Please create an account to participate in the Slashdot moderation system


Forgot your password?

Comment: Re:A Bridge Fuel... (Score 1, Interesting) 387

The reasoning is that natural gas releases less carbon than coal, so if we switch from coal to natural gas, then we'll reduce climate change.

Yes, I'm perfectly aware of that, and unlike you I know the science behind it. The problem is the next sentence of my post that you conveniently left out of your quote -- which is, if we don't actually reduce energy demand, we'll eventually run out of natural gas and have to burn the coal/oil anyway. So we just end up in the same place, just a few decades later.

Also note my primary objection is to the beginning of TFS which implies we could STOP climate change by this substitution, which is in fact idiocy if anyone thought it true.

Comment: Re:A Bridge Fuel... (Score 3, Interesting) 387

by AthanasiusKircher (#48592025) Attached to: The Shale Boom Won't Stop Climate Change; It Could Make It Worse

The solution to climate change isn't finding ever-more-exotic carbon to extact and burn - it's to stop burning carbon as soon as possible.

Agreed. TFS has got to be one of the most "duh"-provoking things I've seen posted here (and that's saying something). What kind of idiot thought we'd reduce climate change (which most scientists agree has something to do with carbon released from fossil fuel production) by switching to another fossil fuel that still emits carbon when burned? Unless we stop dumping carbon into the atmosphere, we'll still be dumping carbon into the atmosphere. We need an article to tell us this? What we need are other reasonable ways to harness and use energy and/or radically cut energy consumption until we only need renewables; until we have that, gas isn't solving our problem of using coal and oil: it's merely postponing our usage of that coal and oil.

Comment: Re:Someone has (Score 2) 270

by AthanasiusKircher (#48575651) Attached to: Keurig 2.0 Genuine K-Cup Spoofing Vulnerability

Considering the impact on the environment of pods that just ends up in the garbage there's now two reasons not to buy them.

OK, the coffee they make isn't bad, but what's wrong with an ordinary espresso machine?

(1) A Keurig doesn't make espresso -- its pressure is nowhere high enough. (2) Cost: if you really want espresso worth making at home, you're going to pay a LOT more than it costs for a Keurig. (Well, in the short-term anyway; if you keep buying the K-cups, maybe not.)

Anyhow, I would never have bought one of these things myself, but I was given one by a family member something like 6 or 7 years ago. She had used it, but had some trouble with hard water clogging things up, and eventually she got Keurig to send a replacement. But they requested that she remove the insert that allowed you to actually use K-cups, rather than sending the whole thing back.

The flaw in that scheme was that Keurig makes a different sort of permanent plastic "cup" that could be refilled with coffee grounds, allowing you to brew whatever kind of coffee you wanted. But in order to use it -- guess what? -- you needed to remove the insert.

Anyhow, after they had already sent the replacement, it too malfunctioned briefly, and this family member tried cleaning the old one -- and now it worked! (But obviously they didn't have the "DRM" insert to actually use K-cups with it, so they could only brew with actual coffee grounds.) Later they got the new one working again, so now they just had a spare sitting around... which was given to me.

It still works, 7 years later. I've never bought a single proprietary K-cup or even any off-brand ones. I've only ever used it to brew whatever coffee I grind at home.

I would sometimes use it for a fast cup of coffee, but eventually I grew tired of the inferior flavor and went back to a french press.

Point is -- at least with older models, you could brew with your own coffee grounds if you removed the insert and bought the special reusable thing for the grounds (which maybe cost $10 -- an amount you'd save even after a couple boxes of K-cups).

In that case, the environmental impact is really quite minimal and probably better than some other traditional home-brewing methods, since you only heat up enough water for a single serving at a time, rather than people who tend to make a pot of coffee in their drip coffee pot and then never finish the pot or let it sit on the burner keeping warm for hours.

Comment: The actual Zipf's law... (Score 5, Informative) 243

Speaking of the richness of languages, TFA oversimplifies some important language tendencies too.

For example, Zipf's law (which is also linked in TFS) has little to do with "familiarity" or being "more likely to select a word or phrase that we're familiar with."

It basically is just an observation that the statistical ranking of word in most natural languages is inversely proportional to its frequency. From the Wiki article:

Thus the most frequent word will occur approximately twice as often as the second most frequent word, three times as often as the third most frequent word, etc. For example, in the Brown Corpus of American English text, the word "the" is the most frequently occurring word, and by itself accounts for nearly 7% of all word occurrences (69,971 out of slightly over 1 million). True to Zipf's Law, the second-place word "of" accounts for slightly over 3.5% of words (36,411 occurrences), followed by "and" (28,852). Only 135 vocabulary items are needed to account for half the Brown Corpus.

Yes, I suppose one might get out of this that "we tend to choose words we're more familiar with," but Zipf's law is a MUCH more specific constraint on distribution of word frequencies. And it's more a statement about what word frequency distributions ARE rather than how we come to choose words or what we may be "familiar with," unless by "familiar with" you just mean "occurs more frequently."

Moreover, there is some research that has shown a distribution somewhat like Zipf's law will emerge even in texts generated with artificial random "languages" composed of random letters... which makes the claims about how we're making conscious or sub-conscious choices about "familiarity" even less likely.

Comment: Re:Free Enterprise (Score 5, Insightful) 184

by AthanasiusKircher (#48561285) Attached to: Swedish Police Raid the Pirate Bay Again

Before I respond, let me be clear that I absolutely think copyright law is broken and needs significant reform -- at a minimum to make the time until something goes into public domain be only a few years (like it was in original 18th century statutes), not a century or more That said...

Sharing/copying should be encouraged as a social good. Sharing of knowledge is what made our civilization, and keeps it alive. Voluntarily allow a few elite control over what may be copied and who can copy, and you weaken civilization.

I always find these sorts of arguments hilarious. Because you know who funded the arts before copyright existed? Rich dudes.

How did one become an artist in the age before artists could make money off of publications and copying? Well, you had two choices:

(1) Be independently wealthy. A lot of art, music, literature, etc. used to be created by only those filthy rich who didn't have to work for a living. So, if you had nothing else to do and were bored, you could afford to make art.

(2) You're not rich? Well, if you want to be an artist, musician, writer, or skilled craftsman, you have to find yourself another rich dude to fund your work. In other words, you found yourself a patron, because otherwise, how are you going to support yourself?

If you actually want art that requires significant SKILL and TRAINING to learn a craft, those are your primary choices without some concept of intellectual property.

There are other ways for artists to earn a living.

Sure, you can say performing musicians have to tour rather than making money off of recordings, but what about the composers who actually write the songs? Lots of pop artists don't make their own songs -- they rely on expert songwriters to do that. How exactly does one make money off of those sorts of creations? One can't exactly become a "touring songwriter." (I mean, yeah, improvisation is fun and you can make up crappy songs on the spot for a paying audience I suppose, but there's little incentive then to spend time crafting an actual good song...)

So far, I've just been talking about pop music, but it gets harder if you want someone with real talent to devote months or even years to an extended project -- like a book, for example. And how about training? Mozart spent maybe 15 years learning the craft of composition before he began writing stuff of a "mature composer" with thorough training in how to write music. Who pays for those 15 years of training before one can even begin to compose?... and then one's compositions are just shared with no reward for the person who spent his life acquiring the skill to make them.

That's ultimately the problem with these arguments. A system without any sort of intellectual property makes it much more difficult for anyone to spend significant time on any given creative project, since no money can be made from that lost time... let alone taking any time to learn a skilled craft.

Art thus becomes only an amateur occupation, something your crappy band in a garage does improving stupid songs on a weekend, but no room for any possible types of refinement or skills. We expect doctors and engineers and scientists and programmers to spend years refining their skills so that they can produce a quality product. And when they do, they are rewarded for their work. But if you're a skilled artist who took years to learn a trade, too bad -- we still want you to make art, but we want you to donate it to us for free. Find some other way to make your money, thanks.

Unless, well, you're a rich dude and can spend the time acquiring random skills and putting time in creative tasks that won't make you any money. Or if you can find a rich dude to serve as your patron.

Yeah, once you're an established artist with a record, you might be able to get some crowd-funding or something today, but good luck to get that money to spend a year or two researching and writing your first novel, which you then generously donate to the public without any reward.

Basically, you're right -- to some extent a bunch of rich dudes control what may be copied now, and that may be restricting art. But guess what? You get rid of intellectual property, and you also risk reverting back to a system where only rich dudes fund art again.

At least in the current system, you can vote a bit with your money -- you can influence the corporations by choosing what art to buy or not to buy. In a patronage system, it's really just the elite who get to determine what art gets created, because they're the only ones with the time or money to fund it.

+ - Deep Neural Networks are Easily Fooled: Is this Snowcrash for AI?-> 1

Submitted by Anonymous Coward
An anonymous reader writes "A new paper on deep learning produces snowcrash to hack deep neural networks, producing fascinating images and raising security concerns. The paper is called "Deep Neural Networks are Easily Fooled: High Confidence Predictions for Unrecognizable Images".

Here is the abstract:

Deep neural networks (DNNs) have recently been achieving state-of-the-art performance on a variety of pattern-recognition tasks, most notably visual classification problems. Given that DNNs are now able to classify objects in images with near-human-level performance, questions naturally arise as to what differences remain between computer and human vision. A recent study revealed that changing an image (e.g. of a lion) in a way imperceptible to humans can cause a DNN to label the image as something else entirely (e.g. mislabeling a lion a library). Here we show a related result: it is easy to produce images that are completely unrecognizable to humans, but that state-of-the-art DNNs believe to be recognizable objects with 99.99% confidence (e.g. labeling with certainty that white noise static is a lion). Specifically, we take convolutional neural networks trained to perform well on either the ImageNet or MNIST datasets and then find images with evolutionary algorithms or gradient ascent that DNNs label with high confidence as belonging to each dataset class. It is possible to produce images totally unrecognizable to human eyes that DNNs believe with near certainty are familiar objects. Our results shed light on interesting differences between human vision and current DNNs, and raise questions about the generality of DNN computer vision."

Link to Original Source

Comment: Re:the evils of Political Correctness (Score 1) 201

by AthanasiusKircher (#48546541) Attached to: James Watson's Nobel Medal Sells For $4.1 Million

Because it is a no-brainer that if you grab two random groups of individuals and measure ANY trait within them, you'd expect to find a difference in the mean. That is true no matter what the groups are, or what the trait is. Heck, if you grabbed 500 white people, took two samples of 50 out of that group, and compared just about any trait between the two groups of 50 you'd find differences. Hence the reason statisticians are interested in things like standard error.

Yes, all of this is true. And I think you've just proved my point. All sorts of "differences" can show up in random groups. The question is whether those differences are significant and meaningful (i.e., not caused by improper control groups or other confounding factors).

The variance among people of any given race in intelligence is larger than variance between races. So the question is whether those relatively minor variances seen between races are meaningful.

Black people and white people tend not to inter-marry. I'm not saying that it never happens - only that it doesn't happen NEARLY as often as intra-racial marriage. That makes it all the more likely for genetic drift to make some genes become more predominant in one population vs the other.

As I also said in my post, when you have appropriate control groups, most of that apparent disparity disappears. And even if it doesn't disappear completely, that doesn't mean that IQ is the sole measure of this monolithic entity called "intelligence" -- there could be many other things that lead to smart decisions and success in life other than that measured on an IQ test. (I'm not saying IQ doesn't measure something, but that doesn't mean it's the only thing....)

Would you expect the genes that govern skin color to be any different between the average african-american and somebody of european descent? Then why not other genes?

We could rephrase this question and say something like, "Would you expect the genes that govern where the heart is located inside the chest to be any different between the average african-american and somebody of european descent?" Answer -- probably not much. "Then why not other genes?"

The amount of COMMON genes between races is HUGE compared to minor differences. Those differences exist. But why would you automatically assume that any particular genes MUST be different when the vast majority of them are the same?

Look -- regardless of all of this, the reality is Watson didn't make a nuanced statement like this, "Oh, yeah, variance can reasonably happen between any group." He said nothing like that. He basically said he thought Africa was unlikely to improve its condition because black people are stupider.

That's not a nuanced statistical argument. That's stating something as a fact, and there just isn't enough evidence to support such a claim. The OP I was originally responding to was arguing that Watson's statements should be believed because he speaks the "truth."

So what I was responding to was an OP who was agreeing with a blatantly racist claim that is not supported by scientific evidence, not some nuanced "Oh, there might be some random variance" hypothesis....

Comment: Re:the evils of Political Correctness (Score 4, Interesting) 201

by AthanasiusKircher (#48544491) Attached to: James Watson's Nobel Medal Sells For $4.1 Million

For science to work you must be able to state an unpopular opinion and not get slaughtered for it.

Agreed. Has anyone discounted any of Watson's other scientific discoveries on the basis of this remark? I don't think so. And if not, science is still working as it should.

We're talking about a very smart guy that helped discover DNA.

I'm sensing a fallacious appeal to authority coming up....

If he says that there is a DNA element to intelligence (and everyone knows there is)

Yes, that's a true statement.

and that it varies by race (again, this is a no brainer)

If you're looking for the place where your post went from "misguided appeal to authority" to "racist rant," this is where it happens. Exactly why is it a "no brainer" that intelligence varies significantly by race?? I've personally met some very smart people of all sorts of races, and I've met idiots from all sorts of races too. I don't feel like I've accumulated enough data to say it's a "no brainer" that one race is smarter than another -- what dataset do you have access to where you feel like this is a "no brainer"?

Also, you referenced IQ earlier, and now you're talking about "intelligence" -- are you rejecting the idea that different races might have evolved different sorts of intelligence if you're presuming they've evolved differently enough to have different adaptations in this area (and maybe those localized adapations might not be measured as precisely as a test designed mostly by white people to test white people)? I'm just mentioning one of many problems with IQ as a proxy for "intelligence," even if there were obvious differences... which there aren't. When you control for demographics and other social aspects, a lot of racial differences narrow significantly.

then what is the big deal, he's speaking the truth.

The big deal is when he made these remarks, he was no longer just some smart young scientist. He was an 80-year-old dude with a history of making racist, sexist, and homophobic remarks with little basis. And, let's be honest here, even many great scientists aren't always going to be "at the top of their game" anymore at 80 years old.

So your appeal to authority here is problematic in a number of ways -- a guy was recognized for an achievement more than a half-century ago, he's old, he tends to say things that aren't true or well-thought-out in public, and yet you just assume he "speaks the truth"?

Why? THAT does not strike me as a very "scientific" attitude.

Comment: Re:the evils of Political Correctness (Score 4, Insightful) 201

by AthanasiusKircher (#48544393) Attached to: James Watson's Nobel Medal Sells For $4.1 Million

You are probably right about the confirmation bias. But one should be able to make that argument without hounding someone out of a profession. That is more-or-less what happened here.

No it's not. The guy has continued to revise his books and memoirs and other publications in recent years, which is more than you can say for most 86-year-olds. He has continued to publish new scientific ideas in recent years.

What actually happened is that he wrote a memoir about his life which was intended for a POPULAR audience, and in the early stages of gearing up for his book tour, he made the remarks everyone's been talking about. Most of his appearances on that book tour were then cancelled, because of reactions to a public figure who basically implied that the science on the genetics of race was settled (when it's really not -- there may be some studies that appear to agree with his claims, but there are about as many that show the opposite) and then made racist implications on the basis of this.

He was not at all "hounded out of a profession," unless you consider "being a public intellectual" a profession. Show me evidence that people have refused to publish his research or took away memberships in academic societies or whatever -- then you can say he was "hounded out of the profession." He wasn't. He did lose a high-profile administrative position, but he continued to advise and do research at that place. He just lost his audence to talk to the public, which he should, given that he has a long history of saying rather nasty things and claiming a scientific basis for them when there generally isn't.

This is a classic case of claims of "Science!" being used as a cover for political correctness. More like "Science! (so shut the hell up)".

Huh? Look, you want to be a "normal scientist" and go about your day, doing research, publishing papers, whatever -- that's great. And chances are if you make some crass or racist remark to some random friends, nothing's going to happen to you.

But if you want to be a world-famous scientist and live in the public eye, you are subject to public scrutiny -- which means when you say something that's not true AND offends people in the process, you might lose your public audience.

That has nothing to do with "science." It's just the reality of being a public figure. It would be one thing if this were a single off-hand comment from Watson. It was not. He has a history of saying things that are racist, sexist, homophobic, etc., and he's been doing it for decades. (He's also, frankly, a bit of a kook in his old age, but that's a separate issue.)

You want press? You get to accept what press you create for yourself....

Comment: Re:Its own editors said so (Score 4, Insightful) 346

The second is that we have this misbegotten notion that "balance" is that we must give both sides of a story equal billing. When one side is flagrantly wrong, it deserves to be dismissed and ignored.

I absolutely agree with you that there's no need to present ideas that are demonstrably false. But "liberalism" is not something that is easily proved "true" or "false." A group stating that they are promoting "liberalism" could mean many things, but a political ideology is NOT a synonym for "truth."

The third is on you to show how they omit facts. I know the extreme right wing of this country loves to manufacture "facts" or omit actual facts when it suits them, there's a whole TV network that excels in such shitflinging.

I know many "conservatives" who use misunderstandings (intentional or not) or outright lying to promote their ideas, but I also know "liberals" who have done the same. And there are people who have fundamental ideas about what they think good policies might be on both sides who try to stick to the truth.

In any case, getting stuck in one's own ideological bubble means that it can be difficult to see the truth -- not always because you're deliberately lying or because any of your ideological buddies are lying. Often the sides talk past each other -- so you always get your "talking points" and never really have to seriously consider rebuttals from the other side... or if they occur, you just laugh and dismiss them, and your group of ideological friends laughs along with you, because it's easier than confronting real philosophical fundamental inconsistencies that are present in any real-world political ideology.

Whether you're a fan of the Rush Limbaugh show or NPR or Fox News or DemocracyNow! or whatever, you get the slice of "news" that best represents what the producers/editors think is important.

"Balance" is an issue not so much about truth, but about making sure that opposing opinions are considered in cases where there are real, actual conflicts with no one "truth." And it's also about running a variety of stories that sometimes might bring up "inconvenient" problems with your pet ideology.

There's absolutely nothing wrong with a magazine or whatever saying, "We're going to slant toward liberalism." But "facts" can always be selected, even if they are all true. Magazines and newspapers have to figure out what stories to run, and they will select them in ways that will promote or emphasize their ideology. For people who don't ever step outside that "ideological bubble," though, they could end up with a pretty skewed perception of the world... even if every single sentence in the magazine is "verifiably true."

That's why "balance" -- in general -- is important. Even more important is diversity of opinions, diversity of experiences, and diversity of ideologies. If you don't have those things, you can still up distorting things to adhere to your chosen ideology... even unintentionally.

(P.S. In case anyone's making assumptions and gearing up for ad hominem, let me be clear that I would never identify myself as a "conservative" (whatever that means). I believe that the one-dimensional idea of a political spectrum that encompasses all possible ideas is fundamentally flawed and leads to distortions, groupthink, and doublethink -- because all possible issues have to be crammed into some space along the spectrum, despite many underlying inconsistencies that arise.)

Comment: Re:Logic fail (Score 1) 186

by AthanasiusKircher (#48522433) Attached to: Pizza Hut Tests New "Subconscious Menu" That Reads Your Mind

So you can't have it out of a 1,000 degree oven, so you're going to put it into one that struggles to reach 500 and which heats your house in the process whether you like it or not?

My home oven easily tops out over 550 with normal bake cycle on. I have multiple thermometers I use to measure it, so yes, I know what temp it is. If I put the broiler on before throwing in the pizza, I can get it even hotter. (I'm not willing to break the lock and use the self-cleaning cycle as some do -- I value my home insurance.)

But the biggest difference is the steel. Sorry, but stones just don't cut it. I didn't realize this until I got one a year ago or so, but having baked pizzas for many years, I was positively shocked the first time I used a steel in terms of the difference it made. The heat transfer is just so much higher than a stone -- it easily cuts my baking time down by at least 30%. Between the increased heat transfer from the steel and the broiler above (with its excess radiant heat), I'm easily getting to a heat transfer rate comparable to a grill with a stone over 700F.

To be a proper hipster you're going to have to at least put it on a stone on your BBQ, which you can get well up into the sevens if it's any good.

I have no idea what "hipsters" have to do with anything. I've been baking pizza and bread long before "hipsters" became a common thing. I just like good food.

Anyhow, if I were to do it right, I'd build a brick pizza oven in my backyard. I don't have the time or energy for that now, and frankly I'm not that obsessed. I do have a friend who has one. Then you get to proper temp. I've used ceramic grills too, but that's a bit more work than I want to deal with every week.

In any case, I mostly make pizza in seasons when the heat for my house is actually useful, and if you are using a ceramic stone on your grill, you're probably not getting the maximum effect. Replace it with a steel plate, and you'll get much better heat transfer, probably enough to get close to a proper sub-90-second Neapolitan bake.

Comment: Re:Who's their test group? (Score 1) 239

by AthanasiusKircher (#48521485) Attached to: Google Hopes To One Day Replace Gmail With Inbox

but of course google inbox 2034 will contain the term "actionize" instead.

Yes, but only for a year. Then it will be replaced, like all words in Google products, with some obscure icon on a button -- containing four seemingly random geometrical shapes and some weird lines between them -- "obviously" (according to Google's design team "experts") having the specific meaning of " actionize."

By this point, Google's support will also have replaced clear simple words with answers to their FAQ in only pictograms. Serious users simply give up and just click random buttons for five minutes every time they need an action ("actionization"?) beyond "read" and "send," hoping to hit the right icon by accident.

Comment: Re:Logic fail (Score 1) 186

by AthanasiusKircher (#48519587) Attached to: Pizza Hut Tests New "Subconscious Menu" That Reads Your Mind

If was Naples wouldn't that be degrees C?

The official regulations state minimum oven temperatures as 430 degrees C for the oven floor and 485 degrees C for the oven dome, with a bake time of 60-90 seconds.

That's MINIMUM of ~800F for the oven floor and minimum 905F for the air temp, to qualify as authentic Neapolitan pizza. In practice, many ovens are higher than this. I was just stating an approximation.

The use of anthropomorphic terminology when dealing with computing systems is a symptom of professional immaturity. -- Edsger Dijkstra