Slashdot is powered by your submissions, so send in your scoop


Forgot your password?

Comment: Not exactly a hack (Score 1) 73

by michelcolman (#49604381) Attached to: Hacking the US Prescription System

So you enter someone's name and date of birth on this website, and it gives you all the details? How exactly is this a hack? If I asked the president of the US for the nuclear launch codes, just for laughs, and to my great surprise he would simply give them to me, would I have "hacked" the US nuclear missile system? Would I be thrown in jail for hacking?

This is just plain irresponsible behaviour by PillPack, nothing to do with hacking.

Comment: Re: Error in headline (Score 1) 292

I was just offering an alternative explanation as a possibility, not taking sides. When I use one part of the article to cast doubt on another part, that doesn't mean that I believe one part or the other. I just like to point out other possibilities when people jump to conclusions based on very scant information. Spectacular headlines often turn out to be less than accurate.

Comment: Re:Error in headline (Score 5, Interesting) 292

I must agree that you can't really say anything useful about this incident without knowing what was in the paper. The title makes it seem like an outrageous situation: "Scientists Have Paper On Gender Bias Rejected Because They Are Both Women". I imagine that what actually happened is probably a lot more nuanced.

It could very well be that the quality of the paper was rather poor. The article says the two female researchers just looked at the number of papers submitted by men and women, the number of jobs they applied to, and how long it took them to get accepted for a position. They then apparently concluded that, since women tended to be less successful, this obviously proves the existence of gender bias because the quality of the work cannot possibly be different, you know, men and women being equal and all.

I know that the article only gives a brief and possibly distorted summary of the paper, but if this was indeed the content, the reviewer has a perfectly valid point saying the results could also be explained by a lower quality of women's work. That doesn't mean that this is indeed likely to be the case, just that it would be an alternative explanation that must be ruled out before you can conclude anything about gender bias.

There have been other studies on academic gender bias, for example the one where identical papers were sent in with either men or women listed as the authors, and noting the discrepancy in their acceptance. And yes, a bias did indeed show up there, so I certainly don't rule it out, but you have to use proper methods instead of jumping to conclusions.

The remark about including a male co-author is obiously not a very smart one, but I kind of understand the reason for that suggestion too: a paper on the Palestinian conflict written by Jewish and Palestinian co-authors is more likely to be neutral than a paper written by only Jews or only Palestinians. So for this particular issue, having a male co-author is probably not a bad idea. Especially if they jumped to a conclusion about gender bias without ruling out alternative explanations, which would actually suggest gender bias in their work.

Once again, I haven't seen the paper so this is all just speculation on my part. The reviewer certainly could have phrased his comments a little better, though. Maybe he was just poking fun at them for writing an obviously flawed paper, but it clearly didn't get interpreted that way.


Australia To Grade Written Essays In National Exam With Cognitive Computing 108

Posted by timothy
from the oh-this-will-work-well dept.
New submitter purnima writes: Australia keeps on giving and giving. Each year school kids in Australia sit The National Assessment Program (NAPLAN) which in part tests literacy. The exam includes a written page-long essay aimed at examining both language aptitude and literacy of students. Of course, human-marking of such essays is costly (twenty teacher-minutes per exam). So some bright spark has proposed that the essays be marked by computer. The government is convinced and the program is slated for the 2017 school year. Aside from the moral issues, is AI ready for this major task?

Comment: Re:Wouldn't it make more sense... (Score 3, Interesting) 34

You could put on three spheres, so that any objects are always visible by at least two of them. Then you can do some processing (quite a lot of processing, actually) to synthesize the images that would have been seen by two cameras at a fixed distance from each other pointing in any particular direction. I'm not saying it's easy, but certainly feasible with today's processing power. And it would result in less lag than actually having to physically move the cameras. Also, multiple people could use the same feed like popo suggested.

There are obviously some disadvantages, one of them being the much higher bandwidth required to capture 360 degree vision from three cameras in sufficiently high resolution so that a relatively small view window still keeps enough pixels to look good.

But lag is an extremely important issue if you don't want people to get seasick within minutes of using the device.

Comment: Re:According to TFA (Score 2) 262

by michelcolman (#49579435) Attached to: Crashing iPad App Grounds Dozens of American Airline Flights

Yes, because articles on the internet and in newspapers always only contain exactly correct details, no information ever gets lost, misunderstood, or altered in transmission. So if they say the screen was black, it couldn't possibly have been any other shade, and certainly could not have had any text on it, like an error message or something like that. Because journalists never get this kind of thing wrong.

OK, back to reality. Since both the captain's and first officer's iPad "went black" (?) at the same time, and this in multiple airplanes, even after many months without this problem ever occuring in a rather large fleet of airplanes, I imagine this is probably some configuration error related to some sort of DRM, licence expiration or other kind of protection. I doubt multiple iPads would all "crash" at the same time. Maybe the database had an incorrect expiration date, for example. Must not let pilots fly with out of date charts, better give them no charts at all. That sort of thing. Wouldn't be the first time, I've had a few experiences like that in different airline companies.

We've had an airbus grounded because a student pilot had messed with the on board clock, for example. The computers decided that the deadline for flap inspection had passed (based on the incorrect date set by the student) and refused to extend the flaps for take-off. Maintenance action was required before the plane could take off again.

Comment: Re:quacks get front page (Score 1) 129

by michelcolman (#49567003) Attached to: Holographic Principle Could Apply To Our Universe

I've never understood the rationale behind a holographic universe. I'm not saying it's impossible, but Occam's razor would seem to suggest it's unlikely.

If I understood correctly, the theory goes something like this: Information cannot be destroyed, yet the actions of a black hole on its surroundings (i.e. what we can observe from outside) are completely determined by its twodimensional event horizon, therefore a volume of that size can only contain as much information as what is present on that surface area. Since you can imagine any area of the universe collapsing into a black hole, and again no information can be lost there, the entire universe can only contain as much information as a twodimensional sphere, so the universe really only has two dimensions.

But why is everyone so sure information cannot be "destroyed" (rendered inaccessible) in a black hole? Many laws of physics break down in black holes, what makes you so sure the second law of thermodynamics will hold?

And it's not like this second law is really a "law" in the strictest sense. Mathematicians would never call this a law. It's like people observing a box with red and blue marbles in it. When they shake the box and then look inside, the marbles are all mixed up. Therefore, they introduce the "law" that the marbles will always be mixed up after shaking the box, and it will be impossible for all the blue marbles to be on one side and the red ones on the other.

Now this second law does turn out to be very practical. Because for all intents and purposes, given enough particles, you may reasonably expect to never see a violation of the second law. And certainly on the scale of the universe, it would be... well... "impossible".

But black holes are not like the rest of the universe. Extrapolating the second law, which isn't even really a real law but kind of a quasi-law, to black holes (which we know little about, apart from speculation) and then jumping to the conclusion that this must mean that the whole universe is twodimensional, is just silly. It's people performing "logical deductions" without asking themselves if that particular step actually makes sense.

Also, the fact that we can only see a certain amount of information from outside the black hole, does not mean that more information could not be present inside. What if you drop into the black hole? Of course you wouldn't survive, but if you disregard that detail, couldn't you find all the "lost" information inside? Applied to the universe, this might mean that you could replace a chunk of the universe with a sphere of that size, and you could calculate the effect of that chunk on its surroundings based on that twodimensional sphere, but that doesn't mean the information inside does not exist.

Also, if gravity inside a black hole is so strong that our usual laws of physics break down, it doesn't seem outrageous to assume that information could indeed really be lost there. Why not? Which is more likely? Information getting lost in a singularity, or our threedimensional world being only twodimensional? I would say Occam's razor favors the former.

Information gets lost all the time anyway, in the expanding universe. There are plenty of galaxies that are moving away from us so quickly that we will never be able to see their light. We are "losing" information all the time, yet this doesn't seem to bother anyone.

Really, maybe I'm missing something, and by all means feel free to enlighten me, but I fail to see a compelling reason to assume the universe has only two dimensions, it seems to be an unlikely solution looking for a problem.

Comment: Re:Cautionary Tale? (Score 1) 182

Keep saying that until, a few millennia from now, pretty much every newborn needs immediate medical attention to fix all sorts of life-threatening problems, breast feeding humans can only be found in history books, everyone needs eye surgery to be able to see normally, we can no longer smell anything, babies can only be conceived using IVF because natural fertility is about zero, and so on, and so on. Hey, we can fix all of that with medical procedures, so they are not serious defects anymore, are they?

Eagles have excellent eyesight because their life depends on being able to spot a mouse from high up in the sky. Moles have nearly zero eyesight because they rarely come out of the dark. Yet moles probably evolved from some other animal that did have good eyes. Evolution is a constant balance between natural selection and random degeneration from mutations. It's not black and white, things just evolve to the point where the two "pressures" match. If some quality no longer makes a difference for your ability to reproduce, random mutations over the course of millennia will slowly erode it away. And this random degeneration happens much more quickly than positive evolution, because it's much easier for a random mutation to mess something up rather than improving it.

Of course this does not mean we should re-introduce natural selection by letting people die or prevent them from reproducing, as certain societies have done in the past. But genetic techniques offer a humane way of achieving the same goal. It's already starting to be used to weed out obvious and serious genetic defects, by selecting embryos not carrying the defect, and we'll slowly get better at it so we'll be able to at least stabilize, and possibly improve our genome. If done responsibly, this needn't result in dystopian future societies with rich superhumans and poor degenerates. Just randomly weeding out some bad genes and introducing a sprinkle of good ones is enough to keep things going in the right direction, they will spread out naturally to the benefit of all mankind, while keeping enough diversity.

Comment: Re: War (Score 1) 154

by michelcolman (#49535215) Attached to: Hubble Spots Star Explosion Astronomers Can't Explain

Or, due to a strange quirk in the space-time continuum, we are actually looking at a future version of our own planet exploding. Don't worry about paradoxes ripping the universe apart, though. The fact that we can see it, means that there's no way to avoid it so there won't be a contradiction. Moving right along.

Comment: Re:Cautionary Tale? (Score 1) 182

and the fact is there's an inverse correlation between IQ and number of children.

Well, that's exactly the opposite of what I remember from that study I mentioned, but then again, you can probably find studies either way depending on what the researchers were trying to prove. I can't find the link for mine, but would be interested to read more. (Actual articles and papers, not just opinions).

Comment: Re:Cautionary Tale? (Score 2) 182

Stupid people tend to have a lot more kids than smart people.

Fortunately that doesn't appear to be true. Sure, quite a few geniuses with a 150 IQ have trouble finding a mate, but there are just as many people (by definition) with an IQ of only 50. How likely are they to find someone and have kids?

Closer to the average, higher intelligence definitely makes people more attractive, not less. Girls don't want to marry someone who's dumb if they can get a smarter partner. I remember reading a study demonstrating a positive correlation between IQ and procreation, but it was a few years ago and I can't find the link. Anyway, it seems to make sense, notwithstanding trailer trash with dozens of kids and business women running out of time to start a family. Apparently we're still doing OK on average.

There are plenty of other areas, especially health-related, where our genome is slowly degenerating for lack of selection since we can fix so many defects with medical procedures (or simply don't need high quality senses anymore to survive), but fortunately the evolution of intelligence appears to still be going in the right direction.

For all those other qualities that are degenerating, we'll need some kind of genetic manipulation or selection at some point, but we'll be OK for quite a while before it really becomes a problem, and by then we will have gotten over our ethical objections (apparently the Chinese are well on the way).

In case of injury notify your superior immediately. He'll kiss it and make it better.