Catch up on stories from the past week (and beyond) at the Slashdot story archive


Forgot your password?

Comment Re:Done us all a favor (Score 1) 629

Who gets to decide what is "absolutely non-disputable"?

There is such a thing as established facts.

There have been a number of extensive court cases in Germany where the Holocaust-deniers on trial had ample opportunity to present any actual evidence they have.

You - like them - make a pseudo-argument that they're a repressed minority opinion. They're not. This is not a question of opinion, but of historic fact, and its misrepresentation for political reasons.

the path from "you cannot praise the Nazis" to "you cannot criticize the government" is steep and slippery.

That's so crazy you should get your head examined.

First: The laws are not about not praising the Nazis, they are a lot more specific and revolve around actions and statements of fact, not about voicing an opinion. You can say "I think the Nazis were great" on public TV in Germany and you won't get jailed.

Second: We have laws providing you the right to critizise the government, right in our constitution.

Third: The slope between these two only exists in your imagination.

Comment Re:"rights" (Score 1) 245

Ok, so you're just talking about how the brain is not just a turing machine.

More precisely, I'm talking about how the brain is not just a machine.

Physicalism means everything that is exists within the realms of physics and let's forget about magic and souls and higher beings.

Mechanism means that everything can be explained with deterministic cause-and-effect chains, and things like randomness and free will are just an appearance of complexity.

I don't doubt physicalism, to the best of our current knowledge it holds true. Mechanism, on the other hand, has been on its deathbed ever since Einstein made his famous quote about the dice.

The limit of a map as accuracy approaches unity is an exact copy of the territory,

There is still a difference you can't get away with: The map will be in a different place. The map, even if it is a 1:1 copy, never is the territory.

Maybe I should've said that no matter how detailled and precise it describes the dishes, you shouldn't eat the menu.

Likewise, whatever function it is that constitutes "feeling", if we built something that perfectly carried out that function, or a function sufficiently similar to it, then it's more accurate to say that we built a thing that feels (perhaps not exactly like a human feels, but feeling nonetheless), than merely a thing that simulates feeling.

You still assume that emotions are a function. I claim it isn't. The details are still poorly understood, but emotions seem to be parts of a bigger network encompassing adaptive states and learnt reactions, processing shortcuts and acquired traits. What all that means is that it is unlikely that you will be able to isolate an emotion from everything else going on in the brain. Basically, you can't create true emotions without creating an entire brain-body replicate with consciousness, memory, education etc. etc. etc.

At which point you're building a human being, not a robot.

Last I checked, the so-called "hard problem" of "phenomenal consciousness", to which the question of emergence applies, was still an open issue, with competing answers to emergentism ranging from total skepticism (there really is no such thing, it's a confused idea) to refinements of panpsychism (such as panexperientialism, panprotopsychism, and panprotoexperientialism); and the easier problem of "access consciousness" was considered a mere triviality now, easily understood in mechanistic terms.

That's a mouthful and I admit I had to look up a term or two. I'm not really sure how any of this relates to the original point, though. If consciousness can be constructed by any means, I hold it to be reasonable to assume that we can engineer it as well, and not just duplicate humans. If we can engineer it, there is no reason to include the bugs of the naturally evolved system. Meaning the social rules we have created in order to not trigger them unnecessarily won't apply to engineered conscious beings.

Comment Re:"rights" (Score 1) 245

but it is still a physical thing carrying out a function that can be mathematically described, and there is no reason in principle why a different thing cannot be built to carry out that same function.

Possibly, we still don't understand everything yet. There might be some quantum effect uncertainty involved, which means you can describe it statistically, but reproducing it is a bit more challenging. But those details aside, it does seem theoretically possible to construct a brain replica.

Which doesn't mean that thing would a anything like a computer. Just because we can build it doesn't mean it's a computer.

o know that we'd first need to properly analyze exactly what it means to feel, before we can know if such a function can be implemented on a turing machine or not.

We know that you need consciousness of some kind to feel, as emotions are internal states. A turing machine does not ascribe "meaning" (I'm using these terms in the broadest possible sense here) to its internal states, so it can't "feel" in any sense that would even resemble what we mean by it.

And again, emulation is not the real thing. A good actor can display an emotion as good or better as someone actually experiencing it. And yet I'm sure you'd agree that playing Romeo on stage and being in love are two entirely seperate things.

You can certainly use genetic algorithms, say, to create a machine that lights an LED labelled "fear" if you show it pictures of, say, spiders. But if you claim that this means the machine has an emotion, then you and I have very, very different understandings of these basic terms.

Comment Re:"rights" (Score 1) 245

What is the functional difference between an emotion and a simulated emotion anyway? If the robot feels sad, then it is sad.

The difference between a simulation and the real thing is appearance vs. existence.

A simulated emotion would not be an internal state, but merely an outward representation of one. The robot would not feel sad, it would simply act as if it were.

You might as well dismiss human emotions as mere illusory products of chemical processes in the brain.

The fact that experiencing an emotion is different from describing one in a textbook doesn't make the experience illusory.

Comment Re:"rights" (Score 2) 245

If you make a computer out of huge piles of lego so the logic gates work the same, is that an exact copy? No. But it can run the same software.

You are missing the point.

Lego or silicon is just a different implementation of the same concept.

Brains and silicon do not share the same conceptual design.

The fact that the software that runs on a brain isn't written in anything like the programming languages we understand, or maybe could ever understand, doesn't stop it from being an entity of information processing that can be replicated by any sufficiently powerful computer.

I invite you to update your knowledge on how the brain works with more recent research. From what I gather (mind you, I'm not a scientist in this area) the brain is not just a powerful computer. It is not running software of any kind. It's not a matter of programming language, but of completely different concepts.

I'm not enough of both an expert on the subject and a science writer to make the important arguments in a paragraph or two, so I'll have to leave the argument at this. Everything recent I've read on the subject indicates that the "brain is like a computer" model is much like Newtonian physics - an interesting approximation that is useful for some superficial estimates, but once you look closer you find that a paradigm change is required and the equivalent of quantum physics is just being started on.

And for the other side: Computers are not unlimited processing engines. They have built-in restrictions and assumptions that are fine for what we're using them for, but do not necessarily translate to other entities.

Comment Two wrongs don't make a right, though (Score 4, Insightful) 572

Well, yes, but my experience is that even if I've never screamed at an admin, nor informed them of their mothers' extramarital activities, the majority seem to make it their duty to keep me from doing my job anyway.

In fact, for some (I'm looking at the fucktard duo administering the MQ server,) the nicer you are and more willing to explain why you need a queue for the application already approved by anyone who had a legitimate say, the more they'll abuse that and your time by MAKING you have to explain for weeks or get nothing from them. The guys who do tell them to STFU and do their own job, now those get what they asked for.

Now I have sympathy for admins, and understand that other people shit on their day. But WTH does it solve to in turn have them shit on MY day and my coworkers' day?

If X bullied admin Y, and Y bullies innocent bystander Z in turn, what did it solve, other than make an extra person unhappy? And how does the former even excuse the latter, anyway? Much less make it right. Two wrongs don't make a right.

Comment Re:"rights" (Score 1) 245

Care to post a link to this revolutionary new research debunking physicalism? I'd have thought it'd have made the news.

I didn't doubt physicalism.

For a while, the thought that brains are just highly sophisticated computers was "in". Then additional research found that brains are quite unlike computers in almost every way. Both do process and store information, but the how is about as different as it gets. Dyson, Bennett, Jaynes, Rucker, Stapp, Hawkins et al have all written interesting books about fragments of this larger body of knowledge.

and a sufficiently perfect emulation of such functions would constitute "feeling" just as much

Uh, no. A sufficiently good map of an area is still a map. Unless you re-define "sufficiently perfect emulation" to mean "exact copy", at which point it ceases to be an emulation.

But that is an important distinction to make, given that consciousness is - to the best of our knowledge - an emergent property.

Comment Re:"rights" (Score 2) 245

What if the only way to achieve the artificial intelligence necessary for them to be useful in their intended role is through processes which mimic biological development?

I know people who suffer from the bugs of these systems. Depression, bipolar disorder, stuff like that. If you intentionally create a being with these faults, you are a monster, on par with serial rapists.

That said, I also believe there is a difference between emotions and simulations of emotions, and I don't see why we would need the former. AI research used to be about mimicking humans, that was a fad for some time. It's long over, AFAIK. So, basically, unless you have some facts and evidence to support your core assumption, that this is the only way, the rest of the argument is moot.

Comment Re:It's opposite land (Score 1) 4

Perhaps you are a little unsure of some basic science. Almost all the food you eat has been genetically modified over thousands of years. What you might think of as natural didn't exist even 1000 years ago in its present form.

GMO food has two differences from selectively-bred food. One, it selects the genes in question extremely precisely, where breeding is hit or miss, choosing all the gene changes as a group when only one or a few are what is wanted. This is like writing a book by choosing random words from the dictionary, but claiming to be selective by limiting your choice to all but ten random pages.

Two, it happens much much faster. What might take thousands of generations and ten thousand years now takes only a few years.

Do you build houses by throwing trees, rocks, and other raw material in thousands of piles and choosing the bets one? That's what natural breeding does.

Anyone who opposes GMO as not being "natural" is scientifically illiterate.

Comment Re:"rights" (Score 0) 245

And what exactly do you think a human being is? It's a robot with a computer brain that has been programmed to simulate 'feeling' things.

Uh, no it isn't. That's a mechanistic view that has been out of fashion for at least 10 years. I strongly recommend reading some more recent research.

Comment "rights" (Score 3, Insightful) 245

Leaving aside the obvious, that we're talking about a fictional space opera, not some serious SciFi here, what's wrong with this approach?

We give rights to people and to animals because they are feeling, living beings. A robot or computer does not feel anything unless you've programmed him to simulate such a thing. The difference between your iPhone and the androids of SciFi is much smaller than the difference between a microbe and a human being, and we kill millions of those every time we use desinfectant spray. Don't recall anyone getting murder charges for that.

Despite all the make-belief, androids aren't human and don't suffer. There's no point in giving them rights. As a matter of fact, programming them so that they can suffer (instead of simulating an avoidance algorithm) would be the cruel part, not what comes after.

Comment Re:So... (Score 1) 467

Where do authors make money

With or without "piracy", that is a real question, because the vast majority of authors don't.

I'm not sure about the exact numbers as it's been a while since I read the articles on this, but roughly 90% of authors don't have an income worth mentioning from their books or articles. Of the remaining 10%, less than 1% makes their living entirely on their writing. The vast majority of authors have a "real" job in addition to writing.

Comment Re:strip (Score 1) 467

I don't think that the robot would change words for ones of similar meaning like a badly done thesaurus on steroids.

That is exactly what it does. The article with the examples is in german, which is my native language. There are examples where it changes linebreaks to spaces and such, but most of the examples actually do change words.

Slashdot Top Deals

EARTH smog | bricks AIR -- mud -- FIRE soda water | tequila WATER