason for spending a hundred times more per bit? If it's a big shipment, I can wait for the plane. If it's a small shipment, I can wait for good old internet 1. If it's interactive, I should fly myself to the computer that's doing the crunching or upload the code to my local platform. I have yet to see a legitimate scientific application for this. Maybe there's a futuristic entertainment angle to internet 2, but should NSF be funding ultra-luxury entertainment?
Internet 2 is a solution looking for a problem.
Or maybe sequelitis. "The first one was a hit, let's hurry up and get another one just like it out."
Bah. I don't pay taxes so people can win pointless expensive races. Show me how this helps anything that is remotely in the public interest.
*sigh* nice troll...
Climate prediction is not about dynamics,it is about statistics. In other words, it is about identifying the shape of the butterfly, not where the dot happens to be on the butterfly.
So you have just made a much more sophisticated version of the same error that everyone who wants to believe that climate principles are somehow unknowable (ooh, "chaos", so let me keep my SUV) are making.
Tuning the model to generate appropriate statistics is very different than tuning the model to generate very specific dynamics. In the latter case you are limited by chaotic nonlinear dynamics to a few weeks. In the former, you are trying to identify processes that are interacting in complex ways but are fundamentally dissipative and hence predictable in principle.
AOL/TW/CNN obviously has risked much to become a major player in the content game. Their discomfort with a world in which anyone is a content producer leaks out here. You'd hope they would find ways to profit from this prospect of freedom, rather than trying to squelch it, but it's not surprising that some folks in that outfit don't get it.
As for me, I'm not anti-big-corporation where big corporations matter. I like airlines and bridge builders and silicon foundries, but I'm not about to set one up in my basement. I don't like Starbucks, because their main value-added is de-localizing what ought to be a lot of small businesses.
If information megacorps want to help me, they'll help me make the most of all the content out there, and they'll help me stay secure even though there's no sensible way to keep bad people out of chat rooms. I don't want to live in a world where people steal my credit card, but even more I don't want to live in a world where significant powers feel free to characterize online chat as subversive.
You claim to like real facts, yet you write as if you didn't read what I wrote. I said the CO2 excursion was absolutely without precedent.
There are strong reasons to expect that the climate system will be settling down from this huge and sudden input for several thousand years. These strong reasons are simple consequences of elementary systems theory and classical physics.' What we've seen so far is only slightly abnormal, but we ain't seen nothin' yet.
Again, if this is in agreement with what some idiot believes it doesn't make it false any more than it makes it true.
Learn to listen to the real debate, not its echoes.
No, but an excursion of CO2 concentrations outside the range of the past million years over the space of a single century is indeed a geological event.
nor can you make the assertion that burning fossil fuels is causing global warming without having to prove it.
Waiting for "proof" is like waiting until after the fire to purchase property insurance.
Of course, you certainly need to amass a lot of coherent evidence before you make the increasing claims of plausibility, statistical significance, and generally accepted. Within the field of physical climatology, that's all happenned over the last twenty years. Of course, to read the libertarian press, which would find this piece of physics hard to reconcile with their politics, you wouldn't know it.
Unfortunately in the real world physics trumps philosophy every time.
Try curl for a reasonable client side solution.
Shame about the PayPal thing, though. What's wrong with my MasterCard?
Assuming you're for real and not trying to subvert the position you claim, your last sentence misses the point at best.
I for one don't care how much Hollywood or Nashville charges for their crap. If it's cheap enough, I may watch or listen, and if it isn't I won't. I have no intention of circumventing any copy protection, and while I think copyright law is already excessive, I'm not overwhelmed by sympathy for your breaking the law for such a self-indulgent pursuit.
The reason we are in *much* more trouble one way is that under this proposal *any* technology is *presumed* illegal unless proven otherwise. Who knows when some technologically illiterate authorities will decide that some perfectly innocent activity is felonious? It's not as if there were no precedent for that!
In fact, it's possible to read this proposal as saying that almost any action whatsoever regarding information technology is considered a felony until proven otherwise by getting some sort of a "certification".
In other words, "Sit back and consume. Production of information is for licensed professionals. We'll bring back the old safe days where information was controlled by a cartel, and the idea that any ordinary person had the right to promulgate their opinions was just a nice theory with no practical implications. We'll take care of your entertainment and we'll decide what you need to know and what you don't."
This isn't just about your two thousand MP3s, kid.
DRM, to whatever extent that it should exist, should seek some sort of license on the host equipment before yielding its goodies. That's one thing.
I should not have to get a license before I plug an IC into a socket or write a line of perl. That's another thing entirely. Obviously, that outcome will be disastrous for many sorts of innovation, including but not limited to open source, and it will also be a real blow to free expression.
The difference is enormous. It's not "either way we are in trouble", it's "one way we don't get to watch lousy movies and listen to bad music without paying the bastards their cut, and the other way we don't get to touch digital technology of any sort without a license". DRM may or may not be a big deal (I think it will eventually backfire and go away, like every copy protection scheme before it). Moving to a world where any technological activity is expressly forbidden without asking for permission is a very big, very bad deal.
"Either way we are in trouble" is silly. It's like saying "hmm, if I hit the brakes hard enough to avoid hitting that jackknifed semi, I'll spill my coffee, so I'm in trouble either way". It may be true, but it's really not a serious question as to which option to prefer.
Email doesn't count for the same reason spam doesn't count. Show them you care enough to spring for an envelope and a stamp. My impression is that legislators take signed postal mail much more seriously than email, and what's more, I think that is a reasonable position for them to take.
When the north pole faces the sun, I predict it will be warmer in the north than when the south pole faces the sun, all else being relatively equal. This is a climate prediction. It predicts the physical properties of the atmosphere in response to a shift in the earth's axis relative to the sun. It is no different in principle from predicting shifts in properties of the atmosphere due to changes in atmospheric composition.
The details are harder to work out because we have less historical evidence to work with. (Whether or not this is good news is left as an exercise for the reader.) However, the chaotic trajectory of the system is not especially relevant.
For those familiar with the pop literature, we aren't discussing where on the butterfly the system is going to be at a particular moment. We are discussing how we are changing the underlying system, and hence the overall shape of the butterfly, much as it changes naturally every season.
*sigh* That again. Look, science is exactly identical to the construction and testing of models. I presume you mean high-order fluid-dynamics based models in particular. These are much more highly constrained than some would have you believe.
It just isn't possible to build a fluid dynamical model that reproduces the contemporary climate well that gives you whatever arbitrary future prediction you want. Unfortunately, you'll have to try it for yourself before you can do other than take my word for it. Anyway, I'll presume you are taliking about that class of model below.
Climate models, like weather models, have a lot of "tweak factors" which are used to adjust for factors that the model cannot incorporate. This means that models are tweaked to produce a match to history, and then their forecast is used.
Assuming that were true (it isn't exactly true of any significant models I know of) that would not constitute a bias in a statistical sense. It would not systematically cause the overestimate or underestimate of any quantity.
Selection (publication selection) leads those models to be included as the best forecasters (fund managers are given more money if they have good track records).
An interesting analogy, but as Tom Peters points out, selection works on available diversity. It's actually much easier to build a mututal fund than a (3D fluid-dynamical atmosphere-ocean) climate model. The behaviors of these funds is much less constrained than the models'. So the analogy doesn't really hold.
This is one reason that the IPCC consensus estimate changes significantly (and not in a convergent direction) from one report to the next.
"Not in a consistent direction" is what one would hope for in converging on a complex truth, isn't it?
In addition, a cost/benefit analysis requires a good analysis of the cost of remediation. In the environmental area, most analysis goes towards the "benefit" (degrees of avoided warming per century, or in your case, avoided economic losses from pessimistic outcomes). But little focus is given to the cost (economic impact with trickle-down costs). Since the economic system seems to be as hard to predict as the climate, this means that we need to take the most pessimistic views of the economic cost of remediation into our cost benefit analysis also!
Absolutely, it's a multidimensional integration across all the uncertainties weighted by the probability distribution. You can ask any competent electrical engineer how to do this if there aren't any economists who can actually contemplate such a thing.
Oh, it's very hard. That doesn't justify not trying! This stuff is important!
First, when doing a proper cost-benefit analysis of a process which stretches over centuries, the choice of a discount rate can swing the decision either way (i.e. if I choose to make a dollar today at the cost of ten dollars in a hundred years, even slight changes in the projected interest rates over the next century can completely change whether it's a wise choice or not)
Yes, unfortunately however sound the discount rate is for individual decisions, it is invariably destructive and immoral for collective decisions that involve items that are fundamntally irreplaceable. Once you get to quibbling about the discount rate, you are quibbling about the rate at which society sacrifices those resources which deliver value on a longer time scale. One of the moral responsibilities of the collective component of the society is to identify such irreplaceable resources and place them out of the reach of decision processes affected by the discount rate.
A stable climate is such a resource. Many other resources are currently valued based on the presumption of a reasonably stable climate. Hint: it's February 24 and the current temperature in Madison WI is 58 F. What's the valuation of nearby winter resorts? What about my old wooden house, which was not designed for a climate where termites are viable?
Second, even if you agree on a discount rate and manage to get some kind of economically realistic cost-weighted estimate of the damage from global warming, you can't just plug in the numbers and go merrily on your way, because you also have to consider the relative uncertainties of the two costs. The situation is exactly analogous to a farmer selling futures on his grain harvest. A rational farmer will sell futures at a price which is slightly less than the time-discounted price he expects to receive for his grain. The reason for this is that by doing so, he is gaining a tangible benefit -- he is insured against a poor harvest. The mathematics quickly get out of hand, but suffice it to say that given a choice between two outcomes, one of which has a small uncertainty in the outcome and the other of which has a large uncertainty, it's rational to choose the first, even if its expected value is slightly smaller than the second (the tradeoff between reducing uncertainty and reducing expected value is the level of risk aversion). The application this has to global warming is that a risk averse policymaker will actually choose to pollute more than the break-even point between economic/environmental harms, because the economic harms are 1)more immediate and 2)have a lower uncertainty given the current state of the science. I've always had a sneaking suspicion that if you wanted to change decision makers' minds, it would be more effective to decrease the uncertainty in the effects of global warming than to inflate people's expectations of harms. If we can ever predict global warming to something like a quarter of a degree, the decision would be a no-brainer. With the current 2 degree spread in the consensus view, a foot-dragger can always say "Yeah, but we might get lucky."
An excellent point, though I think it's closely related. The political career of a politician has a discount rate for the future much as does money. The planet's climate system has a range of time constants ramging from a decade to hundred million years and is being subjected to a hundred-year transient. All we are arguing about from a physics point of view is the size of the impulse. A little grasp of systems theory is enough to show that once the impulse is let loose there is no reversing it. Thinking in terms of a 5 or 10 percent discount rate is ludicrously out of proportion.
A nice book called "The Hidden Order" by somebody Friedman but not Milton, explains the market libertarian argument clearly, rigorously and succinctly, not to mention smugly. It is worth reading and utterly wrong.
The easiest way to disprove a hypothesis containing the word "always" is to find a single counterexample. In this case I find it easy enough. Requiring and funding universal public education removes the freedom of the individual to decide whether it is to their own advantage to attend school. Which societies, then, are better off in the aggregate? Those with universal mandated public education, or those where poor people are allowed the free choice of sending their children to work at an early age?
The fundamental error of the Friedmanesque argument (I think he is related to Milton, IIRC) is the presumption without demonstration that the net change in well-being is the sum of the aggregate individual changes in well being, and that the well-being of the surrounding context is not affected in any way except by direct effects on individuals.
While my children may be worse off by a few pennies by not stitching together sneakers for rich people, in other words, they are better off for being more likely to live in a world with better opportunities.
This is the fundamental intention behind all regulations, whether they are well-designed or horribly botched. Of course, every regulation reduces the universe of choices and hence has an immediate negative impact. Stopping the analysis there is pretty silly, though.
No, and I should have pushed this point harder when I was actually in the climate change business. You have encouraged me to reconsider how to elaborate and defend the argument. Many thanks.