Come on, if you have never used Bluesecurity, then you were obviously not in their database, and your email could not have been leaked to the spammers! Obviously, the spammers just sent out these FUD spam mails to everyone, just like spammers generally do.
I have to agree with the above poster. We've had ID cards in Sweden for ages, and I've never noticed any significant breach of privacy. What I have noticed is how difficult it is in the UK (where I'm living at the moment) to e.g. open a bank account. Proving your identity here requires obscene arrangements like producing a "utility bill" (gas, electricity etc.) which mean tough luck if your name is not on the contract. On the other hand, once you've got your foot in the system, you can register for a lot of things in the wrong name. The system without ID cards is perverse and I'm frankly amazed that it somehow works after all.
I've been trying to track down the details on this experiment, to find out how the brain culture does its learning. Is it supervised learning? Reinforcement learning? How does the network know when it's doing well? If it's given a reinforcement signal/target values, how does it know how to use this data?
The research referred to in the article doesn't seem to be easily available on the web. What you can find on the guy's (DeMarse's) website are a few older articles, which describe the setup with the culture disk etc., but no mention of getting the brain slice to learn anything. In fact, they seem not to understand anything about what happens between input and output, more than that things change.
So, does anyone know how they do this? My guess is that they have just trained a perceptron or something similar on the output of the brain culture.
As for what you call "ethical cheats", that is what evolutionary algorithms are really, really good at. Trust me. You have design your fitness function (scoring system) very carefully for this not to happen. It is a major source of frustration, disappointment and thoughts of getting a normal job among neuroevolution researchers. E.g., you want evolution to come up with a nice neural network that drives smoothly around a track, but evolution (that bastard!) finds out that it can actually score higher faster by creating something that drives in circles, bounces between walls etc.
I don't know about the other tactics, but it is certainly not impossible, given that NEAT is more open-ended that most NE systems out there. Let's find out!
The only way for IT workers in western countries to survive is to gain additional skills which workers in other countries lack.
Alternatively, we could just move to India. While developers' paychecks over there are certainly lower than here, they are usually worth more, compared to the overall cost of living in the country. I, for one, wouldn't mind moving abroad to do equally stimulating work for relatively more money, and at the same time escape the bitter cold of nothern Europe.
Living every day like it was their last is exactly what infants do before they acquire the concept of "delayed gratification", which they do (according to Piaget) at about the age of three.
Given that the trend of the quality of my life is positive, I suspect that my life was pretty hopeless before I was three. And by induction from one case, I suspect that this goes for anyone else also.
So, whatever you do, don't live your life like every day was the last.
I do know Scheme - I used to TA a course on it, which is admittedly not the same as using it for something productive - and there are some really nifty features in it that I wish I had available in non-functional languages sometimes, such as the ability to pass a method (rather than an object) as an argument to another method.
But just to underscore how much certain things are matters of taste, I find dynamic typing horrible. In my (strange?) world, strong typing = stricter checking = more control = easier debugging = better. Such a "feature" is for me an argument against any language that has it.
But yes, I will check out Ruby, for fear of being lost behind as the earth keeps revolving...
If you use Struts, Spring, etc., you configure every action in mulitple XML files. You map everything around so that the application knows what to do. This means you must define knowledge about your application's architecture in multiple places.
With Tomcat, I just drop JSP files in one directory and classes in another. If MVC thinking is required, the model can reside in the classes, and view in the JSPs. Not very complicated. But I've probably missed something important here...
I suppose I still have to try out RoR seriously, then... And I can buy the syntactic sugar argument - while which code looks better is to a large extent a matter of personal taste, Java code certainly has many characters per method call, aka long lines. (Is there an established measure for this?)
One thing I've never really understood, though, is why people generally consider Java development slow. To me, getting a basic web app going in JSP takes no longer than doing the same in PHP. Probably less, if there is debugging involved. It might be that it could go even faster with Rails.
Another factor is the ease of integration with other apps. I don't spend most of my days developing for the web, rather I do bio-inspired AI, but when I do do web I often want to integrate it with some other fanciness I've written, often in Java. Doing both web and other applications written in the same language makes integration infinitely much easier. But I guess I'm a minority here, as most web apps only ever communicate with a database.
As I understand it, Ruby/Rails has got a lot of attention lately as it seems to be an easy to learn yet powerful way of creating web applications, for people that normally use php. However, while php is great for quick hacks, it really caters most to those who have never had any proper programming education, and is not very well suited to anything more advanced than a shopping cart. (No, this is not flamebait, it's fact. Sorry for offending some of you, I don't know how to put it otherwise.)
So, my question is: is there anything in all that RoR buzz for those who actually have CS degrees and usually do web development with JSP/servlets? Is RoR in any way better than Java, or just easier?
As for a solid database, why don't they just serve up a nice gui and some integration with the other office-style apps for mysql or something similar? Mysql is already available as a very simple installation for OS X, so I way there is already a solid database, just not the 'app', or graphical fancy.
And given that you have months and years, that makes defaulting to the 'instant, cheap, and easy' strategy even more reprehensible.
Ok, I will try to disregard the parts of your statement that was just a unsolicited and pure insult. Hmm... suddenly there was nothing left to answer to.
Utter bilge. Papers that are correct and/or important, and relevant should get the most citations.
Ah, I see. Principles. But welcome to the real world, where success is partly dependent on how well you can convince others that you work is correct, important and relevant. The real world, where there are literally thousands of papers that just could be relevant, and there is no way that you could physically get hold of all of them. Even if you have access to the Bodleian library, which I incidentally have not. I am sure I can get hold of more relevant research on the web than I could do if I spent the same time running around the library and/or filling out distance-loan applications.
I don't understand what you have against freely available research. Do you get a paycheck from Springer-verlag?
In other words, cheap and easy is more important than complete and correct.
I would say that:
1. Even a proper literature survey has time limits. You know you're supposed to finish a PhD in a few years?
2. Cheap and easy doesn't need to be, and really shouldn't be, opposed to complete and correct. That papers available online gets more citations is only fair, given that the people who make their papers freely available help make life easier for all of us.
Uhm, no. Let's face it: there is too much research out there that _could_ be relevant. There isn't enough time to order all those articles from faraway libraries, or money to buy them online. To be able to do a proper literature survey you need to be able to get the paper instantly and read just as much as you need of it.
Therefore, I rarely cite papers that are not available for free online. The exceptions tend to be important papers that were written before the age of personal homepages. Most researchers, at least in the digital sciences, put their papers up at their homepages nowadays, even if they are also published in conventional journals of conference proceedings, as there really is no reason not to.
I agree. But the notion that there are ability-affecting genetic differences between individuals is pretty much universally accepted in the scientific (as in natural science, as opposed to social science) community.
What is very far from accepted at the moment is the idea that there may be similar differences between groups (races, genders, etc). When the subject is brought up, scientists often point to studies that show that humans all over the world due to intense traveling and cross-breeding are very genetically similar.
On the other hand, there are studies that show that we share 95% (or something like that) of our genome with chimpanzees. Apparently, what is important is not how many alleles we share with group X, but which particular ones we don't share. In the study described in TFA they seem to only be talking about a few specific genes.
X11 ports look horrible, and their UI are not as responsive as a Cocoa one. A shame, as everybody would want a well-integrated Mac Open Office.
For Java apps on the other hand, there is actually an Aqua look and feel, which is just brilliant. Load up the top-notch IDE IntelliJ, for example, on a Mac and change that one look and feel setting (under Apperance), and you get something which is impossible to tell from a native app.
Looking at how well JetBrains (who makes IntelliJ) pulled this off, it makes you wonder why more productivity apps don't use Java. Probably some old prejudice from developers.
Well, a year or so ago I participated in designing and implementing a system that handles instant messaging and session handling for Swedens largest web community. It has well over a million users, 50 000 of which are online simultaneously - and that's fifty thousand persistent tcp connections, not fifty thousand http requests now and then.
All in Java. Works flawlessly. Even on top of Windows servers.
It's true that Java was once not very well suited for heavy duty network tasks, but since the advent of nio (New I/O) in Java 1.4 that is very much a thing of the past.
"Kids growing up in areas damaged by radiation from the plant have a higher IQ and faster reaction times, say Russian doctors."
Note: say Russian doctors, says The Sun!
Do the two layers on untrustworthyness somehow cancel each other out, so as to make the statement trustworthy?
For example, there is this common saying that "opposites attract", meaning that people are sexually/romantically attracted to people that are the opposites of themselves in some respects (e.g. length, tidyness, aggressiveness or whatnot). On the other hand, you also hear that "same seeks same" - meaning more or less the opposite of the first saying.
Both of these sayings are perfectly acceptable common sense. For people who believe in common sense. In scientific reality, the question of who attracts who is of course infinitely more complex than that, and we need many more "obvious studies" and "academic disputes" before we come to something even resembling a conclusion.
no matter how little I had to pay him. Especially if he was whining about there "being" no jobs and wanting someone to "give" a job to him.
Come on, you're supposed to be americans, the most industrious and innovative people in the world, why are you all crying foul and wanting daddy Bush to protect you when you get a little bit of competition? Cowards.
Importing skills is how the US became a world leader in technology in the first place, what with all Jewish and other European scientists around the time of WWII. There's nothing wrong with that - it's good both for the US and for talented people around the world who couldn't develop to their full potential in their native countries.
So, I think this recent US protectionism is very sad. For us, and for you. Noone gains from US tech companies not being able to hire the most qualified people. Not even underskilled US engineers (there are underskilled engineers in every country).
Why? Because with an underdeveloped or receding US tech industry it will be even harder to find a tech job, and those you find will require higher skills. With an expanding tech industry, on the other hand, it will be easier to find a job for anyone, regardless of how many foreign workers are employed, as competitive companies can continue to grow and take market share from foreign companies and/or expand the world market with it's superior products. And to become competitive, companies need to hire the most qualified personnel possible. It's that simple.
Slashdot Mirror
Come on, if you have never used Bluesecurity, then you were obviously not in their database, and your email could not have been leaked to the spammers! Obviously, the spammers just sent out these FUD spam mails to everyone, just like spammers generally do.
People always blame society.
The question is, what does it say about you?
Brilliant! Too bad it was modded OT...
I mean, it sure is OT, but it's actually really good! So, who's recording this?
No, nuclear power is hot.
I have to agree with the above poster. We've had ID cards in Sweden for ages, and I've never noticed any significant breach of privacy. What I have noticed is how difficult it is in the UK (where I'm living at the moment) to e.g. open a bank account. Proving your identity here requires obscene arrangements like producing a "utility bill" (gas, electricity etc.) which mean tough luck if your name is not on the contract. On the other hand, once you've got your foot in the system, you can register for a lot of things in the wrong name. The system without ID cards is perverse and I'm frankly amazed that it somehow works after all.
I've been trying to track down the details on this experiment, to find out how the brain culture does its learning. Is it supervised learning? Reinforcement learning? How does the network know when it's doing well? If it's given a reinforcement signal/target values, how does it know how to use this data?
The research referred to in the article doesn't seem to be easily available on the web. What you can find on the guy's (DeMarse's) website are a few older articles, which describe the setup with the culture disk etc., but no mention of getting the brain slice to learn anything. In fact, they seem not to understand anything about what happens between input and output, more than that things change.
So, does anyone know how they do this? My guess is that they have just trained a perceptron or something similar on the output of the brain culture.
As for what you call "ethical cheats", that is what evolutionary algorithms are really, really good at. Trust me. You have design your fitness function (scoring system) very carefully for this not to happen. It is a major source of frustration, disappointment and thoughts of getting a normal job among neuroevolution researchers. E.g., you want evolution to come up with a nice neural network that drives smoothly around a track, but evolution (that bastard!) finds out that it can actually score higher faster by creating something that drives in circles, bounces between walls etc.
I don't know about the other tactics, but it is certainly not impossible, given that NEAT is more open-ended that most NE systems out there. Let's find out!
The only way for IT workers in western countries to survive is to gain additional skills which workers in other countries lack.
Alternatively, we could just move to India. While developers' paychecks over there are certainly lower than here, they are usually worth more, compared to the overall cost of living in the country. I, for one, wouldn't mind moving abroad to do equally stimulating work for relatively more money, and at the same time escape the bitter cold of nothern Europe.
Why was parent modded flamebait? It's a completely legitimate opinion, even if the moderator does not agree, expressed without expletives.
Living every day like it was their last is exactly what infants do before they acquire the concept of "delayed gratification", which they do (according to Piaget) at about the age of three.
Given that the trend of the quality of my life is positive, I suspect that my life was pretty hopeless before I was three. And by induction from one case, I suspect that this goes for anyone else also.
So, whatever you do, don't live your life like every day was the last.
I do know Scheme - I used to TA a course on it, which is admittedly not the same as using it for something productive - and there are some really nifty features in it that I wish I had available in non-functional languages sometimes, such as the ability to pass a method (rather than an object) as an argument to another method.
But just to underscore how much certain things are matters of taste, I find dynamic typing horrible. In my (strange?) world, strong typing = stricter checking = more control = easier debugging = better. Such a "feature" is for me an argument against any language that has it.
But yes, I will check out Ruby, for fear of being lost behind as the earth keeps revolving...
If you use Struts, Spring, etc., you configure every action in mulitple XML files. You map everything around so that the application knows what to do. This means you must define knowledge about your application's architecture in multiple places.
With Tomcat, I just drop JSP files in one directory and classes in another. If MVC thinking is required, the model can reside in the classes, and view in the JSPs. Not very complicated. But I've probably missed something important here...
I suppose I still have to try out RoR seriously, then... And I can buy the syntactic sugar argument - while which code looks better is to a large extent a matter of personal taste, Java code certainly has many characters per method call, aka long lines. (Is there an established measure for this?)
One thing I've never really understood, though, is why people generally consider Java development slow. To me, getting a basic web app going in JSP takes no longer than doing the same in PHP. Probably less, if there is debugging involved. It might be that it could go even faster with Rails.
Another factor is the ease of integration with other apps. I don't spend most of my days developing for the web, rather I do bio-inspired AI, but when I do do web I often want to integrate it with some other fanciness I've written, often in Java. Doing both web and other applications written in the same language makes integration infinitely much easier. But I guess I'm a minority here, as most web apps only ever communicate with a database.
As I understand it, Ruby/Rails has got a lot of attention lately as it seems to be an easy to learn yet powerful way of creating web applications, for people that normally use php. However, while php is great for quick hacks, it really caters most to those who have never had any proper programming education, and is not very well suited to anything more advanced than a shopping cart. (No, this is not flamebait, it's fact. Sorry for offending some of you, I don't know how to put it otherwise.)
So, my question is: is there anything in all that RoR buzz for those who actually have CS degrees and usually do web development with JSP/servlets? Is RoR in any way better than Java, or just easier?
As for a solid database, why don't they just serve up a nice gui and some integration with the other office-style apps for mysql or something similar? Mysql is already available as a very simple installation for OS X, so I way there is already a solid database, just not the 'app', or graphical fancy.
And given that you have months and years, that makes defaulting to the 'instant, cheap, and easy' strategy even more reprehensible.
Ok, I will try to disregard the parts of your statement that was just a unsolicited and pure insult. Hmm... suddenly there was nothing left to answer to.
Utter bilge. Papers that are correct and/or important, and relevant should get the most citations.
Ah, I see. Principles. But welcome to the real world, where success is partly dependent on how well you can convince others that you work is correct, important and relevant. The real world, where there are literally thousands of papers that just could be relevant, and there is no way that you could physically get hold of all of them. Even if you have access to the Bodleian library, which I incidentally have not. I am sure I can get hold of more relevant research on the web than I could do if I spent the same time running around the library and/or filling out distance-loan applications.
I don't understand what you have against freely available research. Do you get a paycheck from Springer-verlag?
In other words, cheap and easy is more important than complete and correct.
I would say that:
1. Even a proper literature survey has time limits. You know you're supposed to finish a PhD in a few years?
2. Cheap and easy doesn't need to be, and really shouldn't be, opposed to complete and correct. That papers available online gets more citations is only fair, given that the people who make their papers freely available help make life easier for all of us.
Uhm, no. Let's face it: there is too much research out there that _could_ be relevant. There isn't enough time to order all those articles from faraway libraries, or money to buy them online. To be able to do a proper literature survey you need to be able to get the paper instantly and read just as much as you need of it.
Therefore, I rarely cite papers that are not available for free online. The exceptions tend to be important papers that were written before the age of personal homepages. Most researchers, at least in the digital sciences, put their papers up at their homepages nowadays, even if they are also published in conventional journals of conference proceedings, as there really is no reason not to.
I agree. But the notion that there are ability-affecting genetic differences between individuals is pretty much universally accepted in the scientific (as in natural science, as opposed to social science) community.
What is very far from accepted at the moment is the idea that there may be similar differences between groups (races, genders, etc). When the subject is brought up, scientists often point to studies that show that humans all over the world due to intense traveling and cross-breeding are very genetically similar.
On the other hand, there are studies that show that we share 95% (or something like that) of our genome with chimpanzees. Apparently, what is important is not how many alleles we share with group X, but which particular ones we don't share. In the study described in TFA they seem to only be talking about a few specific genes.
X11 ports look horrible, and their UI are not as responsive as a Cocoa one. A shame, as everybody would want a well-integrated Mac Open Office.
For Java apps on the other hand, there is actually an Aqua look and feel, which is just brilliant. Load up the top-notch IDE IntelliJ, for example, on a Mac and change that one look and feel setting (under Apperance), and you get something which is impossible to tell from a native app.
Looking at how well JetBrains (who makes IntelliJ) pulled this off, it makes you wonder why more productivity apps don't use Java. Probably some old prejudice from developers.
Well, a year or so ago I participated in designing and implementing a system that handles instant messaging and session handling for Swedens largest web community. It has well over a million users, 50 000 of which are online simultaneously - and that's fifty thousand persistent tcp connections, not fifty thousand http requests now and then.
All in Java. Works flawlessly. Even on top of Windows servers.
It's true that Java was once not very well suited for heavy duty network tasks, but since the advent of nio (New I/O) in Java 1.4 that is very much a thing of the past.
"Kids growing up in areas damaged by radiation from the plant have a higher IQ and faster reaction times, say Russian doctors." Note: say Russian doctors, says The Sun! Do the two layers on untrustworthyness somehow cancel each other out, so as to make the statement trustworthy?
For example, there is this common saying that "opposites attract", meaning that people are sexually/romantically attracted to people that are the opposites of themselves in some respects (e.g. length, tidyness, aggressiveness or whatnot). On the other hand, you also hear that "same seeks same" - meaning more or less the opposite of the first saying.
Both of these sayings are perfectly acceptable common sense. For people who believe in common sense. In scientific reality, the question of who attracts who is of course infinitely more complex than that, and we need many more "obvious studies" and "academic disputes" before we come to something even resembling a conclusion.
I'm not sure I would hire someone with the motto
while () { drink_guinness(); }
no matter how little I had to pay him. Especially if he was whining about there "being" no jobs and wanting someone to "give" a job to him.
Come on, you're supposed to be americans, the most industrious and innovative people in the world, why are you all crying foul and wanting daddy Bush to protect you when you get a little bit of competition?
Cowards.
Importing skills is how the US became a world leader in technology in the first place, what with all Jewish and other European scientists around the time of WWII. There's nothing wrong with that - it's good both for the US and for talented people around the world who couldn't develop to their full potential in their native countries.
So, I think this recent US protectionism is very sad. For us, and for you. Noone gains from US tech companies not being able to hire the most qualified people. Not even underskilled US engineers (there are underskilled engineers in every country).
Why? Because with an underdeveloped or receding US tech industry it will be even harder to find a tech job, and those you find will require higher skills. With an expanding tech industry, on the other hand, it will be easier to find a job for anyone, regardless of how many foreign workers are employed, as competitive companies can continue to grow and take market share from foreign companies and/or expand the world market with it's superior products. And to become competitive, companies need to hire the most qualified personnel possible. It's that simple.