I developed a not (yet?) open-sourced POS program, and i could go on and on about the benefits of using linux on cash registers - one of them being that it just makes THAT much sense that it's relatively easy to convince managements to take that 'risk' (license fees, secutiry, etc.). some recommendations/problems i ran into:
printing is a bitch - even though basically all POS printers support some nice standards like ESC/POS, and are trivial to control, text-mode is all you'll get (AFAIK). No 'real' printer drivers for linux available anywhere i looked. This is usually not too bad a problem, but some people like rasterized text.
barcodes, as others have pointed out, are trivial - they just 'look like' a keyboard although some models can be connected to a serial port, too
even though it's harder for a user to screw up the OS installation, i'd still recommend a standard image you just roll out to new registers, and which run in read-only mode. additionally, you can run the program as directly from X as a kind of 'window manager', which effectively turning your PC into a kiosk-like macine
use the register for only that. ans get a really cheap and crappy box. those PCs suffer a lot, so why waste a nice box on this?
remote support via SSH is cool (even from a cellphone with a crappy 3rd world GPRS connection!), but i dearly miss something like VNC running in the same mode as it does on windows, which allows you to see what's on the screen i that very moment. Anybody know a program for doing this on linux? remote support is crucial, since people can't usually call their neighborhood geek fr fixing linux problems
there are a lot of FOSS pos programs on the internet, but none of them fitted my bill, and developing your own is surprisingly easy, and you get exactly what you need. give it shot
i always recommend using install-less programs running from a central server - otherwise updating can be a PITA
some credit card processor's 'merchant accounts', which are usually used for online stores, can be used to interface your own little program so that it can accept credit cards. but just using a separate 'swipe device' is usually easier.
See? That's something i hate about PHP - i'm a kind of sloppy perl-coder myself, but this whole shielding-bad-code-from-the-evil-internet mentality has to stop. Why would apache, using a special module, have to baby-sit PHP applications? The correct way of doing this is to simply always check the input you're getting, and only letting known good values pass - especially if you're going to use some function that might pull data from the internet. Obviously, adding additional layers of security around applications that are already as safe as possible doesn't cause harm, but i've just seen PHP coders get away with the belief that their "safe mode" or some other patch or hack will protect them from evil, or considerably increase the security of their programs, even though their programs are just deeply flawed in the core. Often, they just run around like headless chicken screaming that the world is going to end if you don't switch on some RewriteRules (they don't understand themselves) that some blog told them would protect them from having to actually consider using placeholders for database calls, and other good practices like that. sigh.
</rant>
Sorry - i don't think YOU meant to say that mod_security is a replacement for basic good programming practices, but i just had to vent a bit, and your post reminded me...
Actually, in Germany taking a basic course on "Life Supporting Measure at the Site of an Accident" is a prerequisite for getting a driver's license, and i never quite understood why this is not implemented in more countries around the world. It's less than a First Aid course, but a lot more than nothing. Also, traffic cops often check that your car has the obligatory first aid kit present, and that it is contains all mandatory equipment, and that they haven't expired yet. I have no idea whether all of this actually saves a lot of lives, but it's definitely a good idea, IMHO.
If Earth's solar constant is 1366 W/m2, and this 'color' absorbs 99.9% of the incoming light's energy (which wavelengths? all of them?), wouldn't this mean that it would be almost trivial to boil water in containers covered with this, and thus power steam turbines? Shouldn't this then be basically the solution to all out energy problems, or is there something i am missing? Losses by black-body radiation, if i understood that problem correctly, depend on the material's temperature, but i'd guess that at 100C, this would still be an incredible energy-source. Just a couple of square meters on the roof would easily power a house.
I think the semantic web would be incredible, once it is widely implemented by content providers - a great example is dbpedia's query
"Soccer player with tricot number 11 from club with stadium with >40000 seats born in a country with more than 10M inhabitants"
but, as far as i can see, it's just too tedious to implement. There should be something in between full-fledged semantics, and stuff like RSS which expose information in a rather un-semantic way.
I just ran in to a problem trying to unify various "Event" feeds using RSS from various websites in a central calendar. Very stupid work, since it just feels wrong writing various parsers for such simple information like "date, subject, text". And if people can't even get around to use readily available standards for stuff like this, how will they ever implement the semantic web?
My 'solution' would be a gigantic awareness-campaign on how cool it would be vor everybody involved to use some kind of standard (even something like microformats)....
All i want is a generic LOW COST I/O board, with driver support for everything from a little Palm PDA to a Linux server. just hook it up to a USB port, or wherever, and read from certain 'registers' to know the voltage of a input-pin, or write to a register to open or close a opto-coupled relais, or even better to set a resistance (so that you could dim a light, or make a motor go slower). but it should be completely fool-proof, in the sense that i can hook up 220V or 1.5V to the ports, and nothing 'bad' happens. I think a device like that would open up a gigantic market for little hard-hacks, and my limited electronic knowledge doesn't see why this ould be so hard to design.
sure, some of this functionality is possible with an old parallel port, but most PCs now don't come with a parallel port, and i have yet to find documentation on how to use those USB Parallel adapters for this (any hints?)
I am an atheist, but i would still even accept the theory that a god-thing created the rules at the very beginning, and then stood back and watched to see what happens, so to speak, without interfering in any way - but you insist that god DOES interfere. In my previous post i was trying to show you that this is, as far as i can see, impossible, because if e.g. you'd pray before an exam that you'll get a good grade, and i'd start to monitor every little physical interaction that occurs to you after your prayer, i would have to notice SOMETHING that couldn't be explained with our scientific toolbox - which would at least leave the possibility open that some god or something supernatural influenced your actions (i.e. responded to your prayer). If i cant measure ANYTHING out of the ordinary, how exactly did god then manage to modify your neurological signals in a way that you get a better grade at the end?
The relationship between me and my brother has nothing to do with science,
But 'relationship' is just a fancy word for 'complex interaction (taking into account their current state, and their vectors of change of properties) between a bunch of molecules, part of them in neurons, but to understand the whole deal, you'd have to have the state-information of almost all molecules in the universe', and if you put it that way, the relationship between your molecules and your brother's molecules certainly do fall in the realm of science. Or, to rephrase the GF's gist: if god had anything to do with your relationship, then point me EXACTLY to the point in the infinite chain of action-reactions that make up your relationship, where god fiddled with things, to make them so, which science can't explain. As far as i can see, the only 'escape routes' for theists are that a) god set up the rules of the game at the beginning or b) god somehow controls the really random stuff like nuclear decay. case a) would make god irrelevant, since he wouldn't be able to influence anything ever again (including the afterlife, since your consciousness is subject to scientific investigation of your molecules), and given case b) would make god's action unstudiable, and unpredictable.
Just an ignorant question from somebody who hasn't opened his physics-books in a while:
If absolute zero is when particles stop their brownian motion, what is the reference-frame of that measurement? If 'd measure the motion and energy of an atom in a lab here on earth, and get it to 'stop' relative to my instruments, i'd be measureing absolute zero, right? But couldn;t earth's motion through the universe be interpreted as a 'unilateral brownian motion', which would mean that the atom actually does have at least some energy left in it?
sorry if i'm overlooking something obvious, but this question really intrigues me...
Why does he always use a wiimote for this? A couple of years ago, i build a basic minority-report-like-wave-your-hand-in-front-of-the-screen UI with a webcam, an IR-LED and a infrared filter in front of the webcam (i simply used the dark-red looking plastic that usually covers the transmitting LED in a regular remote control). I then simply pulled the images from the webcam using some v4l-utility, and ran some (extremly basic) pattern-recognition in perl to detect the white 'dot' that represents the IR-LED that was stuck to my hand, and used it to move the cursor around on the screen - really not too difficult.
If someone would write a nice library for this (using intel's obencv library for the pattern-recognition, for example), almost all the 'hacks' from this guy could be implemented without the wiimote, and this would seriously rock.
apart from that - AFAIK, face-detection (just detecting a face and the position of he eyes in an image) usually works well enough, so this particular hack should even be possible without using any infrared LEDs, and this would be really cool if implemented correctly. Any takers? Becasue it really looks cool in the demo.
I don't like it when people see evolution as something that can be measured, or something that only happens genetically, etc. Isn't evolution simply a synonym for "things change, and some things change in a way that sucks for them"? What i mean to say is that yes, we are at a special moment in history, starting to have the power of genetics, and thus the basis of (a part of) biological evolution in reach of having it under our control. But genes are just the means to an end. So is the fashion-du-jour of what is sexually attractive, or what way of thinking is 'better' than others. In a certain way, evolution doesn't even have to be limited to biology (maybe it is by definition, but IMHO shouldn't be), since every physical and chemical reaction also strifes to equilibrium points, in accordance with their environment - just as animals and therefore humans do. Whether this happens because genes change, or because we decide consciously what we want our offspring to be like, is kind of irrelevant, since transhumanism (which i am a big fan of) wouldn't be the end of evolution, but just another form of it. Just as when fishes first started crawling on the land, it just changed some of the rules of the game, but the game is still the same.
Sorry if this is kind of off-topic, but i remember that when IBM sold it's PC department to Lenovo (even though Lenovo was actually producing the machines before that, anyhow), folks were very afraid that the great ThinkPad would fall in quality and robustness and all the things that make ThinkPads great - does anyone have anectodal evidence whether this actually happens, or are ThinkPads still the laptop-of-choice for the people that value quality over the bling-factor?
Hey!
Too bad that I cant moderate your post! I just read it out to my girlfriend, and she and myself completely agree with you (she even wanted to know more about what this slashdot place is, where people leave such insightful comments!).
I, for one, completely agree with you - but it still bugs me a lot that scientific principles seem to be way too under-valued in the practicing medical community - which seems to be caused in part by the factors you described (it's just not useful in everyday cases).
My girlfriend happens to be a doctor, and currently works in a 'obesity clinic', and she is going for a PhD in public health, with a focus on obesity, and she left me with the impression that:
- Real medicine never was a 'real' science. It's absolutely shocking how many publications, treatments and diagnosis are based purely on 'gut feelings', or incoherent theories. Just pull up any statistics on malpractices, and be shocked. No other 'science' could get away with so many errors, after such a long time of experimenting. This happens in part because medicine is a rather unique applied science: there're a lot of psychological factors, and incredible amount of measuring errors, a gigantic level of complexity and tons of historic 'baggage' that doctors have to face every day.
- Medicine is getting a lot better in this aspect - there seems to be a relatively new way of thinking (in the medical community, at least) called "Evidence based medicine", which, if i understood correctly, could be basically summed up as applying scientific principals to the medical processes
- Obesity in specific is extremely complex. Almost everything you do has some influence on you body-weight and composition. Of course the laws of thermodynamics apply to human beings, too, but there are a gazillion factors that influence just how exactly the body deals with excessive calorie intake, or lack thereof, ranging from genetical to psychological and social factors. Just a basic example would be that if you simply stop eating for a week, you usually lose LESS weight compared to if you start 'snacking' all the time, eating 5 little meals a day (basic theory behind this sems to be that the body switches to 'emergency mode' if there's no food around, trying to save as may energy reserves as it can)
- Most theories seem to me to be a wild mixture of anecdotal observations mixed with biochemistry, somehow resembling Freudian theories - they are coherent in them selves, but lack a level of 'scientific interconection' to other knowledges. So it's quite common for a specific theory in obesity to me contradictory to a theory of e.g. neuroscience. As long as both theories "kind of" work, it doesn't seem to be a top priority to resolve that discrepancy (in contrast to what i have observed in 'hard sciences'). AS far as I can tell, thee's no real proof or reason why Whiskey shouldn't be as bad as Vodka in a diet, yet (here, at least) it's common knowledge that whiskey's ok, but vodka will make you fat - and as long as this works, it doesnt matter that much why this happens, or if it happens at all.
I just want to point out, as others have, that K-Meleon is a really nice, full featured Gecko-browser for low-end machines. Its basically Firefox without the XUL layer, and is even more customizable than FF out of the box. It really a life-saver on old machines. I run it on a old Pentium II, 300MHz, 64MB RAM laptop, and it's quite snappy and very usable!
I am very surprised that so many fellow slashdotters seem to believe in dowsing - but i think that's actually a good thing, because they at least try to explain how to do it, thus satisfying the requirement of fallibility of their theory. I'd really love to know how many of them would recant their beliefs when someone would put them to the test.
Well, anyhow: dowsing can be a very profitable art, it seems - Randi's 1,000,000 $ challenge seems to be open to dowsers, as well. So I'd recommend that one of you believers put their money where their mouth is, and take that challenge! But, i guess a lot of people have tried and failed, up to now: The Matter of Dowsing
IIRC, RFID nowadays has failure rates between 10% and 40% - and even though it would be incredibly revolutionary if i could get an exact tally of my inventory by just walking through the aisles with an RFID reader once, a failure rate of even 5% would be way to high - people's jobs (HOW much was stolen in the store you manage?!?), long term supply planning and stuff like that are on the line with this, so people are doing anything to reduce the error rate to the bare minimum, and as long as nothing fundamentally changed since the last time i looked into RFID, it's still nowhere close to being viable. Just imagine that nice "instant checkout by driving you cart through some antennas" scenario - but with a 10% failure rate.
I might be completely mistaken here, but IIRC, if you send e.g. a theoretical laser beam straight up into space, after a couple of billion years, it should return to the point of origin from the other side (even though it was traveling in a straight line, for all that we can tell) because of space-time being bent into itselve ('unendlich in sich geschlossen', was the term in german, it think).
So, how far 'back' do we hae to be able to see into space in order to see the milky way being formed? Or am i completely outdated with that theory?
Yes, but i somehow doubt that the army is promoting mathematical game theory in their ads. Would make a nice campaign, though!:)
Anybody know if there are military strategy books/essays based using game theory?
That last point of yous is exactly my problem with the Chinese Room Argument - for all that it's worth, the Room (assuming it's a closed unit) as such actually does seem to understand chinese - and my point is that the word "seems" in that sentence is irrelevant. A neuron is also simply doing what it was told to do (my laws of nature), so to speak. a neuron isn't intelligent. Neither are Glia cells or what-have-you. Thus, the infrastructure of intelligence doen't matter. Only the result. Of course AI isn't there yet, but we're getting there, using a combination of ugly hacks, smart heuristics and a bit of voodoo - just the way nature got there.
There was a half-joke amongst cognitive scientists every time the endless discussions about the definition of the word intelligence sprang up:
"Intelligence is what the brain does"
The sad part is that i think that that is actually one of the best definitions around.
I used to study AI for a while, and i just wanted to point out how unfair this line of reasoning is. Stuff like this ("Very nice, but it isn't *real* AI, because...") always comes up every time there's some AI break-though being discussed.
1. It's almost trivial to make a program 'learn' from mistakes. Just store some negative value for that specific decision-point. Depends on your definition of 'learning', of course. But the principle is the same in humans and AI
2. Kasparov also adjusted his style (i believe there are certain playing-styles that are beneficial when playing against an AI), and i bet he had coaches and consultants
3. So what?
4. See above.
My point is that every time some AI people actually manage to out-do humans, humans tend to re-define what intelligence is. I bet if you'd tell somebody 100 years ago that a machine would be the world's best chess player, that alone would have been enough to consider the machine 'intelligent of sorts'. But as soon as we know how it works, it somehow looses the right to be called 'intelligent' (mechanical turk). I think this is because it seems to hurt humans that AI shows them that whatever gives us the right to call ourselves 'intelligent' is nothing more than the result of zillions of relatively simple interactions of little protein-machines.
IIRC (its been a while) the best way to determine what language a given text is written in, is amazingly 'stupid': just compare the ratio of how many times the different characters appear. The result is still amazing and should be considered 'kind of intelligent'.
So, just give AI some kudos, accept that there's a lot left to be done, and that the heuristics dint really matter, as long as the result is cool. (and please dont give me none of that Chinese Room Argument crap)
I don't get it. According to the video, this is simply a big remote-controlled car with a couple of cameras. I really couldn't make out anything really advanced or innovative about this. Yet, I cringe at the thought how many millions of dollars went into these things.
I know this was only a PR stunt, but it does show point out the trend that users are starting to get very decent bandwidths nowadays. My only worry is: how can the servers keep up? Most dedicated server i saw still 'only' have 100MBit pipes, and even at 1GBit, it only takes 100 simultaneous 10MBit connections (and only 10 for the 100MBit connections! Kneel before my math skills!) to saturate that link - so, what are servers supposed to do?
I'm really worried that this trend will complicate matters for people like me who manage normal servers, instead of akamai'd server farms....
Slashdot Mirror
real LISPers seem to think ANYTHING looks like LISP, only uglier.....
See? That's something i hate about PHP - i'm a kind of sloppy perl-coder myself, but this whole shielding-bad-code-from-the-evil-internet mentality has to stop. Why would apache, using a special module, have to baby-sit PHP applications? The correct way of doing this is to simply always check the input you're getting, and only letting known good values pass - especially if you're going to use some function that might pull data from the internet. Obviously, adding additional layers of security around applications that are already as safe as possible doesn't cause harm, but i've just seen PHP coders get away with the belief that their "safe mode" or some other patch or hack will protect them from evil, or considerably increase the security of their programs, even though their programs are just deeply flawed in the core. Often, they just run around like headless chicken screaming that the world is going to end if you don't switch on some RewriteRules (they don't understand themselves) that some blog told them would protect them from having to actually consider using placeholders for database calls, and other good practices like that. sigh.
</rant>
Sorry - i don't think YOU meant to say that mod_security is a replacement for basic good programming practices, but i just had to vent a bit, and your post reminded me...
Actually, in Germany taking a basic course on "Life Supporting Measure at the Site of an Accident" is a prerequisite for getting a driver's license, and i never quite understood why this is not implemented in more countries around the world. It's less than a First Aid course, but a lot more than nothing. Also, traffic cops often check that your car has the obligatory first aid kit present, and that it is contains all mandatory equipment, and that they haven't expired yet. I have no idea whether all of this actually saves a lot of lives, but it's definitely a good idea, IMHO.
If Earth's solar constant is 1366 W/m2, and this 'color' absorbs 99.9% of the incoming light's energy (which wavelengths? all of them?), wouldn't this mean that it would be almost trivial to boil water in containers covered with this, and thus power steam turbines? Shouldn't this then be basically the solution to all out energy problems, or is there something i am missing? Losses by black-body radiation, if i understood that problem correctly, depend on the material's temperature, but i'd guess that at 100C, this would still be an incredible energy-source. Just a couple of square meters on the roof would easily power a house.
I think the semantic web would be incredible, once it is widely implemented by content providers - a great example is dbpedia's query
"Soccer player with tricot number 11 from club with stadium with >40000 seats born in a country with more than 10M inhabitants"
but, as far as i can see, it's just too tedious to implement. There should be something in between full-fledged semantics, and stuff like RSS which expose information in a rather un-semantic way.
I just ran in to a problem trying to unify various "Event" feeds using RSS from various websites in a central calendar. Very stupid work, since it just feels wrong writing various parsers for such simple information like "date, subject, text". And if people can't even get around to use readily available standards for stuff like this, how will they ever implement the semantic web?
My 'solution' would be a gigantic awareness-campaign on how cool it would be vor everybody involved to use some kind of standard (even something like microformats)....
All i want is a generic LOW COST I/O board, with driver support for everything from a little Palm PDA to a Linux server. just hook it up to a USB port, or wherever, and read from certain 'registers' to know the voltage of a input-pin, or write to a register to open or close a opto-coupled relais, or even better to set a resistance (so that you could dim a light, or make a motor go slower). but it should be completely fool-proof, in the sense that i can hook up 220V or 1.5V to the ports, and nothing 'bad' happens. I think a device like that would open up a gigantic market for little hard-hacks, and my limited electronic knowledge doesn't see why this ould be so hard to design.
sure, some of this functionality is possible with an old parallel port, but most PCs now don't come with a parallel port, and i have yet to find documentation on how to use those USB Parallel adapters for this (any hints?)
I am an atheist, but i would still even accept the theory that a god-thing created the rules at the very beginning, and then stood back and watched to see what happens, so to speak, without interfering in any way - but you insist that god DOES interfere. In my previous post i was trying to show you that this is, as far as i can see, impossible, because if e.g. you'd pray before an exam that you'll get a good grade, and i'd start to monitor every little physical interaction that occurs to you after your prayer, i would have to notice SOMETHING that couldn't be explained with our scientific toolbox - which would at least leave the possibility open that some god or something supernatural influenced your actions (i.e. responded to your prayer). If i cant measure ANYTHING out of the ordinary, how exactly did god then manage to modify your neurological signals in a way that you get a better grade at the end?
But 'relationship' is just a fancy word for 'complex interaction (taking into account their current state, and their vectors of change of properties) between a bunch of molecules, part of them in neurons, but to understand the whole deal, you'd have to have the state-information of almost all molecules in the universe', and if you put it that way, the relationship between your molecules and your brother's molecules certainly do fall in the realm of science. Or, to rephrase the GF's gist: if god had anything to do with your relationship, then point me EXACTLY to the point in the infinite chain of action-reactions that make up your relationship, where god fiddled with things, to make them so, which science can't explain. As far as i can see, the only 'escape routes' for theists are that a) god set up the rules of the game at the beginning or b) god somehow controls the really random stuff like nuclear decay. case a) would make god irrelevant, since he wouldn't be able to influence anything ever again (including the afterlife, since your consciousness is subject to scientific investigation of your molecules), and given case b) would make god's action unstudiable, and unpredictable.
Just an ignorant question from somebody who hasn't opened his physics-books in a while:
If absolute zero is when particles stop their brownian motion, what is the reference-frame of that measurement? If 'd measure the motion and energy of an atom in a lab here on earth, and get it to 'stop' relative to my instruments, i'd be measureing absolute zero, right? But couldn;t earth's motion through the universe be interpreted as a 'unilateral brownian motion', which would mean that the atom actually does have at least some energy left in it?
sorry if i'm overlooking something obvious, but this question really intrigues me...
Why does he always use a wiimote for this? A couple of years ago, i build a basic minority-report-like-wave-your-hand-in-front-of-the-screen UI with a webcam, an IR-LED and a infrared filter in front of the webcam (i simply used the dark-red looking plastic that usually covers the transmitting LED in a regular remote control). I then simply pulled the images from the webcam using some v4l-utility, and ran some (extremly basic) pattern-recognition in perl to detect the white 'dot' that represents the IR-LED that was stuck to my hand, and used it to move the cursor around on the screen - really not too difficult.
If someone would write a nice library for this (using intel's obencv library for the pattern-recognition, for example), almost all the 'hacks' from this guy could be implemented without the wiimote, and this would seriously rock.
apart from that - AFAIK, face-detection (just detecting a face and the position of he eyes in an image) usually works well enough, so this particular hack should even be possible without using any infrared LEDs, and this would be really cool if implemented correctly. Any takers? Becasue it really looks cool in the demo.
I don't like it when people see evolution as something that can be measured, or something that only happens genetically, etc. Isn't evolution simply a synonym for "things change, and some things change in a way that sucks for them"? What i mean to say is that yes, we are at a special moment in history, starting to have the power of genetics, and thus the basis of (a part of) biological evolution in reach of having it under our control. But genes are just the means to an end. So is the fashion-du-jour of what is sexually attractive, or what way of thinking is 'better' than others. In a certain way, evolution doesn't even have to be limited to biology (maybe it is by definition, but IMHO shouldn't be), since every physical and chemical reaction also strifes to equilibrium points, in accordance with their environment - just as animals and therefore humans do. Whether this happens because genes change, or because we decide consciously what we want our offspring to be like, is kind of irrelevant, since transhumanism (which i am a big fan of) wouldn't be the end of evolution, but just another form of it. Just as when fishes first started crawling on the land, it just changed some of the rules of the game, but the game is still the same.
Sorry if this is kind of off-topic, but i remember that when IBM sold it's PC department to Lenovo (even though Lenovo was actually producing the machines before that, anyhow), folks were very afraid that the great ThinkPad would fall in quality and robustness and all the things that make ThinkPads great - does anyone have anectodal evidence whether this actually happens, or are ThinkPads still the laptop-of-choice for the people that value quality over the bling-factor?
Hey! Too bad that I cant moderate your post! I just read it out to my girlfriend, and she and myself completely agree with you (she even wanted to know more about what this slashdot place is, where people leave such insightful comments!).
:)
I, for one, completely agree with you - but it still bugs me a lot that scientific principles seem to be way too under-valued in the practicing medical community - which seems to be caused in part by the factors you described (it's just not useful in everyday cases).
Well. We just wanted to say 'thanks', I guess
My girlfriend happens to be a doctor, and currently works in a 'obesity clinic', and she is going for a PhD in public health, with a focus on obesity, and she left me with the impression that:
- Real medicine never was a 'real' science. It's absolutely shocking how many publications, treatments and diagnosis are based purely on 'gut feelings', or incoherent theories. Just pull up any statistics on malpractices, and be shocked. No other 'science' could get away with so many errors, after such a long time of experimenting. This happens in part because medicine is a rather unique applied science: there're a lot of psychological factors, and incredible amount of measuring errors, a gigantic level of complexity and tons of historic 'baggage' that doctors have to face every day.
- Medicine is getting a lot better in this aspect - there seems to be a relatively new way of thinking (in the medical community, at least) called "Evidence based medicine", which, if i understood correctly, could be basically summed up as applying scientific principals to the medical processes
- Obesity in specific is extremely complex. Almost everything you do has some influence on you body-weight and composition. Of course the laws of thermodynamics apply to human beings, too, but there are a gazillion factors that influence just how exactly the body deals with excessive calorie intake, or lack thereof, ranging from genetical to psychological and social factors. Just a basic example would be that if you simply stop eating for a week, you usually lose LESS weight compared to if you start 'snacking' all the time, eating 5 little meals a day (basic theory behind this sems to be that the body switches to 'emergency mode' if there's no food around, trying to save as may energy reserves as it can)
- Most theories seem to me to be a wild mixture of anecdotal observations mixed with biochemistry, somehow resembling Freudian theories - they are coherent in them selves, but lack a level of 'scientific interconection' to other knowledges. So it's quite common for a specific theory in obesity to me contradictory to a theory of e.g. neuroscience. As long as both theories "kind of" work, it doesn't seem to be a top priority to resolve that discrepancy (in contrast to what i have observed in 'hard sciences'). AS far as I can tell, thee's no real proof or reason why Whiskey shouldn't be as bad as Vodka in a diet, yet (here, at least) it's common knowledge that whiskey's ok, but vodka will make you fat - and as long as this works, it doesnt matter that much why this happens, or if it happens at all.
I just want to point out, as others have, that K-Meleon is a really nice, full featured Gecko-browser for low-end machines. Its basically Firefox without the XUL layer, and is even more customizable than FF out of the box. It really a life-saver on old machines. I run it on a old Pentium II, 300MHz, 64MB RAM laptop, and it's quite snappy and very usable!
I am very surprised that so many fellow slashdotters seem to believe in dowsing - but i think that's actually a good thing, because they at least try to explain how to do it, thus satisfying the requirement of fallibility of their theory. I'd really love to know how many of them would recant their beliefs when someone would put them to the test.
Well, anyhow: dowsing can be a very profitable art, it seems - Randi's 1,000,000 $ challenge seems to be open to dowsers, as well. So I'd recommend that one of you believers put their money where their mouth is, and take that challenge! But, i guess a lot of people have tried and failed, up to now: The Matter of Dowsing
IIRC, RFID nowadays has failure rates between 10% and 40% - and even though it would be incredibly revolutionary if i could get an exact tally of my inventory by just walking through the aisles with an RFID reader once, a failure rate of even 5% would be way to high - people's jobs (HOW much was stolen in the store you manage?!?), long term supply planning and stuff like that are on the line with this, so people are doing anything to reduce the error rate to the bare minimum, and as long as nothing fundamentally changed since the last time i looked into RFID, it's still nowhere close to being viable. Just imagine that nice "instant checkout by driving you cart through some antennas" scenario - but with a 10% failure rate.
I might be completely mistaken here, but IIRC, if you send e.g. a theoretical laser beam straight up into space, after a couple of billion years, it should return to the point of origin from the other side (even though it was traveling in a straight line, for all that we can tell) because of space-time being bent into itselve ('unendlich in sich geschlossen', was the term in german, it think).
So, how far 'back' do we hae to be able to see into space in order to see the milky way being formed? Or am i completely outdated with that theory?
Yes, but i somehow doubt that the army is promoting mathematical game theory in their ads. Would make a nice campaign, though! :)
Anybody know if there are military strategy books/essays based using game theory?
According to this ad for the army, the army seems to think that war is just another game on another level. sickening.
That last point of yous is exactly my problem with the Chinese Room Argument - for all that it's worth, the Room (assuming it's a closed unit) as such actually does seem to understand chinese - and my point is that the word "seems" in that sentence is irrelevant. A neuron is also simply doing what it was told to do (my laws of nature), so to speak. a neuron isn't intelligent. Neither are Glia cells or what-have-you. Thus, the infrastructure of intelligence doen't matter. Only the result. Of course AI isn't there yet, but we're getting there, using a combination of ugly hacks, smart heuristics and a bit of voodoo - just the way nature got there.
There was a half-joke amongst cognitive scientists every time the endless discussions about the definition of the word intelligence sprang up:
"Intelligence is what the brain does"
The sad part is that i think that that is actually one of the best definitions around.
I used to study AI for a while, and i just wanted to point out how unfair this line of reasoning is. Stuff like this ("Very nice, but it isn't *real* AI, because...") always comes up every time there's some AI break-though being discussed.
1. It's almost trivial to make a program 'learn' from mistakes. Just store some negative value for that specific decision-point. Depends on your definition of 'learning', of course. But the principle is the same in humans and AI
2. Kasparov also adjusted his style (i believe there are certain playing-styles that are beneficial when playing against an AI), and i bet he had coaches and consultants
3. So what?
4. See above.
My point is that every time some AI people actually manage to out-do humans, humans tend to re-define what intelligence is. I bet if you'd tell somebody 100 years ago that a machine would be the world's best chess player, that alone would have been enough to consider the machine 'intelligent of sorts'. But as soon as we know how it works, it somehow looses the right to be called 'intelligent' (mechanical turk). I think this is because it seems to hurt humans that AI shows them that whatever gives us the right to call ourselves 'intelligent' is nothing more than the result of zillions of relatively simple interactions of little protein-machines.
IIRC (its been a while) the best way to determine what language a given text is written in, is amazingly 'stupid': just compare the ratio of how many times the different characters appear. The result is still amazing and should be considered 'kind of intelligent'.
So, just give AI some kudos, accept that there's a lot left to be done, and that the heuristics dint really matter, as long as the result is cool. (and please dont give me none of that Chinese Room Argument crap)
I don't get it. According to the video, this is simply a big remote-controlled car with a couple of cameras. I really couldn't make out anything really advanced or innovative about this. Yet, I cringe at the thought how many millions of dollars went into these things.
I know this was only a PR stunt, but it does show point out the trend that users are starting to get very decent bandwidths nowadays. My only worry is: how can the servers keep up? Most dedicated server i saw still 'only' have 100MBit pipes, and even at 1GBit, it only takes 100 simultaneous 10MBit connections (and only 10 for the 100MBit connections! Kneel before my math skills!) to saturate that link - so, what are servers supposed to do?
I'm really worried that this trend will complicate matters for people like me who manage normal servers, instead of akamai'd server farms....