So it seems the task is coming up with a standard format and enforcing it.
Which will cost FAR more than $100 billion, and be done so badly as to render the system nearly useless.
Ever parse a MAGE-ML doc that turns out to have the actual gene expression values in an "other" or "comments" field? Most "standard formats" are so arcane, complex and counter-intuitive that most people using them can't figure out the appropriate place to put the information.
Furthermore, medical terms change with time as new procedures are introduced and old procedures modified. The proposed format is going to either have to handle that or become the kind of straight-jacket that 501(k) process has been in medical devices.
Anyone contemplating touching any aspect of this project simply MUST read Stephen Flowers' "Software Failure: Management Failure", which is a collection of case studies of failed major software initiatives of just this kind. The book is in fact worth reading for anyone with an interest in why software systems fail, which should be everyone involved in software development.
And, once you get a completely nationalised health system, you effectively have a system equivalent to insurance with the largest possible pool.
Actually, no, because there is virtually no risk involved: EVERYONE gets sick, and EVERYONE dies, and about half of EVERYONE's health care costs come in the last six months of life.
Although costs vary, they don't vary by that much, although the tail of the distribution is long. See figure B1 in this report on Canadian health care costs to see the actual distribution. For something over 70% of the population the average cost of a single hospital stay is less than $10,000, and virtually everyone has a couple of those stays in their lifetime (I've had one despite being in extremely good health generally.)
This is utterly unlike true insurance models--auto, home and term life--where the majority of people who pay premiums never collect a claim.
It is interesting to note that both the Canadian and American health care systems use insurance models, and suffer from similar problems of access and spiralling costs. I believe this is due to the inherent inappropriateness of an insurance model for a service that everyone will need and everyone which has a relatively low variance of total payouts.
A reasonable model of health insurance would deal with catastrophic costs only, say in excess of $10,000 per hospital stay as indicated by these data. As not everyone falls into that category, one could actually use insurance to spread RISK, which is not really possible under an "everyone pays, everyone benefits" model because the tails are not that relevant to the overall cost of the system, so you basically have a situation where there is very little risk to be spread (closer analysis of the numbers could contradict that, but that's my impression from a first look.)
Now that the Republican Party is passing out of power in the US we can expect that there will be a good deal less socialism, so whether or not it is desirable (I believe mostly not) there is going to be a good deal less of it now that the more liberal Democratic Party is in control.
I've never been clear exactly WHY Americans call their Socialist Party "the Republican Party". This is the party that has overseen massive growth in government both in responsibilities and costs, intrusive imposition of the federal government into areas normally reserved for the states or the people, and huge transfers of risk "in solidarity" from private individuals and organizations to the public.
It has capped all of this by actually taking ownership of significant parts of your financial system, which must in today's world be counted as firmly amongst "the means of production." Now that the liberal Democratic Party is replacing the socialist Republican party perhaps free market solutions will be prescribed for some of the things that ails the US, like allowing badly-run businesses that make products no one wants to FAIL.
They can protest all they like but no Israeli is going to see it.
I don't recall there being any TV cameras at Amritsar, yet somehow the word got out.
The cowardly, craven apologists for terrorist violence of the kind being committed by BOTH SIDES of the Israeli-Palestinian conflict always have to make trivially false claims of this kind to support their case, but you aren't fooling anyone but your fellow cowards, who are looking for any excuse to deny the trivial truth that non-violent RESISTANCE is almost always more effective and more efficient that violent ATTACK.
You mean like whitespace-based indentation a la Fortran?
Unlike all those OTHER languages, whose indentation isn't based on whitespace?
Even if your quip was coherent, it is a-historical. Any indentation other than the basic first six columns was forbidden prior to the whole "structured FORTRAN" thing in the late '70's, and that was just a convenience for humans trying to read the code. Scoping in FORTRAN has always been a matter of keywords, not whitespace, and in fact the language explicitly ignores even whitespace that one might wish it didn't.
For example, the famous one-character bug that blew up an American Venus probe in the '70's was due in part to FORTRAN's removal of whitespace in numerical constants during compilation, so the loop header:
DO I = 1. 10 WHILE
became:
DO I = 1.10 WHILE
"I" is of course an integer, so this is equivalent to:
DO I = 1 WHILE
and rather than the list of ten items being fully initialized only the first one got handled. If the "." had been a "," like it was supposed to, everything would have been fine. (I haven't written FORTRAN for a decade or so, so my syntax in the above may not be quite correct, but you get the idea.)
With regard to Python, getting over the "oh my god it's full of whitespace" thing was an issue for me in adopting the language. Having worked in SGML-based text processing in the '90's I knew only too well that that "whitespace is not actually evil--it is just misunderstood." But once you get over it, it becomes a matter of habit just like curly brackets did when I moved from FORTRAN to C in the late '80's.
As another person replying to my original comment said, Python + scipy + rpy is an excellent environment for routine scientific programming, and if you add wxPython into the mix you've got a pretty good application development language too, that is deservedly getting a lot of uptake in the scientific community.
(a) it allows for easy access of Fortran and C library routines
(b) it allows you to pass large blobs of data by name
(c) it makes it easy to pass data to and from your own compiled C and Fortran routines
So, it's exactly like Python, except with an outdated 1970's syntax that was frankly pretty weird to start with:-)
I've used R, and found it useful for some of its relatively esoteric capabilities, but currently use it almost exclusively via rpy now, the Python binding to R.
Furthermore, I've been using it less in recent years as the native statistical capability of Python has continued to improve. I can appreciate that people who work strictly in data analysis could find R an appropriate tool, but as someone whose work spans multiple areas, from analysis to application design and development, R is too limiting a tool, and using it always feels a little alien and weird.
Reducing package size is green and it costs less to produce. Why increase the price if cost is lowered?
The problem is that faux environmentalists have for decades tried to convince people to change they way they live based on a kind of puritanical moral argument. They want people to change to become supposedly more morally pure through self-sacrifice, but they don't have the guts to say that--instead they hide behind the genuine need to protect the environment.
This propaganda campaign has been so successful that "green" as been equated in the public's mind with "costly, difficult, limited and painful," to the extent that businesses sometimes reject green solutions out of hand because of the up-front perception that "green costs more."
In reality, the underlying economics of genuinely green technology is almost always better than non-green tech, unless the non-green tech is specifically exploiting a loophole in the system of property by dumping effluent into the commons, which is becoming more difficult due to better protection on private property around the world.
Even the cases where people are claiming in this discussion that "green is not always cheaper" they are typically asking us to look at the problem in a completely myopic way, by considering only exactly the same industrial process that is currently being used and post-processing the effluent from it, which is in most cases silly compared to re-engineering the process so that the amount or toxicity of effluent is reduced.
Green companies have been doing this kind of process re-engineering for DECADES, with the primary goal of saving money. Look at the well-known example of Interface. And if an older company can't afford to replace existing plant... well, let 'em get competed into the ground by newer companies that can.
The opposite of "green" is "wasteful", and what economically sane company is in favour of being wasteful?
"Virtually no waste" translates into "Japan believes it has correctly accounted for all the plutonium in its reprocessing network even though accounting errors have resulted in a thousand kilograms going missing."
Reprocessing is clearly a desirable thing to do from all kinds of standpoints, but the issues of safe transport for the cores and actually maintaining a sufficiently accurate inventory on the fissile material create huge practical problems.
Remember, over a decade or two you will be moving millions of kg of material around for reprocessing, so if you inventory control is 99.9% accurate you will still have enough missing plutonium to make a hundred-odd nice little nuclear bombs.
Only to anyone who understands that temperature is an intensive thermodynamic quantity and therefore cannot be meaningfully averaged in an inhomogeneous medium such as the atmosphere. It is a bit scary to see a meaningless number continually used as the basis for a massive power grab.
Atmospheric heat content can be averaged, if we had a clue what it was, and a surrogate average temperature generated from that. As near as I can tell all global temperature records of note are dry-bulb temperatures which are, by themselves, thermodynamically useless for this sort of averaging.
Global OCEAN temperatures are rising, and that should be a matter of concern because that is a clear indicator of increasing ocean heat content, but as I said, it is a bit scary to see people who believe absolutely in their own righteousness base their arguments on a literally meaningless number.
And the point about complex models failing is very well taken. The climate system is far more complex and we have far less information about it than the financial system, and anyone relying on computer models for their beliefs about climate systems ought to be willing to invest everything they own based computer models of the financial system.
Nocera explores the age-old debate between those who assert that the best decisions are based on quantification and numbers, and those who base their decisions on more subjective degrees of belief about the uncertain future.
Why is anyone still making this distinction, as we now know that the only self-consistent numerical representation of risk follows directly from our subjective degrees of belief about the uncertain future? Furthermore, we have known this for over a generation... isn't it about time that the knowledge start filtering into the popular discourse?
While Bayesian methods are not always all that useful for practical problems (I use them on occasion in my work) the conceptual foundations and deeper understanding of the nature of plausible reasoning and its relation to probability theory needs to be more widely understood.
One of the big take-home messages from the Bayesian revolution is that probability theory is nothing but quantification of what we do subjectively, insofar as our subjective impressions are self-consistent, so the only people who are still debating quantitative vs subjective approaches as such are people who do not understand the question.
As such, the implementation is hardware-dependent, which is why there isn't a standard implementation of this function for Windows CE.
It is nevertheless the case, as others here have pointed out, that the correct way of handling date problems is to use Julian dates and day numbers, for which all these algorithms exist in well-documented forms. Getting the Julian Day number in this case should be trivial, as the function gets the number of days since January 1st 1980 as an argument, so it is just a matter of adding that to the Julian Day of the start date.
If you can't use a third-party calender/time lib (which is obviously the preferred solution given the hellish complexity and enormous subtlety of correct date/time representation) year/month/day divisions should be kept STRICTLY in the UI, and all internal date representations and calculations should be in terms of Julian Day or Date.
It is global warming, as the global average temperature has increased, as has the average temperature of most locations on Earth.
What is this 'global average temperature' of which you speak? Temperature is an intensive thermodynamic property and as such cannot be averaged meaningfully in an inhomogeneous medium like the atmosphere.
Global atmospheric heat content is meaningful. Global average temperature is at best ambiguous and at worst meaningless. It is perfectly possible for the temperature to fall everywhere on Earth and the heat content of the atmosphere to increase.
Global ocean temperature, now... that's meaningful, and increasing, but I would take global warming a lot more seriously if the people who are all het up about it showed an even rudimentary grasp of basic physics, rather than prating on about that nominal effects of CO2 as if you'd have to be a moron to be sceptical about their magnitude.
A sceptic about the magnitude of the effect of CO2 on global heat content may be reasonable. A person who puts there argument in terms of a problematic quality such as global average temperature is not.
Furthermore, since every single story we have seen recently about global warming has announced loudly and clearly that climate scientists have no clue whatsoever about what is actually happening in the climate, why should we trust them with geo-engineering?
Surely everyone here has noticed the spate of stories in the past year or two saying that arctic ice is melting far faster than climate scientists expected or that glaciers are retreating far faster than climate scientists expected or that some other climate system is changing far faster than climate scientists expected?
Each of these stories, if read by someone who hadn't bought into a religious belief in the infallibility of climate scientists, would be taken as clear-cut evidence that climate science as a whole is terrible at predicting anything to do with climate, even when they deign to make predictions about physical qualities, rather than made-up nonsense terms like "global average temperature."
While there may be a larger than average representation of sociopaths in terrorist groups, they are far from the majority within any cohesive, and hence long lasting, terrorist group.
Wherever did I say that the number of sociopaths in terrorist groups and their supporters is small? For that matter, wherever did I say that people who find killing others a deeply satisfying act are sociopaths? If you think that, you really should do the research that you've recommended I do: go experience real conflict, and read some history. Add some psychology and sociology in there is well, especially stuff with an evolutionary bent.
My point is not that the number of people who support terrorism is small, but that ALL people who think that terrorism is a good idea are wilfully ignorant of the probability of its success. This includes the majority of human beings, who generally find killing others satisfying, particularly when the others are "OTHER".
What we cannot do is find any rational or empirical justification for killing others in the claim that we are doing so to further some political end, because that history you think I haven't read teaches us again and again and again that terrorism is one of the worst possible ways of actually achieving any political end.
People don't choose terrorism because it works to achieve terrorist group's stated political ends. They choose it because they are plains apes who get a deep feeling of satisfaction from showing the troop across the river that their own troop has more powerful killers.
Ok, please go back to missing my point now. This wouldn't be the Internet if you didn't.
Most terrorists outside the Middle East are doing it for purely secular (usually wanting a seperate state or something simlilar) causes.
No, false, wrong.
No terrorist anywhere cares one whit about achieving any end except to blow things up and kill people. Anyone who argues otherwise must have failed to notice something: that terrorism routinely and completely fails to achieve the stated aims of the people who commit it.
The British are still in Ireland. The Jews still in Palestine. Every one of the groups you list have been around for DECADES and not one of them has come remotely close to getting what they claim to want.
Everyone who belongs to or supports any of those groups has chosen a method that is known to fail pretty much 100% of the time--and some pedantic idiot is already writing a reply to me claiming that the Galambosians or someone were successful with terrorism in the early 1800's, having utterly missed my main point, which is that TERRORISM IS NOT THE METHOD OF CHOICE FOR ANYONE WHO ACTUALLY WANTS TO ACHIEVE THEIR STATED END, BECAUSE TERRORISM ALMOST ALWAYS FAILS.
Choosing terrorism as your method for political change is like choosing winning the lottery as your method of retirement savings. I can't prove it won't work, and you can point to a tiny handful of cases where it has, but you're still an idiot if you do it, and if you're stupid enough to defend anyone doing it or suggesting what they are doing is remotely rational then you need either your mental or your moral capacity checked.
So why do people continue to believe that terrorists are interested in ANYTHING other than simply killing people, given that terrorism is such a failure at achieving anyone's stated ends? We have examples from people like Gandhi as to what actually works, and the first question anyone should ask a terrorist is, "Have you thought of creating an ashram?"
Wake up, people! Terrorism is committed and supported by people for whom KILLING PEOPLE IS A DEEPLY SATISFYING ACT, and nothing more.
Can you give examples of good Exchange replacements? Lack of such is one of the most frequently cited reasons for MS's continued dominance in the enterprise, because while there are trivial replacements for Windows, IE, Office and Outlook, replacing Exchange has been a show-stopper for a lot of places.
OK, maybe I'm missing something - how does writing funny books actively inspire and help others?
You'd be amazed. A friend of mine has two kids (teenage boys) who had no interest in books before I introduced them to Pratchett. Now they read much more, and have whole worlds available to them that they didn't before.
I like Pratchett's work a lot, but my real appreciation for it comes from seeing how it works as a gateway drug for people who might otherwise never walk through the doors of the imagination. That's worth alittle acknowledgement, isn't it?
The LHC turn-on was an important milestone, but given its current state (broken) it's probably better to wait for actual physics results before talking too much more about it.
Maybe the 2009 "breakthrough" in journalism will be to stop using the word "breakthrough" to describe interesting but incremental improvements.
Because Americans and Canadians don't so much engage in cross-border trade as build things together. In the (soon to be defunct) automotive industry it is not uncommon for a finished vehicle to contain parts that have crossed the border a couple of times in the course of manufacturing.
Border impediments would hurt Americans in the only thing they care about: their wallets.
...I'm quite comfortable knowing that unless I'm actively trying to destablize the government, they don't care about what I am up to.
I wonder if George Washington or Thomas Jefferson would have been quite as comfortable about that as you are, seeing as the freedom to be "actively trying to destablize the government" was a right they were willing to die to protect.
American was founded by anti-government terrorists who wanted to create a new government that was far more limited than the one they had been fighting. The limited government they created has been eroded over the past 200+ years and replaced with the Leviathan you have now, which is limited only by its physical inability to actually process all the information on citizens (or should that be 'subjects'?) that it collects.
The answer is two-fold: money, and existing taxonomies are mostly correct.
Biologists have limited resources, so comprehensive reassessment of the entire tree of life based on genetic analysis is going to get done bit by bit over a long time, and we know we're pretty safe going with what we've got in the meantime.
And while enough genetic analysis has been done to confirm traditional taxonomy on quite a few species, it is only the cases where there is a disagreement that it makes the news. In all the other cases they agree, so traditional taxonomy is left intact.
There are a few dramatic cases like this one, though. There are a couple of species of lizard in the Yucatan that have an extra cervical vertebrae that turn out to have independently evolved that way (I prefer the term "independent evolution" rather than "convergent evolution", as the latter may confuse laypeople into thinking that distinct species have somehow become one.)
In those cases, genetic taxonomy wins, but they are always going to be in the minority.
This is the fundamental error of advocates of the Semantic Web: that data have any meaning at all, much less a single "true meaning."
Think of any string of bytes as having the same relationship to a thought as a lossy-compressed document has to the uncompressed document. Lossy compression algorithms depend on shared assumptions, and you can uncompress such a document on the basis of different assumptions than it was compressed with, but it won't give you the same document, which may or may not be what you're looking for.
Words compress thoughts with such an astonishingly high loss rate that it's amazing we can communicate at all. Because people do the uncompression/inference process unreflectively, they come to think that words and thoughts are isomorphic, when nothing could be further from the truth.
Apparently light from my point will not be reaching you for at least a few thousand years, as you've made a suggestion (AI involvement) that is doomed to fail for exactly the reasons I laid out in my original post.
You need to distinguish between document semantics, which is what the SGML purists wanted for HTML, and real world semantics, which is what the Semantic Web people want. It is indeed instructive to note that the document semantics crowd completely failed in their fight to separate the presentation layer from the document model.
The only mechanism that has gained any general traction is CSS, which is about as far from a "real" document-semantics-based styling language as you can get, and I'm betting I'll have to wait a long time for an XSLT-compliant browser.
Regardless of cause, given the failure of semantic markup in such a limited and controlled scope, it is very unlikely that it will succeed with the richness and complexity of all the data in the world.
The Semantic Web is a failed attempt to extend the WWW via "semantic markup", which allows users/editors/etc to tag content (text, images, data) using a standard format that can be read, processed and exchanged by machines which can then give users more useful pointers to stuff that they care about.
The Semantic Web has failed for a bunch of reasons, with many people tending to blame the tools. However, those of us of a particular epistemological bent believe that it is doomed in principle as current conceived because "meaning" is a verb, not an adjective.
"These data mean X" is completely incoherent on this view of meaning, like saying "This smell of orange blossoms has Republican leanings." "Meaning" is simply not an attribute of data, any more than political tendencies are an attribute of scents.
The Semantic Web fails to capture almost everything about the entities that do the meaning (people) but instead is based on the belief that meaning is a property of data. Data inspires meaning, but meaning is something that humans do, and the Semantic Web has no effective mechanism for capturing this, although with sufficient markup by many individuals on the same data it should be possible to do something similar to ROC evaluation of the ways people mean, which would greatly enhance the utility of the Semantic Web.
A colleague who works in GIS pointed out an consequence of this phenomena to me many years ago when he described an experiment involving a bunch of geologists mapping a particular terrain. At the end of the day, after integrating all their inputs, he could tell who mapped where, but not what anybody mapped.
Slashdot Mirror
So it seems the task is coming up with a standard format and enforcing it.
Which will cost FAR more than $100 billion, and be done so badly as to render the system nearly useless.
Ever parse a MAGE-ML doc that turns out to have the actual gene expression values in an "other" or "comments" field? Most "standard formats" are so arcane, complex and counter-intuitive that most people using them can't figure out the appropriate place to put the information.
Furthermore, medical terms change with time as new procedures are introduced and old procedures modified. The proposed format is going to either have to handle that or become the kind of straight-jacket that 501(k) process has been in medical devices.
Anyone contemplating touching any aspect of this project simply MUST read Stephen Flowers' "Software Failure: Management Failure", which is a collection of case studies of failed major software initiatives of just this kind. The book is in fact worth reading for anyone with an interest in why software systems fail, which should be everyone involved in software development.
And, once you get a completely nationalised health system, you effectively have a system equivalent to insurance with the largest possible pool.
Actually, no, because there is virtually no risk involved: EVERYONE gets sick, and EVERYONE dies, and about half of EVERYONE's health care costs come in the last six months of life.
Although costs vary, they don't vary by that much, although the tail of the distribution is long. See figure B1 in this report on Canadian health care costs to see the actual distribution. For something over 70% of the population the average cost of a single hospital stay is less than $10,000, and virtually everyone has a couple of those stays in their lifetime (I've had one despite being in extremely good health generally.)
This is utterly unlike true insurance models--auto, home and term life--where the majority of people who pay premiums never collect a claim.
It is interesting to note that both the Canadian and American health care systems use insurance models, and suffer from similar problems of access and spiralling costs. I believe this is due to the inherent inappropriateness of an insurance model for a service that everyone will need and everyone which has a relatively low variance of total payouts.
A reasonable model of health insurance would deal with catastrophic costs only, say in excess of $10,000 per hospital stay as indicated by these data. As not everyone falls into that category, one could actually use insurance to spread RISK, which is not really possible under an "everyone pays, everyone benefits" model because the tails are not that relevant to the overall cost of the system, so you basically have a situation where there is very little risk to be spread (closer analysis of the numbers could contradict that, but that's my impression from a first look.)
It is, however, some expression of socialism.
Now that the Republican Party is passing out of power in the US we can expect that there will be a good deal less socialism, so whether or not it is desirable (I believe mostly not) there is going to be a good deal less of it now that the more liberal Democratic Party is in control.
I've never been clear exactly WHY Americans call their Socialist Party "the Republican Party". This is the party that has overseen massive growth in government both in responsibilities and costs, intrusive imposition of the federal government into areas normally reserved for the states or the people, and huge transfers of risk "in solidarity" from private individuals and organizations to the public.
It has capped all of this by actually taking ownership of significant parts of your financial system, which must in today's world be counted as firmly amongst "the means of production." Now that the liberal Democratic Party is replacing the socialist Republican party perhaps free market solutions will be prescribed for some of the things that ails the US, like allowing badly-run businesses that make products no one wants to FAIL.
They can protest all they like but no Israeli is going to see it.
I don't recall there being any TV cameras at Amritsar, yet somehow the word got out.
The cowardly, craven apologists for terrorist violence of the kind being committed by BOTH SIDES of the Israeli-Palestinian conflict always have to make trivially false claims of this kind to support their case, but you aren't fooling anyone but your fellow cowards, who are looking for any excuse to deny the trivial truth that non-violent RESISTANCE is almost always more effective and more efficient that violent ATTACK.
Note that the above paper does not mention the "wildly speculative" spiraling magnetic fields idea.
But this is /., where no one cares about science unless it is wildly speculative.
Good critique of Lerner: http://www.astro.ucla.edu/~wright/lerner_errors.html Dunno why the summary mentions him at all.
You mean like whitespace-based indentation a la Fortran?
Unlike all those OTHER languages, whose indentation isn't based on whitespace?
Even if your quip was coherent, it is a-historical. Any indentation other than the basic first six columns was forbidden prior to the whole "structured FORTRAN" thing in the late '70's, and that was just a convenience for humans trying to read the code. Scoping in FORTRAN has always been a matter of keywords, not whitespace, and in fact the language explicitly ignores even whitespace that one might wish it didn't.
For example, the famous one-character bug that blew up an American Venus probe in the '70's was due in part to FORTRAN's removal of whitespace in numerical constants during compilation, so the loop header:
DO I = 1. 10 WHILE
became:
DO I = 1.10 WHILE
"I" is of course an integer, so this is equivalent to:
DO I = 1 WHILE
and rather than the list of ten items being fully initialized only the first one got handled. If the "." had been a "," like it was supposed to, everything would have been fine. (I haven't written FORTRAN for a decade or so, so my syntax in the above may not be quite correct, but you get the idea.)
With regard to Python, getting over the "oh my god it's full of whitespace" thing was an issue for me in adopting the language. Having worked in SGML-based text processing in the '90's I knew only too well that that "whitespace is not actually evil--it is just misunderstood." But once you get over it, it becomes a matter of habit just like curly brackets did when I moved from FORTRAN to C in the late '80's.
As another person replying to my original comment said, Python + scipy + rpy is an excellent environment for routine scientific programming, and if you add wxPython into the mix you've got a pretty good application development language too, that is deservedly getting a lot of uptake in the scientific community.
(a) it allows for easy access of Fortran and C library routines
(b) it allows you to pass large blobs of data by name
(c) it makes it easy to pass data to and from your own compiled C and Fortran routines
So, it's exactly like Python, except with an outdated 1970's syntax that was frankly pretty weird to start with :-)
I've used R, and found it useful for some of its relatively esoteric capabilities, but currently use it almost exclusively via rpy now, the Python binding to R.
Furthermore, I've been using it less in recent years as the native statistical capability of Python has continued to improve. I can appreciate that people who work strictly in data analysis could find R an appropriate tool, but as someone whose work spans multiple areas, from analysis to application design and development, R is too limiting a tool, and using it always feels a little alien and weird.
Reducing package size is green and it costs less to produce. Why increase the price if cost is lowered?
The problem is that faux environmentalists have for decades tried to convince people to change they way they live based on a kind of puritanical moral argument. They want people to change to become supposedly more morally pure through self-sacrifice, but they don't have the guts to say that--instead they hide behind the genuine need to protect the environment.
This propaganda campaign has been so successful that "green" as been equated in the public's mind with "costly, difficult, limited and painful," to the extent that businesses sometimes reject green solutions out of hand because of the up-front perception that "green costs more."
In reality, the underlying economics of genuinely green technology is almost always better than non-green tech, unless the non-green tech is specifically exploiting a loophole in the system of property by dumping effluent into the commons, which is becoming more difficult due to better protection on private property around the world.
Even the cases where people are claiming in this discussion that "green is not always cheaper" they are typically asking us to look at the problem in a completely myopic way, by considering only exactly the same industrial process that is currently being used and post-processing the effluent from it, which is in most cases silly compared to re-engineering the process so that the amount or toxicity of effluent is reduced.
Green companies have been doing this kind of process re-engineering for DECADES, with the primary goal of saving money. Look at the well-known example of Interface. And if an older company can't afford to replace existing plant... well, let 'em get competed into the ground by newer companies that can.
The opposite of "green" is "wasteful", and what economically sane company is in favour of being wasteful?
"Virtually no waste" translates into "Japan believes it has correctly accounted for all the plutonium in its reprocessing network even though accounting errors have resulted in a thousand kilograms going missing."
Reprocessing is clearly a desirable thing to do from all kinds of standpoints, but the issues of safe transport for the cores and actually maintaining a sufficiently accurate inventory on the fissile material create huge practical problems.
Remember, over a decade or two you will be moving millions of kg of material around for reprocessing, so if you inventory control is 99.9% accurate you will still have enough missing plutonium to make a hundred-odd nice little nuclear bombs.
Average temperature predictions scary?
Only to anyone who understands that temperature is an intensive thermodynamic quantity and therefore cannot be meaningfully averaged in an inhomogeneous medium such as the atmosphere. It is a bit scary to see a meaningless number continually used as the basis for a massive power grab.
Atmospheric heat content can be averaged, if we had a clue what it was, and a surrogate average temperature generated from that. As near as I can tell all global temperature records of note are dry-bulb temperatures which are, by themselves, thermodynamically useless for this sort of averaging.
Global OCEAN temperatures are rising, and that should be a matter of concern because that is a clear indicator of increasing ocean heat content, but as I said, it is a bit scary to see people who believe absolutely in their own righteousness base their arguments on a literally meaningless number.
And the point about complex models failing is very well taken. The climate system is far more complex and we have far less information about it than the financial system, and anyone relying on computer models for their beliefs about climate systems ought to be willing to invest everything they own based computer models of the financial system.
Any takers?
Nocera explores the age-old debate between those who assert that the best decisions are based on quantification and numbers, and those who base their decisions on more subjective degrees of belief about the uncertain future.
Why is anyone still making this distinction, as we now know that the only self-consistent numerical representation of risk follows directly from our subjective degrees of belief about the uncertain future? Furthermore, we have known this for over a generation... isn't it about time that the knowledge start filtering into the popular discourse?
While Bayesian methods are not always all that useful for practical problems (I use them on occasion in my work) the conceptual foundations and deeper understanding of the nature of plausible reasoning and its relation to probability theory needs to be more widely understood.
One of the big take-home messages from the Bayesian revolution is that probability theory is nothing but quantification of what we do subjectively, insofar as our subjective impressions are self-consistent, so the only people who are still debating quantitative vs subjective approaches as such are people who do not understand the question.
As such, the implementation is hardware-dependent, which is why there isn't a standard implementation of this function for Windows CE.
It is nevertheless the case, as others here have pointed out, that the correct way of handling date problems is to use Julian dates and day numbers, for which all these algorithms exist in well-documented forms. Getting the Julian Day number in this case should be trivial, as the function gets the number of days since January 1st 1980 as an argument, so it is just a matter of adding that to the Julian Day of the start date.
If you can't use a third-party calender/time lib (which is obviously the preferred solution given the hellish complexity and enormous subtlety of correct date/time representation) year/month/day divisions should be kept STRICTLY in the UI, and all internal date representations and calculations should be in terms of Julian Day or Date.
It is global warming, as the global average temperature has increased, as has the average temperature of most locations on Earth.
What is this 'global average temperature' of which you speak? Temperature is an intensive thermodynamic property and as such cannot be averaged meaningfully in an inhomogeneous medium like the atmosphere.
Global atmospheric heat content is meaningful. Global average temperature is at best ambiguous and at worst meaningless. It is perfectly possible for the temperature to fall everywhere on Earth and the heat content of the atmosphere to increase.
Global ocean temperature, now... that's meaningful, and increasing, but I would take global warming a lot more seriously if the people who are all het up about it showed an even rudimentary grasp of basic physics, rather than prating on about that nominal effects of CO2 as if you'd have to be a moron to be sceptical about their magnitude.
A sceptic about the magnitude of the effect of CO2 on global heat content may be reasonable. A person who puts there argument in terms of a problematic quality such as global average temperature is not.
Furthermore, since every single story we have seen recently about global warming has announced loudly and clearly that climate scientists have no clue whatsoever about what is actually happening in the climate, why should we trust them with geo-engineering?
Surely everyone here has noticed the spate of stories in the past year or two saying that arctic ice is melting far faster than climate scientists expected or that glaciers are retreating far faster than climate scientists expected or that some other climate system is changing far faster than climate scientists expected?
Each of these stories, if read by someone who hadn't bought into a religious belief in the infallibility of climate scientists, would be taken as clear-cut evidence that climate science as a whole is terrible at predicting anything to do with climate, even when they deign to make predictions about physical qualities, rather than made-up nonsense terms like "global average temperature."
While there may be a larger than average representation of sociopaths in terrorist groups, they are far from the majority within any cohesive, and hence long lasting, terrorist group.
Wherever did I say that the number of sociopaths in terrorist groups and their supporters is small? For that matter, wherever did I say that people who find killing others a deeply satisfying act are sociopaths? If you think that, you really should do the research that you've recommended I do: go experience real conflict, and read some history. Add some psychology and sociology in there is well, especially stuff with an evolutionary bent.
My point is not that the number of people who support terrorism is small, but that ALL people who think that terrorism is a good idea are wilfully ignorant of the probability of its success. This includes the majority of human beings, who generally find killing others satisfying, particularly when the others are "OTHER".
What we cannot do is find any rational or empirical justification for killing others in the claim that we are doing so to further some political end, because that history you think I haven't read teaches us again and again and again that terrorism is one of the worst possible ways of actually achieving any political end.
People don't choose terrorism because it works to achieve terrorist group's stated political ends. They choose it because they are plains apes who get a deep feeling of satisfaction from showing the troop across the river that their own troop has more powerful killers.
Ok, please go back to missing my point now. This wouldn't be the Internet if you didn't.
Most terrorists outside the Middle East are doing it for purely secular (usually wanting a seperate state or something simlilar) causes.
No, false, wrong.
No terrorist anywhere cares one whit about achieving any end except to blow things up and kill people. Anyone who argues otherwise must have failed to notice something: that terrorism routinely and completely fails to achieve the stated aims of the people who commit it.
The British are still in Ireland. The Jews still in Palestine. Every one of the groups you list have been around for DECADES and not one of them has come remotely close to getting what they claim to want.
Everyone who belongs to or supports any of those groups has chosen a method that is known to fail pretty much 100% of the time--and some pedantic idiot is already writing a reply to me claiming that the Galambosians or someone were successful with terrorism in the early 1800's, having utterly missed my main point, which is that TERRORISM IS NOT THE METHOD OF CHOICE FOR ANYONE WHO ACTUALLY WANTS TO ACHIEVE THEIR STATED END, BECAUSE TERRORISM ALMOST ALWAYS FAILS.
Choosing terrorism as your method for political change is like choosing winning the lottery as your method of retirement savings. I can't prove it won't work, and you can point to a tiny handful of cases where it has, but you're still an idiot if you do it, and if you're stupid enough to defend anyone doing it or suggesting what they are doing is remotely rational then you need either your mental or your moral capacity checked.
So why do people continue to believe that terrorists are interested in ANYTHING other than simply killing people, given that terrorism is such a failure at achieving anyone's stated ends? We have examples from people like Gandhi as to what actually works, and the first question anyone should ask a terrorist is, "Have you thought of creating an ashram?"
Wake up, people! Terrorism is committed and supported by people for whom KILLING PEOPLE IS A DEEPLY SATISFYING ACT, and nothing more.
Can you give examples of good Exchange replacements? Lack of such is one of the most frequently cited reasons for MS's continued dominance in the enterprise, because while there are trivial replacements for Windows, IE, Office and Outlook, replacing Exchange has been a show-stopper for a lot of places.
OK, maybe I'm missing something - how does writing funny books actively inspire and help others?
You'd be amazed. A friend of mine has two kids (teenage boys) who had no interest in books before I introduced them to Pratchett. Now they read much more, and have whole worlds available to them that they didn't before.
I like Pratchett's work a lot, but my real appreciation for it comes from seeing how it works as a gateway drug for people who might otherwise never walk through the doors of the imagination. That's worth alittle acknowledgement, isn't it?
The LHC turn-on was an important milestone, but given its current state (broken) it's probably better to wait for actual physics results before talking too much more about it.
Maybe the 2009 "breakthrough" in journalism will be to stop using the word "breakthrough" to describe interesting but incremental improvements.
Because Americans and Canadians don't so much engage in cross-border trade as build things together. In the (soon to be defunct) automotive industry it is not uncommon for a finished vehicle to contain parts that have crossed the border a couple of times in the course of manufacturing.
Border impediments would hurt Americans in the only thing they care about: their wallets.
...I'm quite comfortable knowing that unless I'm actively trying to destablize the government, they don't care about what I am up to.
I wonder if George Washington or Thomas Jefferson would have been quite as comfortable about that as you are, seeing as the freedom to be "actively trying to destablize the government" was a right they were willing to die to protect.
American was founded by anti-government terrorists who wanted to create a new government that was far more limited than the one they had been fighting. The limited government they created has been eroded over the past 200+ years and replaced with the Leviathan you have now, which is limited only by its physical inability to actually process all the information on citizens (or should that be 'subjects'?) that it collects.
The answer is two-fold: money, and existing taxonomies are mostly correct.
Biologists have limited resources, so comprehensive reassessment of the entire tree of life based on genetic analysis is going to get done bit by bit over a long time, and we know we're pretty safe going with what we've got in the meantime.
And while enough genetic analysis has been done to confirm traditional taxonomy on quite a few species, it is only the cases where there is a disagreement that it makes the news. In all the other cases they agree, so traditional taxonomy is left intact.
There are a few dramatic cases like this one, though. There are a couple of species of lizard in the Yucatan that have an extra cervical vertebrae that turn out to have independently evolved that way (I prefer the term "independent evolution" rather than "convergent evolution", as the latter may confuse laypeople into thinking that distinct species have somehow become one.)
In those cases, genetic taxonomy wins, but they are always going to be in the minority.
What is this "true 'meaning'" of which you speak?
This is the fundamental error of advocates of the Semantic Web: that data have any meaning at all, much less a single "true meaning."
Think of any string of bytes as having the same relationship to a thought as a lossy-compressed document has to the uncompressed document. Lossy compression algorithms depend on shared assumptions, and you can uncompress such a document on the basis of different assumptions than it was compressed with, but it won't give you the same document, which may or may not be what you're looking for.
Words compress thoughts with such an astonishingly high loss rate that it's amazing we can communicate at all. Because people do the uncompression/inference process unreflectively, they come to think that words and thoughts are isomorphic, when nothing could be further from the truth.
Apparently light from my point will not be reaching you for at least a few thousand years, as you've made a suggestion (AI involvement) that is doomed to fail for exactly the reasons I laid out in my original post.
You need to distinguish between document semantics, which is what the SGML purists wanted for HTML, and real world semantics, which is what the Semantic Web people want. It is indeed instructive to note that the document semantics crowd completely failed in their fight to separate the presentation layer from the document model.
The only mechanism that has gained any general traction is CSS, which is about as far from a "real" document-semantics-based styling language as you can get, and I'm betting I'll have to wait a long time for an XSLT-compliant browser.
Regardless of cause, given the failure of semantic markup in such a limited and controlled scope, it is very unlikely that it will succeed with the richness and complexity of all the data in the world.
The Semantic Web is a failed attempt to extend the WWW via "semantic markup", which allows users/editors/etc to tag content (text, images, data) using a standard format that can be read, processed and exchanged by machines which can then give users more useful pointers to stuff that they care about.
The Semantic Web has failed for a bunch of reasons, with many people tending to blame the tools. However, those of us of a particular epistemological bent believe that it is doomed in principle as current conceived because "meaning" is a verb, not an adjective.
"These data mean X" is completely incoherent on this view of meaning, like saying "This smell of orange blossoms has Republican leanings." "Meaning" is simply not an attribute of data, any more than political tendencies are an attribute of scents.
The Semantic Web fails to capture almost everything about the entities that do the meaning (people) but instead is based on the belief that meaning is a property of data. Data inspires meaning, but meaning is something that humans do, and the Semantic Web has no effective mechanism for capturing this, although with sufficient markup by many individuals on the same data it should be possible to do something similar to ROC evaluation of the ways people mean, which would greatly enhance the utility of the Semantic Web.
A colleague who works in GIS pointed out an consequence of this phenomena to me many years ago when he described an experiment involving a bunch of geologists mapping a particular terrain. At the end of the day, after integrating all their inputs, he could tell who mapped where, but not what anybody mapped.