Oh no you don't like the syntax. That's a great reason to turn away from a technology that has been implemented enough times and had enough research to bring it to where it is today.
If you can get over not liking the syntax, the SQL standard is pretty awesome, as are many (but not all) actual databases that use it. It's powerful enough to let you do some pretty complex queries, it's reasonably easily optimisable (and there is a lot of literature about that) provided you're not using a lousy database engine (like MySQL which can't even handle basic relational calculus planning in a sane way), it's pretty fast, and it offers some great guarantees. I have absolutely no idea what you mean by being difficult to test - either you know how to test or you don't. SQL doesn't get in the way there. You have a production data store and a test data store; you test changes together.
Stored procedures are not so widely used because they're not standard enough. However, they're not hard to use with source code management - you're making the wrong argument.
Your last gripe is fair, and if you are *really* sure you don't need ACID overhead and you have a reasonable alternative database, go for it. You're giving up on all the other research that's gone into the common platform, but that's a tradeoff that might be worth it for some purposes.
Although some of it is frills, doing it underground does largely eliminate seasonal variation and might make security and general environmental control easier. For ordinary server rooms that can be a big expense.
No society is entirely free nor entirely not. When there are challenges that are based on actual facts, we cannot afford to be so afraid of doing anything that we don't deal with them. History is full of examples of challenges societies have had to face, sometimes making pretty big sacrifices to do so (metal rationing, occasionally meal rationing, conscription, etc). Sometimes the sacrifices were inappropriate, but to condemn beforehand even the idea of having society adapt is just foolishness.
I don't consider Al Gore to be doing much more than popularising the already existing consensus. If he were going way out on a limb and breaking with academia, I'd be quite willing to criticise him, as (I hope) any sensible person would. Al Gore is not a scientist, but so far he gets the details mostly right and he's doing an important task of letting people know.
Of course it's imperfect. It's done by real people with egos, doing experiments that might not always be replicable. However, it's the best method we've got - if you find some method of determining facts that's better than the distributed social and methodological tools of academia, tell the world!
Science provides the best approximation we have for truth, and steering is how we move society towards appropriate action. They're separate processes. Recognising a problem is not the same thing as finding the appropriate policy and other changes needed to respond appropriately.
Our long term survival and health as a species is far more important than your petty complaints over taxes. To whatever extent we can steer this, we must.
Occasionally there are legal reasons why *some* research records can't be released, particularly those containing subject information in human subject experiments. That's not likely to be the case there.
For other kinds of information, it can be an issue if people outside of broad academia try to jump in and play politics with the conduct of research - we already have peer review for that, done by people with a solid understanding of statistics and the standards of the field. We don't want our research popularised and then misrepresented by loud and uninformed people - perfectly reasonable studies, conducted and analysed well that would easily pass peer review could easily have some very stupid criticisms attached to them by showmen who cliaim to understand them, and people would be none the wiser.
Openness is great in principle, but the actual conduct and analysis of the studies belongs to the academes worldwide who understand it, not the common people. Voting, pundits, and public debate have no place and are not welcome in the discussion. It is our community - academia, that has driven knowledge.
It's not really a flaw in your line of argument, but the DSM is itself a rather poor thing to cite. If you compare the DSMIV to previous editions, for example, there is a lot to be embarassed about.
If that print edition is vetted by experts, yes. Otherwise no. Citing an encyclopedia is a bad idea. Citing a project like Wikipedia that isn't exactly an encyclopedia is worse. Wikipedia is great for a lot of things and the articles that get enough eyes usually coalesce into something that's reasonably reliable, but it's not as good as traditional research and education.
GRUB is a bootloader. Policywise it makes sense for early boot stuff to live outside partitions, in the MBR and elsewhere. It does not make sense for applications residing in an OS to reach outside their container and scribble on things - they neither do nor should assume they understand what lies outside their box, and it is all-elbows for them to just make assumptions and do it. What if an unusual partitioning scheme is being used (other than traditional PC partitioning)? What if sometime down the road disks are treated in a very different way and they literally can't do stuff like this?
The stuff outside of a partition is always an ill-defined area without a lot of standards. It's traditionally the realm of the person setting up the system combined with ad-hoc decisions of authors of operating systems and the like. This roughly works - sometimes people step on each others toes or need to dance out of somebody else's way, but provided nobody is actually malicious it ends up ok in the end. What does not work is when applications running on any OS demands, particularly with little documentation or choice, to be much ruder than its host OS. What GRUB is doing is acceptable; what these other bits of software are doing is not.
You understanding of the reasoning behind Microsoft's forking of OS/2 is faulty - it was related to their managing to pull Windows up into protected mode as a cute hack, making the switch to MS/IBM OS/2 (and associated "long-term planning for PC platform") less urgent and making the kernel team for the joint product (mostly in MS) suddenly working on a product that had a questionable future. It was a conscious choice by Microsoft. You're also off on your claim on.net
Really though, while MS is not very trustworthy, no corporation is. It'd be strange to expect them to do anything but what makes them a profit.
Instead of consuming and throwing away less and living sustainably, our future is Febreze. If only that were a solution to the floating garbage island in the pacific.
This only counts as "the past" if you ignore that the US has never stopped pursuing blatant self-interest. We can understand or approve of the US pursuing decent and reasonable ideals that it can hold up as good for all humanity. Too often, the US has instead acted in pursuit of things it could not hold up to the world as decent, the kind of things that remain classified long enough for people to neatly categorise them as "in the past" and pretend "we've moved beyond that".
Of course we should try to improve the world, but we need to improve ourselves too. US history is so full of atrocities that you couldn't swing a dead cat around. We need to admit them, talk about them, and change our political system and institutions enough that when we take off the gloves, it's for a good purpose.
Unions are an important engine for decent treatment and pay of workers, and we're overall much better off with them than without. Still, they occasionally make mistakes, and this is one of them. Unions tend to push for a seniority-based payscale - seniority is not a bad foundation for pay, but there should be a performance-based metric as well (many unions support this too) in order to ensure that wayward workers don't bring the profession or union into disrepute or cause poor product. Concern over alienation from labour goes both ways.
They're breakfast if we eat them for breakfast. We may also map our notion of calculation onto their history if we like and if our notion fits. I don't see how that's really an interesting question though - the "are a result of calculation" quality is so abstract that it applies to basically everything in the universe (unless we want to be more precise about the kind of calculation).
I don't see it as anti-empirical, although I do consider subjectivity to be a delusion. Whether it would map well to the concept of being epiphenomenal is an interesting question, but what if it does? Your rejection doesn't hold water in that epiphemnomenality in the philosophical sense isn't a rejection of past events feeding into the "event", just the feeding forward. As for eliminativism, I think a lot hinges on the definitions, but your summary of the idea is more than a bit unusual. On that topic, I think it's more likely and reasonable that mental states are patterns over the physical state (broadly considered, including electrical and chemical state rather than just the gross structure) of the brain system, and I think it's likely we'll understand in great detail what we already understand in rough detail.
In any case, all of this philosophy still is unlikely to do more than provide an intuition on the question - you could "prove" all you like, but if the framework of thought in which your proofs exist isn't rooted firmly enough in reality, you're risking reaching some highly dubious conclusions. What might people do when their logic is self-consistent but the very idea of their system of logic (or some of its axioms) is wrong? The most reliable path to truth we have is through thinking and observing and making pragmatic assumptions that form the basis for our system of knowledge. Philosophy may be decent for other things (value theory, for example), but for truth claims it's weak. Philosophers making the kinds of arguments you are here simply don't have a seat at this table.
I think it's reasonable to have a strong belief that consciousness is computing, based on my general thoughts in philosophy of science and experience with and studies in neuroimaging. I would suggest you look into the state of the field - there's serious progress into making broad maps of brain function, and the characteristics of the neuron strongly suggest reasonable parsimony with computational models.
I would not care to make a guess on timing or methods - I think Kurzweil may have stuck his neck out much further than he should've with his comments were he to be a conservative guesser. Still, I wouldn't say he's clearly wrong. The tools available for neuroimaging have improved remarkably (I worked with fMRI, but the new TMS-based research opens a lot of doors).
I don't think that style of reasoning really holds water. If you like, I'll say that I reject your second step in the 3-step argument you present, but in general I think that reasoning must be subordinate to empiricism. We might hope that our notions of reason would not allow us to prove things that don't line up with reality (or false by some other metric of truth), but if they do, we might have to either discard them or adopt some variant that works better. I reject the idea that philosophy of the sort you use can in principle restrain or predict reality. I reject the notion that the concepts in our heads have reality outside that system, that proofs are or can be anything but ideas that were pragmatically useful for one thing being repurposed for other tasks.
Why should we reject epiphenomenalism or eliminativism are wrong anyhow, if we're to humour the philosophers? It's not the tack I (or other people with experience in neuroimaging) take - I believe that mental states are useful abstractions (visible mainly in self-reflection) over certain patterns of brain states - not exactly eliminativism unless one takes a certain stance on the "deep meaning" of these states. Still, if you want to say "oh, but that'd be X or Y" and you don't tell people why X or Y are invalid, you're not making a great argument - it amounts to name-calling (with the added oddness that most people won't know what you're talking about - equiv of "Oh I just got called a putz, what does that mean? Is it bad?"
Of course, but it also will definitively solve many of these questions. Looking at the traditional philosophical questions is interesting in our spare time, but in the neurosciences it's better to just plow ahead and let that stuff sort itself out as an afterthought. We have the assumptions of science (methodological and perhaps philosophical naturalism), so far they've been shown pragmatically pretty decent, and there's been no coherent challenge yet to the idea that the mind is strictly rooted in phenomena that are theoretically modelable by a computer. The degree of isomorphism to computer software as-we-know-it may be complex though.
Actually, there's a chance we agree, depending on what you mean by "certain kind of computation". Based where we are, thinking of it as "some kind of computation" is very solid ground, but "what kind" is a vexing question for many areas of brain function (the visual cortex is a huge exception to this - it's very well understood).
It has to fit contemporary science because forcing consistency between our models is how we advance them. If we had some object that had "magical" properties and we had physics that didn't cover that object, we'd do well to figure out how to mix them together. Likewise, when we have an unknown object, it makes a lot of sense to assume it's covered by the laws of physics as we know them unless we see strong indicators otherwise. That's how we learn.
Maybe consciousness in the popular usage is 90% delusion. In neuroimaging we're understanding progressively more about the brain.
Maybe a housecat's brain does the equivalent of trig internally for some movement calculations, but you're really asking why can't it do so on a aware level, I think. That's because awareness has taken such a radically different form in our species (and to a lesser degree some others) than in housecats. If cats had more significant awareness and teachability, they probably could do trig in a deeper way. Their brains are not architected for it now though.
Stressing lack of knowledge may be useful for humility, but it's bollocks as an actual approach to learning - we're going to stumble forward, and we already have done so to the point that your thought experiments are inadequate.
I'm not bothered by the government considering a mandate that would advocate for some notion of the public interest. This is the government's job, and it often produces very good results. I don't think it's a good idea in this case though. Many people don't use their phone as a media/convergence device (meaning it's of no theoretical interest to them), and even for those who do, FM radio tends to be of little interest to many technology enthusiasts among them, being more of a legacy stack belonging to prior generations.
I have some sympathy to the idea that, this being an established public venue, it's a good idea to steer devices to supporting it for awhile. It's a bit late for that though, I think. The right time for this may have been when phones were just beginning to act as media devices, or perhaps before, or perhaps there never was a good time. Likewise, there's some value in mandating device access to media channels that are not not strongly controlled by the device manufacturer or the cell provider, but radio is too much on the trailing edge of technology to serve that role.
Slashdot Mirror
Oh no you don't like the syntax. That's a great reason to turn away from a technology that has been implemented enough times and had enough research to bring it to where it is today.
If you can get over not liking the syntax, the SQL standard is pretty awesome, as are many (but not all) actual databases that use it. It's powerful enough to let you do some pretty complex queries, it's reasonably easily optimisable (and there is a lot of literature about that) provided you're not using a lousy database engine (like MySQL which can't even handle basic relational calculus planning in a sane way), it's pretty fast, and it offers some great guarantees. I have absolutely no idea what you mean by being difficult to test - either you know how to test or you don't. SQL doesn't get in the way there. You have a production data store and a test data store; you test changes together.
Stored procedures are not so widely used because they're not standard enough. However, they're not hard to use with source code management - you're making the wrong argument.
Your last gripe is fair, and if you are *really* sure you don't need ACID overhead and you have a reasonable alternative database, go for it. You're giving up on all the other research that's gone into the common platform, but that's a tradeoff that might be worth it for some purposes.
Although some of it is frills, doing it underground does largely eliminate seasonal variation and might make security and general environmental control easier. For ordinary server rooms that can be a big expense.
I'm reasonably sure Slashdot did a story on this underground data center about a year ago, maybe a bit more. I know I've seen these photos before.
No society is entirely free nor entirely not. When there are challenges that are based on actual facts, we cannot afford to be so afraid of doing anything that we don't deal with them. History is full of examples of challenges societies have had to face, sometimes making pretty big sacrifices to do so (metal rationing, occasionally meal rationing, conscription, etc). Sometimes the sacrifices were inappropriate, but to condemn beforehand even the idea of having society adapt is just foolishness.
I don't consider Al Gore to be doing much more than popularising the already existing consensus. If he were going way out on a limb and breaking with academia, I'd be quite willing to criticise him, as (I hope) any sensible person would. Al Gore is not a scientist, but so far he gets the details mostly right and he's doing an important task of letting people know.
Of course it's imperfect. It's done by real people with egos, doing experiments that might not always be replicable. However, it's the best method we've got - if you find some method of determining facts that's better than the distributed social and methodological tools of academia, tell the world!
Science provides the best approximation we have for truth, and steering is how we move society towards appropriate action. They're separate processes. Recognising a problem is not the same thing as finding the appropriate policy and other changes needed to respond appropriately.
Our long term survival and health as a species is far more important than your petty complaints over taxes. To whatever extent we can steer this, we must.
Occasionally there are legal reasons why *some* research records can't be released, particularly those containing subject information in human subject experiments. That's not likely to be the case there.
For other kinds of information, it can be an issue if people outside of broad academia try to jump in and play politics with the conduct of research - we already have peer review for that, done by people with a solid understanding of statistics and the standards of the field. We don't want our research popularised and then misrepresented by loud and uninformed people - perfectly reasonable studies, conducted and analysed well that would easily pass peer review could easily have some very stupid criticisms attached to them by showmen who cliaim to understand them, and people would be none the wiser.
Openness is great in principle, but the actual conduct and analysis of the studies belongs to the academes worldwide who understand it, not the common people. Voting, pundits, and public debate have no place and are not welcome in the discussion. It is our community - academia, that has driven knowledge.
It's not really a flaw in your line of argument, but the DSM is itself a rather poor thing to cite. If you compare the DSMIV to previous editions, for example, there is a lot to be embarassed about.
If that print edition is vetted by experts, yes. Otherwise no. Citing an encyclopedia is a bad idea. Citing a project like Wikipedia that isn't exactly an encyclopedia is worse. Wikipedia is great for a lot of things and the articles that get enough eyes usually coalesce into something that's reasonably reliable, but it's not as good as traditional research and education.
See this for a bit of humour on the topic:
http://www.youtube.com/watch?v=eaADQTeZRCY
GRUB is a bootloader. Policywise it makes sense for early boot stuff to live outside partitions, in the MBR and elsewhere. It does not make sense for applications residing in an OS to reach outside their container and scribble on things - they neither do nor should assume they understand what lies outside their box, and it is all-elbows for them to just make assumptions and do it. What if an unusual partitioning scheme is being used (other than traditional PC partitioning)? What if sometime down the road disks are treated in a very different way and they literally can't do stuff like this?
The stuff outside of a partition is always an ill-defined area without a lot of standards. It's traditionally the realm of the person setting up the system combined with ad-hoc decisions of authors of operating systems and the like. This roughly works - sometimes people step on each others toes or need to dance out of somebody else's way, but provided nobody is actually malicious it ends up ok in the end. What does not work is when applications running on any OS demands, particularly with little documentation or choice, to be much ruder than its host OS. What GRUB is doing is acceptable; what these other bits of software are doing is not.
You understanding of the reasoning behind Microsoft's forking of OS/2 is faulty - it was related to their managing to pull Windows up into protected mode as a cute hack, making the switch to MS/IBM OS/2 (and associated "long-term planning for PC platform") less urgent and making the kernel team for the joint product (mostly in MS) suddenly working on a product that had a questionable future. It was a conscious choice by Microsoft. You're also off on your claim on .net
Really though, while MS is not very trustworthy, no corporation is. It'd be strange to expect them to do anything but what makes them a profit.
Instead of consuming and throwing away less and living sustainably, our future is Febreze. If only that were a solution to the floating garbage island in the pacific.
This only counts as "the past" if you ignore that the US has never stopped pursuing blatant self-interest. We can understand or approve of the US pursuing decent and reasonable ideals that it can hold up as good for all humanity. Too often, the US has instead acted in pursuit of things it could not hold up to the world as decent, the kind of things that remain classified long enough for people to neatly categorise them as "in the past" and pretend "we've moved beyond that".
Of course we should try to improve the world, but we need to improve ourselves too. US history is so full of atrocities that you couldn't swing a dead cat around. We need to admit them, talk about them, and change our political system and institutions enough that when we take off the gloves, it's for a good purpose.
Unions are an important engine for decent treatment and pay of workers, and we're overall much better off with them than without. Still, they occasionally make mistakes, and this is one of them. Unions tend to push for a seniority-based payscale - seniority is not a bad foundation for pay, but there should be a performance-based metric as well (many unions support this too) in order to ensure that wayward workers don't bring the profession or union into disrepute or cause poor product. Concern over alienation from labour goes both ways.
They're breakfast if we eat them for breakfast. We may also map our notion of calculation onto their history if we like and if our notion fits. I don't see how that's really an interesting question though - the "are a result of calculation" quality is so abstract that it applies to basically everything in the universe (unless we want to be more precise about the kind of calculation).
I don't see it as anti-empirical, although I do consider subjectivity to be a delusion. Whether it would map well to the concept of being epiphenomenal is an interesting question, but what if it does? Your rejection doesn't hold water in that epiphemnomenality in the philosophical sense isn't a rejection of past events feeding into the "event", just the feeding forward. As for eliminativism, I think a lot hinges on the definitions, but your summary of the idea is more than a bit unusual. On that topic, I think it's more likely and reasonable that mental states are patterns over the physical state (broadly considered, including electrical and chemical state rather than just the gross structure) of the brain system, and I think it's likely we'll understand in great detail what we already understand in rough detail.
In any case, all of this philosophy still is unlikely to do more than provide an intuition on the question - you could "prove" all you like, but if the framework of thought in which your proofs exist isn't rooted firmly enough in reality, you're risking reaching some highly dubious conclusions. What might people do when their logic is self-consistent but the very idea of their system of logic (or some of its axioms) is wrong? The most reliable path to truth we have is through thinking and observing and making pragmatic assumptions that form the basis for our system of knowledge. Philosophy may be decent for other things (value theory, for example), but for truth claims it's weak. Philosophers making the kinds of arguments you are here simply don't have a seat at this table.
I think it's reasonable to have a strong belief that consciousness is computing, based on my general thoughts in philosophy of science and experience with and studies in neuroimaging. I would suggest you look into the state of the field - there's serious progress into making broad maps of brain function, and the characteristics of the neuron strongly suggest reasonable parsimony with computational models.
I would not care to make a guess on timing or methods - I think Kurzweil may have stuck his neck out much further than he should've with his comments were he to be a conservative guesser. Still, I wouldn't say he's clearly wrong. The tools available for neuroimaging have improved remarkably (I worked with fMRI, but the new TMS-based research opens a lot of doors).
I don't think that style of reasoning really holds water. If you like, I'll say that I reject your second step in the 3-step argument you present, but in general I think that reasoning must be subordinate to empiricism. We might hope that our notions of reason would not allow us to prove things that don't line up with reality (or false by some other metric of truth), but if they do, we might have to either discard them or adopt some variant that works better. I reject the idea that philosophy of the sort you use can in principle restrain or predict reality. I reject the notion that the concepts in our heads have reality outside that system, that proofs are or can be anything but ideas that were pragmatically useful for one thing being repurposed for other tasks.
Why should we reject epiphenomenalism or eliminativism are wrong anyhow, if we're to humour the philosophers? It's not the tack I (or other people with experience in neuroimaging) take - I believe that mental states are useful abstractions (visible mainly in self-reflection) over certain patterns of brain states - not exactly eliminativism unless one takes a certain stance on the "deep meaning" of these states. Still, if you want to say "oh, but that'd be X or Y" and you don't tell people why X or Y are invalid, you're not making a great argument - it amounts to name-calling (with the added oddness that most people won't know what you're talking about - equiv of "Oh I just got called a putz, what does that mean? Is it bad?"
Of course, but it also will definitively solve many of these questions. Looking at the traditional philosophical questions is interesting in our spare time, but in the neurosciences it's better to just plow ahead and let that stuff sort itself out as an afterthought. We have the assumptions of science (methodological and perhaps philosophical naturalism), so far they've been shown pragmatically pretty decent, and there's been no coherent challenge yet to the idea that the mind is strictly rooted in phenomena that are theoretically modelable by a computer. The degree of isomorphism to computer software as-we-know-it may be complex though.
Actually, there's a chance we agree, depending on what you mean by "certain kind of computation". Based where we are, thinking of it as "some kind of computation" is very solid ground, but "what kind" is a vexing question for many areas of brain function (the visual cortex is a huge exception to this - it's very well understood).
It has to fit contemporary science because forcing consistency between our models is how we advance them. If we had some object that had "magical" properties and we had physics that didn't cover that object, we'd do well to figure out how to mix them together. Likewise, when we have an unknown object, it makes a lot of sense to assume it's covered by the laws of physics as we know them unless we see strong indicators otherwise. That's how we learn.
Maybe consciousness in the popular usage is 90% delusion. In neuroimaging we're understanding progressively more about the brain.
Maybe a housecat's brain does the equivalent of trig internally for some movement calculations, but you're really asking why can't it do so on a aware level, I think. That's because awareness has taken such a radically different form in our species (and to a lesser degree some others) than in housecats. If cats had more significant awareness and teachability, they probably could do trig in a deeper way. Their brains are not architected for it now though.
Stressing lack of knowledge may be useful for humility, but it's bollocks as an actual approach to learning - we're going to stumble forward, and we already have done so to the point that your thought experiments are inadequate.
Still, it's very reasonable to believe that it is - what else could it be that fits with modern science?
1) Unauthorised by whom?
2) Didn't a school district try this recently and get some bad press for it?
I'm not bothered by the government considering a mandate that would advocate for some notion of the public interest. This is the government's job, and it often produces very good results. I don't think it's a good idea in this case though. Many people don't use their phone as a media/convergence device (meaning it's of no theoretical interest to them), and even for those who do, FM radio tends to be of little interest to many technology enthusiasts among them, being more of a legacy stack belonging to prior generations.
I have some sympathy to the idea that, this being an established public venue, it's a good idea to steer devices to supporting it for awhile. It's a bit late for that though, I think. The right time for this may have been when phones were just beginning to act as media devices, or perhaps before, or perhaps there never was a good time. Likewise, there's some value in mandating device access to media channels that are not not strongly controlled by the device manufacturer or the cell provider, but radio is too much on the trailing edge of technology to serve that role.
I think you mean CSV