Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 



Forgot your password?
typodupeerror
×

Comment Re:Freedom (Score 1) 250

Monopoly on what?
Ebooks? Books? Written content? Information? Entertainment? Leisure provider?

Anything can be defined as a monopoly if it is scoped narrowly enough. Apple has a monopoly on iphones. Yet it is under great competitive pressure from substitutes.
Amazon is in the same boat (plenty of competitive pressure) and consumers don't seem to be fleeing away from the supposed monopolistic abuse (higher prices and limiting supplies, according to textbooks).

If you are interested in the history of antitrust, see Dominick Armentano for a critical review of efficacy of government improving on market competition.

Comment Re:Freedom (Score 1) 250

I agree. Not only is it easier than ever to publish, Kindle Unlimited is just one more option. So if you are a noticeable author, you can still publish on hardcover/paperback/Kindle and escape the fixed-pie game of Kindle Unlimited.

I don't know how many people buy one book every month, but Kindle Unlimited is probably more revenue from an average person. If anything it sounds dangerous for readers, more than authors, as it could be like paying for the gym you don't end up visiting often...

Comment Bayesian senses (Score 1) 244

The above explanations for mondegreens seem very consistent with the recent understanding of neuroscience.
All perception of the world requires inference, as the signals coming into the brain are ambiguous, conflicting and noisy.
For instance, the brain tries to reconstruct a stable 3 dimensional perception of the world from constantly moving and imperfect 2 dimensional projections.

An increasing number of studies show that the low-level processing in the brain is surprisingly similar to Bayesian inference. In particular, it demonstrably relies on priors learnt from the environment (for example, vertical lines should be interpreted as corresponding to longer distances than horizontal ones) and by fusing sources of information (for example, the ambiguous local motion detected in one part of the image is reconciled with other ambiguous local motions into a perceived motion of objects).

For anyone interested, I'd recommend some material by Stanislas Dehaene.

Comment Cost? (Score 1) 103

Look, this great shiny technology. Oops, it's uneconomical...
Costs are an important question as this story happens quite often (mag trains anyone?).
For space projects the cost may not matter, since it's paid for by taxpayers. But for applications for the rest of us making this affordable is as important, if not more.

Comment Already addressed (Score 1) 652

Reads the articles:
"Across the board, we need solutions that don’t require subsidies or government regulations that penalize fossil fuel usage. Of course, anything that makes fossil fuels more expensive, whether it’s pollution limits or an outright tax on carbon emissions, helps competing energy technologies locally. But industry can simply move manufacturing (and emissions) somewhere else. So rather than depend on politicians’ high ideals to drive change, it’s a safer bet to rely on businesses’ self interest: in other words, the bottom line."

Comment Re: liability, and necessity of randomized trials (Score 1) 193

The issue of liability can be solved with consent, just as it is in the randomized trial. The number of patients to receive the treatment doesn't seem to affect liability, nor does the cost of the treatment.
Also showing product safety is generally much easier than showing effectiveness, and different people (especially those with a terminal condition) have different tolerances for risk.

The question your raise is whether you learn something from non-randomized trials (which I agree are the gold standard procedure). If there is a very lethal disease (that we're able to test for but not treat), and the people taking the treatment (in various locations, conditions, etc) fare significantly better, while the people who don't continue to die, can you draw any conclusions as to the effectiveness of the treatment? I think you can make inferences, although with lesser confidence.
The question that follows from that is, if you are a doctor or producer and have such a treatment, is it ethical or desirable for you to withhold it? That seems selfish to me (putting your preference for experiments over my preferences as a patient).
Is it ethical for prohibit the doctor or producer from delivering to consenting patients? To me, that seems paternalistic and unhelpful (not to mention of questionable political authority). It would seem much better to help inform the patients and give them support for their own choices.

As I suggested in another comment you could imagine a two prong approach: customers are allowed to get a safe but unproven product, or participate in experiments with some compensation/incentives. So you could still conduct experiments on people who accept that bundle (risk and compensation including the satisfaction of knowing they are helping science).
Whether people SHOULD participate in those experiments and therefore those experiments SHOULD continue to exist doesn't seem a categorical/universal question to me. It seems a question of preferences (demand). Assuming enough people demand the extra confidence and rigorousness, and enough people are willing to take the risk of receiving a placebo, then they will exist. Otherwise you are imposing your own will and preferences on people, which hardly seems civilized or peaceful.

Comment Re:So, of course, it goes without saying (Score 1) 193

You seem to be forgetting that some people in randomized trials are receiving the drug. The people who participate in the trial volunteer and are indeed bearing risks. The doctors delivering the drug would be no more liable if they give the other half the treatment too, or to even more people that want to join the experiment.
The question raised here is whether it is ethical to withhold a safe and potentially effective treatment from suffering patients. Of course, the patients would still have to consent, you wouldn't force the unproven drug on them.

Comment Withholding potential treatment (Score 2) 193

Randomized trials are essential withholding potential treatment from some suffering patients, to satisfy some of the experimenters' goals.
So why not allow the experimental product to be administered outside of the trial, having only passed safety standards? That can provide solid evidence for a large category of illnesses (those that people are not known to recover from spontaneously).

Then, if some folks want more rigorous evidence of efficacy, they are welcome to find patients who accept to participate in the randomized trials. Maybe by compensating them in some way for the risk they are taking? (It's possible that nobody would accept such a deal, so what?)
Or maybe by modifying the protocol so that they get the treatment for sure if their condition worsens?

Comment Not enough (Score 1) 549

Specialization and dvision of labor are essential factors of technological development. You need a large population (demand) to justify some investments such as education in advanced fields, advanced machinery for high productivity and pursuing economies of scale.
I doubt that a population of a million could support the level of technology required for life on Mars. Working knowledge would be lost (knowledge in a book in little use compared to knowledge in someone's brain).
There have been historical instances of such regressions where groups were cut off from the larger population and regressed technologically as their market shrunk.
As Adam Smith put it, "The division of labor is limited by the extent of the market".

Comment Nirvana fallacy of neoclassical economics (Score 1) 218

Fortunately for us, competition is pretty effective even without perfect information (or perfect substitutes or large number of competitors for that matter). The neoclassical model of perfect competition does little to understand the world, because of its unrealistic assumptions (including that government can fix those supposed problems without being subject to similar ones).
It is true that information is easier to exchange than before, but such aggregation was always possible. That's the whole purpose of brands, reputation, certification and insurance. Licensing (restricting entry) is never necessary to protect consumers.

I don't know much about the history of this specific industry, but the same claims have been made in many industries (lawyers, doctors, plumbers, architects), if not most. The various historical analyses that I read on those cases show that cartelization interests were a stronger factor than consumer protection.

Comment Nonesense (Score 1) 418

Putting two words together does an argument make.
Pollution is an externality, whereby bystanders are harmed. Technically, it's a form of trespass on those people's property. But advertisement on the internet is no such thing.
YouTube does not belong to you, it belongs to Google. If you choose to use that service, then you don't get to call the costs or downsides "pollution". Similarly, if you don't like the music that is played in a restaurant, or you don't like their drink minimum rule, you do not get to call those parts of the deal "pollution". Either you go there or you don't.

"When words lose their meaning, people lose their freedom." -- Confucius

Slashdot Top Deals

There are two ways to write error-free programs; only the third one works.

Working...