Please create an account to participate in the Slashdot moderation system


Forgot your password?
Slashdot Deals: Cyber Monday Sale! Courses ranging from coding to project management - all eLearning deals 25% off with coupon code "CYBERMONDAY25". ×

Comment Re:LOL ... (Score 1) 63

As I understand it, random bets should yield about the same outcome. You just get fewer, larger payouts. Either way, the house is taking its piece, and you share the rest with the other players.

Ideally, the stats that matter are matching your knowledge of the race against everybody else's. If you know that this horse does better than people expect, or you know that some horse is a sentimental favorite but isn't likely to perform well, you can beat the other players and walk away with more than your randomly-determined share of the money.

It would be interesting to see how well the bettors actually do. The outcome I described above only works if the odds are mostly equivalent to the true odds. If a lot of bettors are betting badly, it would be easy to beat the market. It's generally hard to beat the stock market, except under narrow circumstances where you really do have better information than most (without ticking over into inside trading). Are horse tracks mostly filled with knowledgeable people, or are they just a bunch of rubes waiting for smart people to take their money? (Besides the house, of course, which always wins.)

Comment Re:I'm not a runner, but... (Score 1) 169

Even without data to back it up, it does seem reasonable that a bike would be dangerous with headphones on. You're sharing the road with cars, and you've moving very fast. There's less room for error, and if you have to ditch, you hit the ground pretty hard.

It's tricky, since there's so much wind noise that it can be hard to hear cars coming anyway. Frankly, I just don't feel all that safe on a bike, and I prefer running. Worse, I find cycling duller than running, since I can't let my mind wander as much; I have to constantly pay attention for anything that might throw off my steering (glass in the road, cracks, objects, etc.). So I often put one headphone in as a compromise, though I'd like to try getting a mount that might let me listen through the external speaker.

Comment Re:Scientists (Score 2) 203

A proof is a proof, regardless how extraordinary, extravagant or hillarious the claim is.

Well... yes and no. Even a mathematical proof isn't just a proof, because humans are involved in creating and checking it. A complex proof requires considerable work, and occasionally even a fairly sturdy result has to be withdrawn and reconsidered. Some examples.

Real-word experiments are never "proofs" in the mathematical sense. Directly, the only thing you can say about an experiment is "this thing yielded this result on this occasion". Everything else is extrapolation, and there are many different ways to extrapolate. The more you want to extrapolate, the the more work you're going to have to do to rule out the alternatives. When you want to extrapolate to something as big as "a new law of physics", you're going to have to rule out a lot of alternatives.

That's what "extraordinary claims require extraordinary proofs" really means. There are, at this point, a lot of far less extraordinary alternatives to "brand new physics", especially since the effect is such a tiny fraction of the input energy. The clearer you can isolate the effect, the more likely your particular extrapolation is the correct one.

Comment Re:He's absolutely delusional (Score 1) 309

I've been finding this rather disconcerting. He wants to reform a system he doesn't even begin to understand.

That's not to say the system is great; it sucks, and it could use a lot of novel ideas. But I have a hard time believing that useful ideas are going to come from somebody who just plain misunderstands the present state of things. It just doesn't fill me with confidence that the new ideas are anything other than armchair ranting. He sounds little better than the Tea Party: his ideas are less repugnant, but he seems more incensed about the imagined flaws in the system than the real ones.

Comment Re:Here we go again (Score 1) 142

Definitely no argument from me there. While computers can do better than people in that regard, people are worse than they should be, by a lot. Paying attention, and a sense of urgency, would make traffic flow a lot more smoothly. Traffic intersections in particular are valuable commodities: get through it, and expand your following distance (if you have to) on the other side. Just getting a few more cars through on this cycle of the lights will greatly improve total throughput.

Comment Re:Here we go again (Score 1) 142

Eventually, it would also be nice to have cars that can talk to each other well enough to safely form "draft trains" to conserve energy

I'm really hoping they can use it to improve traffic flow in congested situations. Human drivers require a lot more space around their cars to move safely both in front and side-to-side. They also require a lot of slack in traffic signalling: yellow lights, slow acceleration off the line, etc. Inter-car communication would greatly improve that. Combined with reclaiming traffic lanes now used for parking (since you can send your car off to park elsewhere), cities could become far more efficient places to travel through.

Just my $.02. But that will take some time.

Comment Re:Explain to me like I'm 5 (Score 5, Informative) 257

It's not intractable, but it is a challenge. (Well, not "five"; I kinda hate that expression. But "scientifically interested layman" isn't beyond reach.)

Try it this way: Quantum mechanics rules are the "real" rules of the universe: objects don't have exact positions or locations. Rather, what you get is a wave that describes the object. One way to interpret that wave is that it predicts the probability that it could be at any particular place. The total behavior of the object is the sum of those probabilities. It really is in every single place, all at once, though "more" some places than others. These waves can even cancel out. That's very much at odds with what we expect.

Here's the thing with probabilities: the more of them you add up, the more they behave like the average. That is, there's a lot of uncertainty in the roll of a 20 sided die. But you know that if you roll it a thousand times, the average is going to be very close to 10.5.

Real-world objects contain far, far, far more than a thousand objects. If you work the sum of the quantum waves for that many objects, what pops out is remarkably like plain classical physics. So, everything you see looks like ordinary physics.

But if you design your experiment carefully, you can make some of the quantummy behavior show up. The most classic one is the two-slit experiment: you restrict the particle's path to one of two places, and you get interference waves. But if you modify the experiment so that it is interacting with large-scale objects like a detector somewhere in the process, the waves vanish. (A detector is something that has large-scale changes between the particle's presence and the particle's absence.) The confusing part is that you can put the detector in places where you wouldn't expect it to have an effect, but since the particle is "everywhere", it affects it in counterintuitive waves.

Proving that for certain turns out to be tricky. The difference between "the particle really is (partly) everywhere at once" and "the particle is actually in only one place, but you can't tell" is pretty subtle. You can show it by carefully counting up "entangled particles", where the two probability waves are linked. It would be natural to think that particles were exchanging information to maintain the linkage, faster than the speed of light, but the quantum rules actually rule that out. Proving it for certain is hard, since you're talking about very tiny things and very fast speeds. We actually have been doing it for decades, but since it's so hard, there were usually loopholes. This experiment finally nails the last of them shut.

The solution to the chicken-egg problem lies in the behavior of the sums: big objects behave like you expect them to because the probability of them not doing so becomes vanishingly small. There's still some fiddly bits: that "vanishingly small" isn't quite zero and nobody exactly knows where it goes. Some say "another universe"; others (like me) just put our fingers in our ears and say "I don't know but shut up and calculate la la la".

Comment Re: Isn't this a no brainer? (Score 1) 474

I was really excited by the long-tail idea. I got less excited soon after because I happened to be dating a rock musician (seriously). Her band worked insanely hard, were incredibly talented, and put on great shows, but they were just never in the right place at the right time to make it big. Of course you have only my word for their ability, which is subjective (and biased) and anecdotal, but I'm just saying it shifted my perspective on what could be accomplished in the long tail. As abominable as the record labels are, they do one thing very well: make people famous by spending lots of money.

I still have high hopes for the long tail. It's now present-seeking season and I do it as much as I can on sites like etsy, the long tail of craft stuff. I occasionally find cool things, which I buy and evangelize by giving as gifts. But I doubt any are eking out even as much as the minimum wage by doing it. That's partly Etsy's fault, and I am gonna try out Amazon's new version of it. Amazon is good at building buzz.

Oh, one other thing that advertising has worked well for, to me. I listen to a lot of podcasts, and I've bought at least two things because they supported things I like. I would not have encountered NatureBox; I'm not going to evangelize it (it's good, not amazing), but I do kinda like having prepackaged snacks on hand for grab-and-go things. I did so at least partly because I wanted to support the podcasts. (In the case of one podcast, they have an adless, longer paid version, but I actually like the free version because the longer version can be tedious.)

I feel like I'm spreading word of mouth for advertising ;-) (The Onion, of course, did that joke better:

Comment Re: Isn't this a no brainer? (Score 1) 474

Yep... when you can get amazing buzz, it's awesome. It helps when your CEO is a gazillionaire who gets free press for building freaking *rockets*. And when your competition gives you this massive opening to build something that's just so much better than what they're putting out. (Cf Uber, who is competing against cab companies who put out a genuinely abysmal but widely-used product.)

I don't expect anybody to get similarly excited by my new OXO spatula. But it got the first brownies out of pan whole because it's somehow flexible enough in one direction to curve under the corner of the pan, while being strong enough in the other direction to support the brownie as I lift it. Kinda spooky, actually. It's definitely better than my previous spatulas, which cost about half as much (though I can spring for eight bucks on a kitchen tool), but I don't expect there to be a waiting list ;-)

Comment Re: Isn't this a no brainer? (Score 1) 474

Few things are so Insanely Great that people are going to rush out and tell their friends about it. I got a really nice new spatula the other day, but I'm not evangelizing for the spatula company. Even if they'd given me a free spatula to encourage me to proselytize, and I managed to encourage a couple of friends to buy one, it's unlikely that they're going to rave about a spatula they bought so much that it's going to push out to six degrees of separation.

This kind of marketing works great when it works, for really exciting consumer gadgets and apps and political candidates. But a lot of the stuff you need to buy is just, ya know, stuff.

Comment Re: Isn't this a no brainer? (Score 1) 474

I wish that were true, but sadly it isn't. To get word-of-mouth, you need somebody to have your product. Somebody has to buy it. You could give it away, but that's just another form of advertising, and it's not clear how effective it is.

I run a theater troupe. Most people don't even know when I run performances. I've built up a following over a decade and a half, partly by word of mouth. But it's very slow, and if I had big budgets (paying actors, big sets, larger theaters) I'd need more people, and I'd need them fast. I need to tell them that I exist, and word of mouth alone won't cut it. I need the people who aren't directly connected to me.

Word of mouth is great, the purest form of marketing you can get. But it doesn't work in every case. Few things are so insanely good (or insanely grabby) that they go viral.

Comment Re:Isn't this a no brainer? (Score 1) 474

They'd definitely like for you to buy now, but they will mostly have to settle for less. You may not need a new job via Monster or a subscription to New Scientist right now (current ads on Slashdot), but they may gradually sneak up on you. If you need a job, where are you going to go? The most popular site (though network effects contribute to that). How do you know what's popular? The one you see all the time.

What do you give your nerdy buddy for Christmas? Maybe a magazine subscription. It might not have occurred to you that you could even do that, but if the two thoughts happen to be in your mind at the same time, they get some business.

You're absolutely right that the ads that demand your attention right now are the ones that have to go. Advertisers think they want them, because those can be tracked to see how successful they are. It's more annoying for New Scientist to say, "Well, we put out an ad, and over the next three months subscriptions went up 3%. Due to the ads? The economy? Random chance? A good article?" Or worse: "Subscriptions went down 1%. But maybe they'd have gone down 1.5% if we hadn't put up an ad..."

The purveyors of "polite" ads should probably demand that web sites not show any other kind. That's the kind that gives advertising a bad name, and makes people rush out to get AdBlock, which kills all ads. It's possible that getting the advertisers together to set standards could result in more money for them. But I'm not holding my breath.

Comment Re:Isn't this a no brainer? (Score 1) 474

Mostly agreed, though what I had in mind for "popular song" was those breakout hits ("Gangnam Style", "Call Me Maybe") that just stick in the mind for no obvious reason. (Though I did hear an interesting podcast a few weeks back that analyzed "Cal Me Maybe" and its follow-up, showing where the formula was successfully repeated and where it failed to recapture the magic.)

Those are rare, though so are the equivalent ads. And even then, only on TV. I don't think anybody's sidebar ad has ever achieved anything close, though occasionally one will be so aggravating that it becomes a meme, e.g. "Punch the Monkey".

Comment Re:Isn't this a no brainer? (Score 4, Interesting) 474

Most ads aren't aimed at getting you to click on them and make a purchase. The goal of an ad is to put a concept in your mind. The purest form of this are the political yard signs. Nobody thinks, "I saw 90 signs for Bob and 110 signs for Mike; therefore I will vote for Mike." But people do respond to pressures like that; it creates biases and impressions that they don't even realize.

Even when you do your research, you are influenced by these. Most of the time, your research is going to be inconclusive. There isn't any "best chair"; at best, it's a matter of personal taste. Most products, from canned peaches to computers, will end up having similar specifications, but you'll have a preference because you like the flavor of this brand or the you had a good experience with that computer in the past.

Advertising helps put those ideas in your head. Just seeing it in the ads will give you a positive feeling toward the brand, if the advertising is well done. A lot of advertising is poorly done, of course, but a well-done ad can influence preferences in very subtle ways. That subtlety means it's aggravatingly hard to tell which ads work and which ads won't, but advertising continues to exist for a reason: it steers consumer preferences during the phase where they don't know what they want and end up trusting their instincts. Which applies to more purchasing decisions than most people realize.

Stupid advertisers want ads that they measure working by clicks, so they optimize the ads to attract clicks, but that doesn't drive purchasing. The best ads are the ones that consumers don't even realize they've seen, but just develop a cumulative effect of exposure. That's hard to do, and requires a lot of time, money, and effort to get right. Even then it's a crapshoot, like trying to write a popular song. But in the end, there's a market for so many chairs and so many peaches and so many computers, and advertising can steer enough purchases towards yours and away from somebody else's equivalent one in a way that merely improving the specifications can't do.

No man is an island if he's on at least one mailing list.