Many many years and beers ago, when I was in high school, I took coursework in calculus and physics. In these courses, some memorization is required for equations and such. However, I remembered just a few things in both and got through. My favorite exam was in physics. What I knew was that
P=MV
F=MA
and concepts from the whole course. I worked on the "equation sheet" first, and took 3 pages to go from
P=MV
to E&M to light&optics, and then collapsed everything back to P=MV (equation for momentum). Using this method, I was able to derive equations needed for the rest of the exam. I worked in a similar fashion on the AP (Advanced Placement) exam, which determined how much, if any, college credit I would get for the coursework.
The best part in all this? The physics teacher was upset and angry, as he "didn't expect [me] to pass the AP exam because [I] didn't memorize all the equations[, but took time to derive them instead]". Heh.
"The student, who was completing a research paper on Communism for Professor Pontbriand's class on fascism and totalitarianism, filled out a form for the request, leaving his name, address, phone number and Social Security number. He was later visited at his parents' home in New Bedford by two agents of the Department of Homeland Security, the professors said."
What better way to learn about fascism and totalitarianism than to live under 'em, eh?
And one more time... YES! Of course... I suppose that we'll all end up being let down and then the sundown clauses will be tacked on to the end of a "don't let kids die" bill of some sort.
If, of course, the wholesale selling of freedom ends up being that covert.
I'd like to say that I believe this will be blasted and subsequently killed off like Clipper, but I think the 'net has become too dilute. Result? I, for one, welcome our (possibly, from TFA) fingerprint-backed TPM.
...is the transiency of information-as-conversation as opposed to information-as-corporate-document. E-mail, as it can be printed out and brought before a judge, is effectively a corporate document. However, a conversation has properties that are transient and ephemeral while also being subjective.
This is what the article gives when it says that
[i]n spring 2006 we will be launching new services such as self-destruct email
If someone told you that you could have $1000000 for nothing except exploiting a system that's getting fucked over anyway, would you really be that bad for taking it?
Yes.
Before I continue, I've got to say that I understand your point and the frustration associated with it (at least, that's what I got out of reading your comment), but two examples follow below that illustrate that, yes, you would be bad for taking advantage of a system flawed in this way.
Example the first, an illustration of Utilitarianism, from a personal anecdote: Four years ago, I had a slip-and-fall injury which slipped 2 disks, broke my tailbone, and gave me short-term sciatica. As the stairs were attached to a building, 12 risers "high", and had a handrail that was torn out, it was a clear case of negligence. All I wanted was an apology. I never got the apology, and ended up filing suit for the following:
(1) Lost wages for the week[1] I was out of work.
(2) Actual medical costs.
(3) Money to cover one MRI exam in the future, should the need arise.
the result was just over $5k or so, which paid my bills and gave me a bit left over for the future. Was it what I could have received? No. Was it what I should have received? Well, although I still cannot sit on concrete, sometimes in life you have to be able to grimace and go on with not much more than a shit happens remark. That builds character.;)
I say that this is an example of utilitarianism because if everyone acted after this fashion, the legal system would be less burdoned and barratry would not exist. The society at a large may not feel so damn entitled and might be more cohesive and productive.
Example the second; a succinct quote:
Evil happens when good men do nothing -- Edmund Burke
By failing to act in a manner which prohibits this sort of behavior, evil-by-proxy is formed. This is bad for the individual, and it is bad for society at large.
[1] After a week of staring at the ceiling, I had to get out of the house and return to life. I should have likely been out for longer, but being bedridden is horribly boring.
As a footnote: Do I have regrets about not "taking them (a state) to the cleaners"? Well, sure - free money is attractive. After all, I sure would like this mortgage to be paid off and so on. However, the regret about taking this far kinder (and fairer) action is far outweighed and tempered by the fact that people should do the right thing, and that this is its own reward. Sounds stupid and hackneyed, maybe, but I'd have a hard time looking at myself in the mirror in a house bought with ill-gotten money. Maybe that's just me, but on the whole I have no regrets - outside of never getting a sincere apology. I should have demanded and forced 'em into giving me that!;)
It now appears that at least 568,200 nameservers have witnessed DNS queries related to the rootkit. How many hosts does this correspond to? Only Sony (and First4Internet) knows...unsurprisingly, they are not particularly communicative....
Hm. I'd like to see Sony pay for all this bandwidth. I know that if I ran any of these networks involved (e.g. backbone or DNS points), I'd be after Sony for the bandwidth used.
I believe they've started doing just this. There was a/. story a couple days ago talking about Singularity (http://research.microsoft.com/os/singularity/) as a new scratch-built OS in development.
Interestingly enough, Tuomas Aura is working on it. I only know (of) him in the steganography domain, but this should be interesting to see Singularity unfold with time.
N
of his Ph.D. Just like the German (likely all over this topic) Jan Hendrik Schön. I'm a big fan of holding people acting as an agent to Science or The Public Knowledge or The Public Good accountable for their actions. If they act in violation of their implicit powers entrusted them by their peers and/or readers (e.g. data theft or total fabrication), then the PhD ought to be revoked.
Just as a Jurispudence Doctorate gives one the ability to publically argue a case on the behalf - or at the behest - of another, so too does a Doctorate of Philosophy imply that the bearer ought act in accord with the advancement of science (and the subsequent betterment of man's plight).
No... it would appear to work more like a salt. We're FINALLY moving to the point of being able to get over the "stolen fingerprint" stage of biometrics. Shame that I didn't get around to publishing 2 years ago. Oh well.
Incidentally, it is likely worth pointing something out here. Psychologists tend to fall into two very broad camps over this point:
The Man-as-evolved-stimulus-response-Units (Learning Theory) group will say that any action can be controlled with S-R chains. This is very sexy as it is possible to model this kind of interraction and behavior.
Or...
The representationalist camp is much more inclined to look at mind as a super-position of brain, and be equally inclined to pull the two elements apart and examine mind.
I fall into the second camp, and always have. As such, it makes me pleased to see that we may be - just MAY be - starting to come around to the conclusion that an Artificial Neural Network is JUST that - an ARTIFICIAL approximation (simulation) of learning. It's handy, yes. But it doesn't encapsulate what we are DOING "inside the black box".
Now for the meat... there are certain things that the brain does, and does exceptionally well. Indeed, it does this because we have been hard-wired for this over the course of our evolution. What does Man - Man that has existed, until recently, at relatively slow speeds and needed to directly or indirectly hunt - need most to survive? Well, we need to be expert at speech-processing so that we can pack-hunt and send each other signals, warnings, and information. We also need to be expert at visual discrimination - especially with respect to face recognition - so that we can see danger, identify cave-pal Ug from enemy Oot, and track a moving prey item.
Now... what do you get when you combine these two? Why, you get an expert system that is BUILT to be an always-on system. Whee. Hence, yes you will see the two systems also interracting and experiencing possible dissonance and interference. In fact, check out some work that Michelle Miller (Northern Arizona University) did a few years back. I did a lot of the back-end for it (hopefully it published, you never know with these things), and it essentially showed that, yes, cognition is not a FSM (Finite State Machine).
That reminds me... let's back off this a step... IS THIS NEWS to anyone? With the exception of chunking (as the cognitive load presented here is VERY low), the Man-cum-FSM view of Mind is inflexible to such an extreme so as to fail to account for a lot of the "bug-in-the-box" kinds of things that we see out in the world.
Argh. This should be a LOW-hanging fruit, I think I'll maybe cherry-pick this into a follow-up article.
Since pedantic mode has apparently been left on...
For the record, real neurons are connected by synapses (missing on the picture) formed between the axon (missing on the picture) of one neuron and the dendritic tree (missing on the picture, which is a real shame because they can be made to look very cool) of another.
No. Real neurons may have axodendritic, axosomatic, axosynaptic, axoaxonic, or axosecretory connections. It is more complex than even this, but this will suffice to demonstrate that are wrong and have vastly over-simplified your rebuttal. The thrust of which, I think, was that the artist had made a real botch of displaying neurons in situ as you state that
Whoever did this clearly has no clue about neurons.
And neither do you. And neither do I, fully. And there you have it - it's art. Take it or leave it (though I'd personally leave it, it's a good first effort).
... then change institutions. Here (in the midwest at a Research I school), the university takes a pretty good appraoch to the problem and typically works with you (the inventor) to do a 50-50 split on things that you have come out with while working on their network/machines/time/etc. This seems pretty sane to me, as everyone wins - YOU may or may not have had the time and legal muscle to get a patent in and approved, and THEY wouldn't have had the invention in the first place.
When going for either P&T or hiring, ASK about these issues, and if you don't like the answer, tell them and move along.
No, Indians are starting to become more expensive than Slovak programmers, at least. Eastern Europe, from my earlier comment, is starting to see an upswing in outsourced work from at least one huge, global (US) company.
This situation (outsourcing for the wrong reasons) seems to have its own "bubble", and it seems like we're headed for its imminent explosion. To wit: the phrase is two words, and starts with "cluster".
I say "for the wrong reasons" because, last year, I was exposed to some thinking (Business Proces Re-engineering) that seemed to make sense - that a company could outsource to allow its internal experts continue to grow and expand their area(s) of expertise, without having to run the day-to-day (mundane) operations. I like to use the writing of software as an example. Given that a company wants to write software, it must first be engineered/architected (this is not "throwing code together"!), and then documentation written up, and finally the software can be developed ("coded") and final deliverable documentation developed and delivered. Given the best possible outcome of outsourcing, it is entirely plausible to keep the engineers and architects at a high-level, while giving compartmentalized code specifications to a third party for the purpose of code generation. This final stage, of course, represents the actual outsourcing.
This, I feel, gives a good approach to outsourcing insofar as it not always having to be an onerous or negative thing for a company. However, the problem as I have seen it has been that very seldom does outsourcing get done in this fashion. In fact, from what I understand from a huge company's perspective (global), India is now becoming too expensive, and so software development (including engineering and architecture/protocol design and specification and documentation) is now being done in Eastern European countries. This is, in itself, fine and well, but I believe that it is leading us to a brain drain in this country, and is bad for everyone all 'round.
I honestly appreciate dialogue on this, and if I'm in the wrong, I'd appreciate hearing back. I am by no means an opponent of globalization (or, indeed, of a more equalized global society), as this would allow me, the worker, an opportunity to choose where to work based on culture, dialect, language, and environment/scenery rather than only the knowledge that I can earn 10x to 20x more working and living in country X over country Y. Unfortunately, I do not see this happening in my lifetime, as it appears that businesses run the global show and we (the West) seem to desire this to continue. Oh well. I also do not see enough inertia left on the 'net to create the kind of critical mass to make a difference anymore. Unlike, say, 1994 and the Clipper chip, when the 'net seemed to be less dilute and be able to create a critical mass of outrage that eventually sank the roll-out effort.
So noted in my response to myself - however, I am still curious as to what provisions have been made for fall-out. This is now between IRB and the subjects, if none were made and IRB signed off on it (at least, that's how I see it).
the IRB Human Subjects form. This was a deception study, clearly. The fact that this was so is fine, but running things like this past IRB requires a strict and rigid understanding between the PIs and the IRB. Also, AFAIK, provisions must be made for "repairing" anyone who is damaged by the research - even if it is incidental (e.g. your research was only "the last straw").
I'd like to see the IRB to determine how things are done at IU. Without seeing the form, I really cannot comment on weather what was done was "ethical" or not. It is a blisteringly simple experiment, and if they can get a paper out of it, it'd be what we call "low-hanging fruit".
However, if no IRB approval was received, then this is an entirely different matter. IRB approval == crap hits IRB if things go horribly wrong. No IRB approval == crap hits PIs and all associated if things go horribly (or publicly) wrong.
My $0.02 on why JPEG *does* serve and have a purpose is this: Due to the fact that JPEG makes use of a quantizer and has frequency-space (thanks to DCT), it is a really nice candidate for image-based steganography. Yes, I say "candidate" knowing full well that MANY MANY MANY methods actually exist for this, thanks.
Check out the Brandeis GOLEM project - they do some model evolution using an ANN as a learning methodology. I know that it's not what you're looking for, but hooking this to sensor input might be very nearly the solution you are looking for. (Yes, much easier said than done)
Slashdot Mirror
The best part in all this? The physics teacher was upset and angry, as he "didn't expect [me] to pass the AP exam because [I] didn't memorize all the equations[, but took time to derive them instead]". Heh.
What better way to learn about fascism and totalitarianism than to live under 'em, eh?
Yes, I'm feeling sardonic today.
And one more time ... YES! Of course ... I suppose that we'll all end up being let down and then the sundown clauses will be tacked on to the end of a "don't let kids die" bill of some sort.
If, of course, the wholesale selling of freedom ends up being that covert.
I'd like to say that I believe this will be blasted and subsequently killed off like Clipper, but I think the 'net has become too dilute. Result? I, for one, welcome our (possibly, from TFA) fingerprint-backed TPM.
Anyone have any silly putty I can borrow?
...is the transiency of information-as-conversation as opposed to information-as-corporate-document. E-mail, as it can be printed out and brought before a judge, is effectively a corporate document. However, a conversation has properties that are transient and ephemeral while also being subjective.
This is what the article gives when it says that
I'm not a lawyer, but I have a crack legal defense strategy for him, and it even comes down to one word...
Satire.
;)
Yes.
Before I continue, I've got to say that I understand your point and the frustration associated with it (at least, that's what I got out of reading your comment), but two examples follow below that illustrate that, yes, you would be bad for taking advantage of a system flawed in this way.
Example the first, an illustration of Utilitarianism, from a personal anecdote: Four years ago, I had a slip-and-fall injury which slipped 2 disks, broke my tailbone, and gave me short-term sciatica. As the stairs were attached to a building, 12 risers "high", and had a handrail that was torn out, it was a clear case of negligence. All I wanted was an apology. I never got the apology, and ended up filing suit for the following:
the result was just over $5k or so, which paid my bills and gave me a bit left over for the future. Was it what I could have received? No. Was it what I should have received? Well, although I still cannot sit on concrete, sometimes in life you have to be able to grimace and go on with not much more than a shit happens remark. That builds character.I say that this is an example of utilitarianism because if everyone acted after this fashion, the legal system would be less burdoned and barratry would not exist. The society at a large may not feel so damn entitled and might be more cohesive and productive.
Example the second; a succinct quote:
By failing to act in a manner which prohibits this sort of behavior, evil-by-proxy is formed. This is bad for the individual, and it is bad for society at large.
[1] After a week of staring at the ceiling, I had to get out of the house and return to life. I should have likely been out for longer, but being bedridden is horribly boring.
As a footnote: Do I have regrets about not "taking them (a state) to the cleaners"? Well, sure - free money is attractive. After all, I sure would like this mortgage to be paid off and so on. However, the regret about taking this far kinder (and fairer) action is far outweighed and tempered by the fact that people should do the right thing, and that this is its own reward. Sounds stupid and hackneyed, maybe, but I'd have a hard time looking at myself in the mirror in a house bought with ill-gotten money. Maybe that's just me, but on the whole I have no regrets - outside of never getting a sincere apology. I should have demanded and forced 'em into giving me that! ;)
I believe they've started doing just this. There was a /. story a couple days ago talking about Singularity (http://research.microsoft.com/os/singularity/) as a new scratch-built OS in development.
Interestingly enough, Tuomas Aura is working on it. I only know (of) him in the steganography domain, but this should be interesting to see Singularity unfold with time.
N
of his Ph.D. Just like the German (likely all over this topic) Jan Hendrik Schön. I'm a big fan of holding people acting as an agent to Science or The Public Knowledge or The Public Good accountable for their actions. If they act in violation of their implicit powers entrusted them by their peers and/or readers (e.g. data theft or total fabrication), then the PhD ought to be revoked.
Just as a Jurispudence Doctorate gives one the ability to publically argue a case on the behalf - or at the behest - of another, so too does a Doctorate of Philosophy imply that the bearer ought act in accord with the advancement of science (and the subsequent betterment of man's plight).
My two cents.
Seemed that way to me; I wish I had "insightful" or "Read the Article" points to give out right now. :-/
No ... it would appear to work more like a salt. We're FINALLY moving to the point of being able to get over the "stolen fingerprint" stage of biometrics. Shame that I didn't get around to publishing 2 years ago. Oh well.
It's Clipper all over again.
Only this time, the Internet is too dilute to react and appropriately halt this.
Welcome to today, I guess.
Incidentally, it is likely worth pointing something out here. Psychologists tend to fall into two very broad camps over this point:
Or...I fall into the second camp, and always have. As such, it makes me pleased to see that we may be - just MAY be - starting to come around to the conclusion that an Artificial Neural Network is JUST that - an ARTIFICIAL approximation (simulation) of learning. It's handy, yes. But it doesn't encapsulate what we are DOING "inside the black box".
No fucking shit.
Now for the meat ... there are certain things that the brain does, and does exceptionally well. Indeed, it does this because we have been hard-wired for this over the course of our evolution. What does Man - Man that has existed, until recently, at relatively slow speeds and needed to directly or indirectly hunt - need most to survive? Well, we need to be expert at speech-processing so that we can pack-hunt and send each other signals, warnings, and information. We also need to be expert at visual discrimination - especially with respect to face recognition - so that we can see danger, identify cave-pal Ug from enemy Oot, and track a moving prey item.
Now ... what do you get when you combine these two? Why, you get an expert system that is BUILT to be an always-on system. Whee. Hence, yes you will see the two systems also interracting and experiencing possible dissonance and interference. In fact, check out some work that Michelle Miller (Northern Arizona University) did a few years back. I did a lot of the back-end for it (hopefully it published, you never know with these things), and it essentially showed that, yes, cognition is not a FSM (Finite State Machine).
That reminds me ... let's back off this a step ... IS THIS NEWS to anyone? With the exception of chunking (as the cognitive load presented here is VERY low), the Man-cum-FSM view of Mind is inflexible to such an extreme so as to fail to account for a lot of the "bug-in-the-box" kinds of things that we see out in the world.
Argh. This should be a LOW-hanging fruit, I think I'll maybe cherry-pick this into a follow-up article.
No. Real neurons may have axodendritic, axosomatic, axosynaptic, axoaxonic, or axosecretory connections. It is more complex than even this, but this will suffice to demonstrate that are wrong and have vastly over-simplified your rebuttal. The thrust of which, I think, was that the artist had made a real botch of displaying neurons in situ as you state that
And neither do you. And neither do I, fully. And there you have it - it's art. Take it or leave it (though I'd personally leave it, it's a good first effort).
... then change institutions. Here (in the midwest at a Research I school), the university takes a pretty good appraoch to the problem and typically works with you (the inventor) to do a 50-50 split on things that you have come out with while working on their network/machines/time/etc. This seems pretty sane to me, as everyone wins - YOU may or may not have had the time and legal muscle to get a patent in and approved, and THEY wouldn't have had the invention in the first place.
...
When going for either P&T or hiring, ASK about these issues, and if you don't like the answer, tell them and move along.
Hope this helps a bit
No, Indians are starting to become more expensive than Slovak programmers, at least. Eastern Europe, from my earlier comment, is starting to see an upswing in outsourced work from at least one huge, global (US) company.
One is a phrase, the other an observation.
This situation (outsourcing for the wrong reasons) seems to have its own "bubble", and it seems like we're headed for its imminent explosion. To wit: the phrase is two words, and starts with "cluster".
I say "for the wrong reasons" because, last year, I was exposed to some thinking (Business Proces Re-engineering) that seemed to make sense - that a company could outsource to allow its internal experts continue to grow and expand their area(s) of expertise, without having to run the day-to-day (mundane) operations. I like to use the writing of software as an example. Given that a company wants to write software, it must first be engineered/architected (this is not "throwing code together"!), and then documentation written up, and finally the software can be developed ("coded") and final deliverable documentation developed and delivered. Given the best possible outcome of outsourcing, it is entirely plausible to keep the engineers and architects at a high-level, while giving compartmentalized code specifications to a third party for the purpose of code generation. This final stage, of course, represents the actual outsourcing.
This, I feel, gives a good approach to outsourcing insofar as it not always having to be an onerous or negative thing for a company. However, the problem as I have seen it has been that very seldom does outsourcing get done in this fashion. In fact, from what I understand from a huge company's perspective (global), India is now becoming too expensive, and so software development (including engineering and architecture/protocol design and specification and documentation) is now being done in Eastern European countries. This is, in itself, fine and well, but I believe that it is leading us to a brain drain in this country, and is bad for everyone all 'round.
I honestly appreciate dialogue on this, and if I'm in the wrong, I'd appreciate hearing back. I am by no means an opponent of globalization (or, indeed, of a more equalized global society), as this would allow me, the worker, an opportunity to choose where to work based on culture, dialect, language, and environment/scenery rather than only the knowledge that I can earn 10x to 20x more working and living in country X over country Y. Unfortunately, I do not see this happening in my lifetime, as it appears that businesses run the global show and we (the West) seem to desire this to continue. Oh well. I also do not see enough inertia left on the 'net to create the kind of critical mass to make a difference anymore. Unlike, say, 1994 and the Clipper chip, when the 'net seemed to be less dilute and be able to create a critical mass of outrage that eventually sank the roll-out effort.
So noted in my response to myself - however, I am still curious as to what provisions have been made for fall-out. This is now between IRB and the subjects, if none were made and IRB signed off on it (at least, that's how I see it).
D'oh - yup, they were filled out:
(from http://www.idsnews.com/subsite/story.php?id=29400)
I still wonder, though, how they (Human Subjects Committee) provisioned for possible fall-out.
the IRB Human Subjects form. This was a deception study, clearly. The fact that this was so is fine, but running things like this past IRB requires a strict and rigid understanding between the PIs and the IRB. Also, AFAIK, provisions must be made for "repairing" anyone who is damaged by the research - even if it is incidental (e.g. your research was only "the last straw").
I'd like to see the IRB to determine how things are done at IU. Without seeing the form, I really cannot comment on weather what was done was "ethical" or not. It is a blisteringly simple experiment, and if they can get a paper out of it, it'd be what we call "low-hanging fruit".
However, if no IRB approval was received, then this is an entirely different matter. IRB approval == crap hits IRB if things go horribly wrong. No IRB approval == crap hits PIs and all associated if things go horribly (or publicly) wrong.
Hopefully the forms were filled out.
My $0.02 on why JPEG *does* serve and have a purpose is this: Due to the fact that JPEG makes use of a quantizer and has frequency-space (thanks to DCT), it is a really nice candidate for image-based steganography. Yes, I say "candidate" knowing full well that MANY MANY MANY methods actually exist for this, thanks.
Hmmm ... I think there is. At least as of a couple years ago, there was. I seem to recall seeing something about "PVM For Windows" somewhere.
Check out the Brandeis GOLEM project - they do some model evolution using an ANN as a learning methodology. I know that it's not what you're looking for, but hooking this to sensor input might be very nearly the solution you are looking for. (Yes, much easier said than done)