This why the technological singularity is preferable and unavoidable if you or I wish to continue to exist.
Ah. Techno-messiahism, where AI or IA ("intelligence amplification") will rapture us all up into cyberspace or augmented reality heaven.
Now there's a solution to global climate threats.
Look, Vinge's "singularity" is an interesting idea; if we create more-intelligent-than-human artifical or amplified intelligence, then that intelligence creates a greater one, and so on, then a rapid cascade would happen. But it's not clear that we can create such intelligence at all, or what the limits of the cascade would be. The idea that AI or IA will save us from the problems we face is literally hoping for a deus ex machina.
Such a singularity is not unprecedented in human history - the development of language, and then of writing, allowed for shared intelligence greater than that of an individual brain, and created a feedback that allowed the individual brain to learn and grow more. That's great. But it didn't repeal the laws of physics, or make us immortal in any but a metaphorical sense. (Which is still pretty good. I have ambition that someone will read a poem or play a song of mine years after I die.)
Heck, it could be argued that we're living in the singularity right now. Is not an internet feed a basic form of intelligence amplification?
Specifically, using object orientation and functional blocks to make everything simpler. i.e. not having a 40,000 line app in one file.
Why in the world would you do that in C?
C's file scoping allows for a great deal of encapsulation. Create your app as a set of modules that interact via a well-defined interface; each module is implemented by one.h file containing the function and extern data declarations and one containing the function definitions (and of course one main.c or app.c that has main()).
It's not OO since you don't get polymorphism, but it gets you the crunchy goodness of encapsulation and abstraction.
You have to trust in the institution, and fix the broken individual people that pollute it.
The institution by its nature is broken. The very nature of a full-time professional police force with special authority is problematic enough, especially with increasing militarization of policing in the past few decades; so long as on top of that they're tasked with enforcing unethical drug, vice, and other "consensual crimes" laws, police forces are broken by design.
but torture most certainly does provide reliable information in certain situations. If you are looking for specific verifiable information (where is the bomb planted) and the person knows, torture will get that information from them.
If the bomb is ticking, the bomber will just give you false information. Yes, it's verifiable information, so you know he lied when it blows up across town as your bomb squad converges on the false leads he gave you, but so what?
Now, you can quibble that we're detaining, tapping, and beating the wrong guys, or not enough guys, and that's fine, we're an open society, get angry and discuss away, but I find it tough to argue against any of these procedures in toto.
Under due process of law of a reasonable government, detention and eavesdropping are fine. We don't have due process of law or a reasonable government at the moment, but yes, that's not an arguement against detention and spying in toto.
Torture, on the other hand, is not only illegal, immoral, a greate recruiting tool for the enemy, and , but it doesn't work as a reliable source of information. People will say anything to make it stop, tell you what they think you want to hear.
If you stick with the rule "simpler, shorter is better" you can almost never go wrong. Once you find out where you can go wrong, you may not need that rule any more.
I would say that even junior developers should be thinking in terms of abstraction and encapsultion - "what is this module or routine's purpose in life, in fifteen words or less".
In fact, maybe we can agree that that's the thing that needs to be short; if you need paragraphs to explain what the thing does, you haven't abstracted the problem very well. And to really determine that, you need to try to write down what it does - you need to comment.
Of course describing parameters, return values, side effects, precondtions, postconditions, design contraints, et cetera, can take pages, but if I can't give a quick snappy answer as to why this thing exists, I probably need to rethink my design. That's something that should be taught to beginning students.
Excellent first step. But are you finished? I'm sure you could find a way to reduce that complexity even further if you spent more time one in.
Of course there was more. I think I ended up shrinking the entire code base on that project by at least ten percent. It was pretty "object-obfuscated" when I got ahold of it, I cleaned up the class hierarchy, factored out some common code, et cetera.
But the issue is the statement that a bunch of short routines are more easily understood than one long one. If you have to jump around six different subroutines to see how some functionality is implemented, that's not necessarily easier to understand than one larger routine.
A routine should be as long as it needs to be to accomplish its function; sometimes that means 1,000 lines nicely encapsulated in one "box" of a routine, rather than a bunch of little boxes tied together.
If you don't know how to refactor, and how to reduce your bloated thousand-line long methods into a series of simple to understand 10-line long methods, you still have much to learn about good code.
If you think that the sole, or even primary, measure of how simple to understand a module is, is the length of the functions or methods, then you still have much to learn about good code.
(For example, I've sometimes found myself refactoring code from a bunch of sort, tightly-coupled routines into one single one to create better encapsultion.)
that comments are not necessary for clarity and can be dangerous if not kept up to date
If you can't keep comments up to date in the code you're responsbile for, you're not competent to be responsbile for the code.
There shouldn't be any debate on the need for documentation. Document your code or hit the road. The only issue is where it goes, in separate docs or in comment blocks. (Doxygen and similar systems make it easy to generate separate docs from comment blocks. Recommended.)
Code reviews are the best enforcement - if you go in and everyone's asking "what the hell does this block do???", you need to comment it.
Bicycle gets it 100 percent due to long lasting lithium grease on the chains and only fabrication using oil
Actually you should proably take into account all the petrochemicals used to grow the food that powered the bike-peddler - fuel for the tractors, feedstocks for fertilizers and pesticides (unless the dude was eating organic). Then there's the energy/oil costs of transporting and refrigerating the food.
Also, this can be set up via education and culture, which teaches you that blacks are stupid, or that you shouldn't read because that's acting white, or that gays are depraved or whatever the popular opinion is.
Education and culture are not neurological pathologies. That's the point.
To say that ideas and opinions can be states of disease is a very dangerous conclusion. If homosexuality can be a "disease" (as it was considered just a few decades ago) and racism can be a "disease", then any beliefs or ideas or emotional states unpopular enough with the official psychiatric community can be diseases. It's defintion by social contract rather than by objective analysis.
Failing to love the State is a disease, Comrade, but we can cure you with re-education.
The brain is not a hardrive that can be erased and reprogrammed, it's a solid object that is relatively static in it's construction.
Of course it's not a computer drive. But if the brain were static than our ideas and behaviours would never change. We'd be unable to learn, and there's be no way to change the troublesome ideas and behaviours we're talking about.
In fact our brain changes every day, thank goodness.
I don't expect to be able to test that after he gets his next few martial arts belts:)
You could sign up and take class with him. It'll help with that extra 40 pounds, and martial arts training is one of the few athletic activities that parents and kids can do together. We have several families where one of the parents trains along with one (or more!) of the kids, and they seem to get a lot out of it.
Say that Evil, Inc runs a banner server banners.evil.com, which puts ads on kinky.xxx and on yourchurch.org (or maybe just an invisible "web bug" on either site). When you visit kinky.xxx, your browser requests the banner from banners.evil.com, which sets a cookie saying "I went to kinky.xxx and all I got was this lousy cookie". That cookie will be sent along with any request your browser makes to banners.evil.com.
Then you log in to yourchurch.org. Their home page has an image tag with a source like "http://banners.evil.com/spyonme.php?username=your name". Your browser makes this request to banners.evil.com, sending along the cookie that server set eariler. Your browser thus tells Evil, Inc your yourchurch.org username (in the image URL) and the fact the you visited kinky.xxx (in the cookie it).
Evil, Inc phones up your pastor and lets him know so that he can shame you in front of the parish the next Sunday (turns out this is all part of your church's anti-porn crusade).
If they combined having an ethical company that is well known with a good product, I think McAfee would bite the dust.
The unethical accounting in question here took place in 1998-2000, and McAfee has cleaned house since then. This settlement is part of that process.
Disclosure: I'm a McAfee stockholder. So I've been watching this play out for year (and I was one of the many stockholders who watched their portfolio's value vanish in the haze of fudged accounting).
what interests me is what norton/symantec is going to do, now that (one of) their biggest competitors is in such a position.
Contrary to the tone of the summary, this isn't a new development that's going to change the anti-virus market. This is McAffe (hopefully finally) settling a problem the occured back in the 20th century - the accounting irregularities in question happened back in 1998-2000.
Disclosure: I'm a McAffe stockholder, so I've been watching this unfold for several years.
This settlement pertains to actions taken in 1998 to 2000. The summary makes it sound like McAffe just got caught with their hand in the cookie jar, when in fact this is the company trying to clean up after an administration long since gone.
The stock went up after the announcement, so the markets seem to think the settlement a good idea.
(Disclosure: I'm a McAffe stockholder, due to stock options from an old employer and a series of mergers (TISX -> NETA -> MFE).
If they're corrupt enough to fuck their shareholders like that, I wonder what other lengths they're willing to sink to.
Hopefully they're not that corrupt anymore. The settlement pertains to actions taken in 1998 to 2000.
The summary makes it sound like McAffe just got caught with their hand in the cookie jar, when in fact this is the company trying to clean up after an administration long since gone.
(Disclosure: I'm a McAffe stockholder, due to stock options from an old employer and a series of mergers.)
According to the Oxford English Dictionary, disease is "a condition of the body, or of some part or organ of the body, in which its functions are disturbed or deranged; a morbid physical condition." Diagnosis, in turn, is "the determination of the nature of a diseased condition... also, the opinion (formally stated) resulting from such investigation."
The core medical concept of disease is a bodily abnormality. Literally, the term "disease" denotes a demonstrable lesion of cells, tissues, or organs. Metaphorically, it may be used to denote any kind of malfunctioning of individuals, groups, economies, etc. (substance abuse, violence, unemployment, et al.).
The psychiatric concept of disease rests on a radical alteration of the medical definition. The mind is not a material object; hence, it can be diseased only in a metaphorical sense. In his classic, Lectures on Clinical Psychiatry, Emil Kraepelin--the founder of modern psychiatry--wrote: "The subject of the following course of lectures will be the Science of Psychiatry, which, as its name implies, is that of the treatment of mental disease. It is true that, in the strictest terms, we cannot speak of the mind as becoming diseased."
...
Diseases are physico-chemical phenomena or processes--for example, the abnormal metabolism of glucose (diabetes). Mental diseases are patterns of personal conduct, unwanted by the self or others. Psychopathology is diagnosed by finding behavioral, not physical, abnormalities in bodies. Disease qua psychopathology cannot be asymptomatic. Changing the official classification of mental diseases can transform non-disease into psychopathology and psychopathology into nondisease (i.e., smoking from a behavioral habit into "nicotine dependence"). In short, medical diseases are discovered and then given a name, such as acquired immune deficiency syndrome (AIDS). Mental diseases are invented and then given a name, such as attention deficit disorder.
(BTW, that's not to say I agree with Szasz on everything.)
Once problems such as depression, schizophrenia, etc. are not considered illnesses, they will not be covered by health care.
Why not? If I break my leg, that's not a illness, but it's covered by my health insurace. No reason why "problems with living" or "psychosocial dysfunctions" couldn't be covered.
I am dismayed by your claims about "neurological expression". Where do you suppose these people's problems reside?
A problem may reside directly at the level of the nervous tissue; alcoholism, for example, is a change in the (IIRC) ion channels along the axon. You could, in principle at least, look at the physiology and see the pathology. (Though the pathology is rather evident when someone gets severely ill or dies from the DTs.)
Other problems do not have such a pathology. Of course beliefs, behaviors, and ideas have some neurophysiological correlation, but not so directly. For example, there's no pathology of the nervous system in holding racist beliefs, which some argue should be considered mental illness. Problems like this are not diseases in need of a treatment, but dysfunctional relationships of the person with themself, other people, and the universe at large in need of adjustment. (Non-coersively, of course, unless the person is a credible threat to the rights of others.)
It's a trite analogy, but think of a "software problem" versus a "hardware problem". The software problem has an certainly has expression within the hardware, but it's not because the hardware is broken. Of course the brain is not a stored-program computer with a clean delination of hardware and software.
To accept homeopathy is to accept the 'law of infinitesimals'...If medication does not follow that, it is not homeopathy.
That's exactly what I mean by setting up straw-men. You're choosing your own defintion of homeopathy that is at odds with what is actually practiced, and then proceeding to rip it to bits.
If "Europe's oldest and the UK's largest manufacturer of homeopathic medicines" is making concoctions with significant concentrations of active ingredients,
that's awfully strong evidence that as homeopathy is actually practiced, not all homoepathic remedies are prepared according to the "law of infinitesimals". If you want to argue, I guess you should take it up with the homeopaths who aren't following your defintion.:-)
And it has nothing to do with how you stir it, or how much water you put in it.
According to the theory (which, again, I'm not saying I believe), yes, it does. Adherents claim that the water takes on some sort of "memory" or "image" of the substance; it "clusters" or absorbs some sort of "electromagnetic vibration" at each stage of dilution.
Slashdot Mirror
Ah. Techno-messiahism, where AI or IA ("intelligence amplification") will rapture us all up into cyberspace or augmented reality heaven.
Now there's a solution to global climate threats.
Look, Vinge's "singularity" is an interesting idea; if we create more-intelligent-than-human artifical or amplified intelligence, then that intelligence creates a greater one, and so on, then a rapid cascade would happen. But it's not clear that we can create such intelligence at all, or what the limits of the cascade would be. The idea that AI or IA will save us from the problems we face is literally hoping for a deus ex machina.
Such a singularity is not unprecedented in human history - the development of language, and then of writing, allowed for shared intelligence greater than that of an individual brain, and created a feedback that allowed the individual brain to learn and grow more. That's great. But it didn't repeal the laws of physics, or make us immortal in any but a metaphorical sense. (Which is still pretty good. I have ambition that someone will read a poem or play a song of mine years after I die.)
Heck, it could be argued that we're living in the singularity right now. Is not an internet feed a basic form of intelligence amplification?
Would that I had mod points, brother. (Er, sister? Can't tell from here.) Amen!
"Morning people are respected. Night people are feared." - calligraphic button I have
"You can look at the moon / Not the sun, it's too bright / Take it as evidence / We were made for the night" - me
Why in the world would you do that in C?
C's file scoping allows for a great deal of encapsulation. Create your app as a set of modules that interact via a well-defined interface; each module is implemented by one .h file containing the function and extern data declarations and one containing the function definitions (and of course one main.c or app.c that has main()).
It's not OO since you don't get polymorphism, but it gets you the crunchy goodness of encapsulation and abstraction.
The institution by its nature is broken. The very nature of a full-time professional police force with special authority is problematic enough, especially with increasing militarization of policing in the past few decades; so long as on top of that they're tasked with enforcing unethical drug, vice, and other "consensual crimes" laws, police forces are broken by design.
Military interrogators say otherwise.
If the bomb is ticking, the bomber will just give you false information. Yes, it's verifiable information, so you know he lied when it blows up across town as your bomb squad converges on the false leads he gave you, but so what?
Nonsense. This is a democratic republic - in a fundamental sense the general public is the government.
Current military deployments and capabilities should be able to be classified for a limited time, say ten years, but that's it.
Two years no, but ten years sounds about right.
The "black budget" gets very little oversight.
You don't "have" to trust them, and you shouldn't. Read up on the Rampart scandal, or the current epidemic of false arrests in Baltimore, or just skim Google for police scandals.
Under due process of law of a reasonable government, detention and eavesdropping are fine. We don't have due process of law or a reasonable government at the moment, but yes, that's not an arguement against detention and spying in toto.
Torture, on the other hand, is not only illegal, immoral, a greate recruiting tool for the enemy, and , but it doesn't work as a reliable source of information. People will say anything to make it stop, tell you what they think you want to hear.
I would say that even junior developers should be thinking in terms of abstraction and encapsultion - "what is this module or routine's purpose in life, in fifteen words or less".
In fact, maybe we can agree that that's the thing that needs to be short; if you need paragraphs to explain what the thing does, you haven't abstracted the problem very well. And to really determine that, you need to try to write down what it does - you need to comment.
Of course describing parameters, return values, side effects, precondtions, postconditions, design contraints, et cetera, can take pages, but if I can't give a quick snappy answer as to why this thing exists, I probably need to rethink my design. That's something that should be taught to beginning students.
Hmm. What field do you work in where you always write trivial code?
Of course there was more. I think I ended up shrinking the entire code base on that project by at least ten percent. It was pretty "object-obfuscated" when I got ahold of it, I cleaned up the class hierarchy, factored out some common code, et cetera.
But the issue is the statement that a bunch of short routines are more easily understood than one long one. If you have to jump around six different subroutines to see how some functionality is implemented, that's not necessarily easier to understand than one larger routine.
A routine should be as long as it needs to be to accomplish its function; sometimes that means 1,000 lines nicely encapsulated in one "box" of a routine, rather than a bunch of little boxes tied together.
If you think that the sole, or even primary, measure of how simple to understand a module is, is the length of the functions or methods, then you still have much to learn about good code.
(For example, I've sometimes found myself refactoring code from a bunch of sort, tightly-coupled routines into one single one to create better encapsultion.)
If you can't keep comments up to date in the code you're responsbile for, you're not competent to be responsbile for the code.
There shouldn't be any debate on the need for documentation. Document your code or hit the road. The only issue is where it goes, in separate docs or in comment blocks. (Doxygen and similar systems make it easy to generate separate docs from comment blocks. Recommended.)
Code reviews are the best enforcement - if you go in and everyone's asking "what the hell does this block do???", you need to comment it.
Actually you should proably take into account all the petrochemicals used to grow the food that powered the bike-peddler - fuel for the tractors, feedstocks for fertilizers and pesticides (unless the dude was eating organic). Then there's the energy/oil costs of transporting and refrigerating the food.
Education and culture are not neurological pathologies. That's the point.
To say that ideas and opinions can be states of disease is a very dangerous conclusion. If homosexuality can be a "disease" (as it was considered just a few decades ago) and racism can be a "disease", then any beliefs or ideas or emotional states unpopular enough with the official psychiatric community can be diseases. It's defintion by social contract rather than by objective analysis.
Failing to love the State is a disease, Comrade, but we can cure you with re-education.
Of course it's not a computer drive. But if the brain were static than our ideas and behaviours would never change. We'd be unable to learn, and there's be no way to change the troublesome ideas and behaviours we're talking about.
In fact our brain changes every day, thank goodness.
You could sign up and take class with him. It'll help with that extra 40 pounds, and martial arts training is one of the few athletic activities that parents and kids can do together. We have several families where one of the parents trains along with one (or more!) of the kids, and they seem to get a lot out of it.
The trick is that the cookie can be linked to your personal information.
The class "compromising cookie" scenario involves a cookie set by an embedded image from a different server.
Say that Evil, Inc runs a banner server banners.evil.com, which puts ads on kinky.xxx and on yourchurch.org (or maybe just an invisible "web bug" on either site). When you visit kinky.xxx, your browser requests the banner from banners.evil.com, which sets a cookie saying "I went to kinky.xxx and all I got was this lousy cookie". That cookie will be sent along with any request your browser makes to banners.evil.com.
Then you log in to yourchurch.org. Their home page has an image tag with a source like "http://banners.evil.com/spyonme.php?username=your name". Your browser makes this request to banners.evil.com, sending along the cookie that server set eariler. Your browser thus tells Evil, Inc your yourchurch.org username (in the image URL) and the fact the you visited kinky.xxx (in the cookie it).
Evil, Inc phones up your pastor and lets him know so that he can shame you in front of the parish the next Sunday (turns out this is all part of your church's anti-porn crusade).
I address this more in this post, so I'll just refer you to that.
Who says you can't get help in solving a problem?
The unethical accounting in question here took place in 1998-2000, and McAfee has cleaned house since then. This settlement is part of that process.
Disclosure: I'm a McAfee stockholder. So I've been watching this play out for year (and I was one of the many stockholders who watched their portfolio's value vanish in the haze of fudged accounting).
Contrary to the tone of the summary, this isn't a new development that's going to change the anti-virus market. This is McAffe (hopefully finally) settling a problem the occured back in the 20th century - the accounting irregularities in question happened back in 1998-2000.
Disclosure: I'm a McAffe stockholder, so I've been watching this unfold for several years.
This settlement pertains to actions taken in 1998 to 2000. The summary makes it sound like McAffe just got caught with their hand in the cookie jar, when in fact this is the company trying to clean up after an administration long since gone.
The stock went up after the announcement, so the markets seem to think the settlement a good idea.
(Disclosure: I'm a McAffe stockholder, due to stock options from an old employer and a series of mergers (TISX -> NETA -> MFE).
Hopefully they're not that corrupt anymore. The settlement pertains to actions taken in 1998 to 2000.
The summary makes it sound like McAffe just got caught with their hand in the cookie jar, when in fact this is the company trying to clean up after an administration long since gone.
(Disclosure: I'm a McAffe stockholder, due to stock options from an old employer and a series of mergers.)
I'l steal a page from Szasz here:
(BTW, that's not to say I agree with Szasz on everything.)
Why not? If I break my leg, that's not a illness, but it's covered by my health insurace. No reason why "problems with living" or "psychosocial dysfunctions" couldn't be covered.
A problem may reside directly at the level of the nervous tissue; alcoholism, for example, is a change in the (IIRC) ion channels along the axon. You could, in principle at least, look at the physiology and see the pathology. (Though the pathology is rather evident when someone gets severely ill or dies from the DTs.)
Other problems do not have such a pathology. Of course beliefs, behaviors, and ideas have some neurophysiological correlation, but not so directly. For example, there's no pathology of the nervous system in holding racist beliefs, which some argue should be considered mental illness. Problems like this are not diseases in need of a treatment, but dysfunctional relationships of the person with themself, other people, and the universe at large in need of adjustment. (Non-coersively, of course, unless the person is a credible threat to the rights of others.)
It's a trite analogy, but think of a "software problem" versus a "hardware problem". The software problem has an certainly has expression within the hardware, but it's not because the hardware is broken. Of course the brain is not a stored-program computer with a clean delination of hardware and software.
That's exactly what I mean by setting up straw-men. You're choosing your own defintion of homeopathy that is at odds with what is actually practiced, and then proceeding to rip it to bits.
If "Europe's oldest and the UK's largest manufacturer of homeopathic medicines" is making concoctions with significant concentrations of active ingredients, that's awfully strong evidence that as homeopathy is actually practiced, not all homoepathic remedies are prepared according to the "law of infinitesimals". If you want to argue, I guess you should take it up with the homeopaths who aren't following your defintion. :-)
According to the theory (which, again, I'm not saying I believe), yes, it does. Adherents claim that the water takes on some sort of "memory" or "image" of the substance; it "clusters" or absorbs some sort of "electromagnetic vibration" at each stage of dilution.
Thanks, I'll check that out.