Become a fan of Slashdot on Facebook


Forgot your password?
Get HideMyAss! VPN, PC Mag's Top 10 VPNs of 2016 for 55% off for a Limited Time ×

Comment Re:AI is just not ready. (Score 3, Interesting) 214

We're not there yet but this effort by Microsoft is, IMHO, as smart as a mouse.

Mice are pretty smart, I'd argue that the current AIs are at insect level of "intelligence".

What's obvious from these results is that the AI has no idea what it's looking at. This is typical for a trained neural net: it finds the best matching pattern in an image, and maps that to one of its output categories. It makes no difference between a random black and white blob, and a penguin, so long as they match the pattern.

A mouse, and true AI, will have spatial understanding. It will (intuitively) know that the images represent objects in space, and will be able to recreate a coarse 3D model of what they see. Then they will break down the scene in basic features, and identify it based on those features. It might say: hey, these blobs remind me of a penguin, but will never say that they *are* a penguin, because the blob will miss the beak and eyes and flippers and feet.

Basically, what we have now are the neural nets we already had 50 years ago, only on much faster hardware, combined with a bot and a web search engine. It's basically ELIZA on steroids, but still a long long way from actual intelligence.

Comment Re:Nah! (Score 1) 184

Some nice theories here but I'm sticking with my own pet theory: our observable universe exists entirely inside a black hole, slowly being compressed at the center across time.

Since we are in the exact center of our observable universe (per definition), that would mean we would be in the exact center of your postulated black hole.
Which would be both an astronomical coincidence (given the near infinite amount of space available), and a very bad place to be (since there should be a universe sized singularity in the center).

In other words: not a chance this is true.

Comment Re:Give us direct access to users (Score 1) 210

I agree with this.

There is on one hand architecture, and on the other usability. The end users shouldn't directly be involved with the architecture, but since they are going to use the thing, they should have a whole lot to say about how it should work. Not some guy who makes up some arbitrary user stories based on a few talks with end users (but those can be a base to start with).

In the (internal) project I'm working on, at first we had a non-programmer collect requests from the end users. He would come back with things like: user X wants a button here that does Y. Then I'd say, hmm OK, and what is he trying to solve/improve by having that extra button? What are the use cases? Then the guy would go back, and it would ping-pong a couple of times. In the end it was often more productive to go talk to that user directly.

There are caveats though.

End users often don't know what they want exactly, only that they have a certain frustration that they want fixed, and they will often propose what they think will solve it (very often a magic button). So you need to take the time to find out what it is they really need, and then see if there is a generic solution for it that fits in your architecture and process flow, that might be useful for more than one user.

They also often have a very narrow view of the problem, so you have to inform them what the impact is of their new feature on other users or other parts of the software. Quite often they didn't think it through and even left big holes in their own use cases. They often request features they need NOW, so you always have to weight the costs against the benefits.

Direct user interaction also helps setting realistic expectations. We can quickly guess if it will be a complex feature, explain why it is so complex, and why we probably won't be able to implement it. End users often expect the software to perform magic, using incomplete data to perform error free tasks. Things a human often cannot even do given all data and experience.

This all leads to less user frustration, and an overall more streamlined workflow.

Comment Re:what is their development strategy? (Score 1) 112

Every commit I make at work is required to have at least one peer review and its' recommended to have two and we are not selling security-related software.

You may vet your code all you want, but if someone compromises your build server's compiler, your binaries may contain backdoors anyway, just ask Ken Thomson

Comment Re:It has to be (Score 4, Informative) 207

No, I really meant constant, not linear. It is indeed odd, and known as color confinement.
But this property only exists at very small distances (sub atomic, nucleus scale), because once the field energy becomes too high with bigger distance, the energy is converted to mass, and these new quarks close the distance.
Outside the nucleus, the color field strength (and thus the strong force) is almost zero, because the colored quarks and gluons in the nucleus have a neutral color charge on average, similar to how positive and negative charges almost completely cancel each other out.

Comment Re:It has to be (Score 4, Informative) 207

It doesn't really make sense to compare the fundamental forces that way. Only the electromagnetic field and gravity propagate far enough to exhibit an inverse square law. This is simply because the field covers a bigger spherical area at larger distances.

The strong force stays roughly constant at growing distances. This is because the color field absorbs the energy used to separate the quarks, and interacts with itself via the color force (generating virtual gluons and quarks). When the separation gets too large (i.e. sub atomic distances), the field energy condenses into new quarks close to the original quarks, and the field between the original quarks disappears (almost, but not completely. The leakage makes nucleons stick together).

The weak force is even harder to describe in this way, since it doesn't really behave like a classical force.

So how do physicists compare these forces then? Each force is associated with a quantum field, and each field has some probability to interact with some particles. This probability is a constant number called a coupling constant, and can be determined by experiment. The fact that C14 has a certain half-life for example is caused by the weak interaction having some probability of turning a neutron in a proton (by changing the flavor of one of its quarks).

So it's the value of the coupling constants that determines the strength of the force, and on average the many quantum interactions between a field (or the bosons that are its quanta) and other particles (which are also just quanta of a field) manifest as a classical force that exhibits an inverse square law.

Comment Re:von Neumann probes (Score 2) 391

A real head-scratching conundrum about the universe is explaining why it's not already overrun with self-replicating robots.

Nah, that's easy : it actually *is* overrun, what else do you think all the dark matter is?
These robots are monoliths with ratios 1:4:5. Because they are black and full of stars, they are very hard to see against the cosmic background.

Comment Re:We may hear from Philae later (Score 1) 337

Of course, the comet will also start shooting off monster steam blasts, which could easily blast Philae off at escape velocity.

Only in Hollywood movies.

Meanwhile they asked this question to the ESA people (watch the press conference), and it turns out that (1) the expected mass emissions of the comet are not that large, so no monster blasts, and (2) the lander is still pretty massive (about a metric tonne) so it has a large inertia, and will be hard to move by the escaping gas. In summary, they didn't expect it to go anywhere soon (unless it got hit by a meteoroid).

Comment Re:95 years but (Score 1) 120

There's no such thing as a passive detector.

Sure there is. There is nothing special about a detector. If you can put a whole cat (=a bunch of atoms) in a superposition of quantum states, you can also include the detector (=a bunch of atoms) in that superposition.

It only works if the inside of the box (including the detector) is isolated from the rest of the universe. Then there is a superposition of 2 states: (1) the radioisotope didn't decay, the detector detected nothing, the cat is alive; and (2) the radioisotope decayed, the detector detected the emitted particle and released the neurotoxin, the cat died.

Once you break isolation (i.e. coherence), the rest of the universe entangles (at random) with one of these states, and the other state "collapses".

The big question is: what happens with the collapsed state? Was it absorbed in the other state? Did it entangle with a parallel universe? Did it become disconnected from reality?

Slashdot Top Deals

Beware of the Turing Tar-pit in which everything is possible but nothing of interest is easy.