Follow Slashdot blog updates by subscribing to our blog RSS feed


Forgot your password?

Comment Re:Stallman's a Brilliant Engineer (Score 2) 649

From the summary: "His solution is elegant ...his measure would create a required minimum 'Return on Investment' scale that corporations need to follow to be viable, and these types of metrics are very industry specific. Another issue is that many large corporations stay in business because they don't take unnecessary risk..."

Which is to say, the proposal requires layers upon layers of kludgy patch-ups to make it even remotely plausible, which will make it highly gamable in ways that mere technologists will never figure out, but the sociopaths who run companies will be all over.

There are some fairly well-known, well-tested ways of dealing with this, Glass-Stegal being the most obvious one. People who are attempting to create new, untested solutions are missing the point.

Comment Re:Recognition Test (Score 1) 417

If an average american from 1910 were suddenly transported to 1960, things would be unrecognizable -- there were so many truly groundbreaking changes. Home electric power, radio, television, refrigerators (and the supermarkets and foods they allowed), automobiles, antibiotics, etc. had all gone from being unknowns to being commonplace in the intervening period. (They may have existed in 1910, but they weren't developed to the point of commercialization.)

Correct. I've been pointing this out for years, and the responses have been pretty much what we've seen to this article:

1) Everything is "incremental" if you equivocate on the meaning of "incremental", somethings implying "builds on past inventions" and sometimes implying "only adds slightly to pre-existing capabilities"

2) Cell phones and the Internets!

Network technology has certainly been a socially important feature of the past 50 years, but my grandmother was born in 1886 and by the time she was 50 (1936) the following things that had been impossible when she was born were commonplace:

* heavier-than-air powered flight...
* ...for commercial travel
* ...and warfare

* moving pictures...
* ...with sound
* ...and in colour

* antibiotics

* electric appliances for the home
* radio communication
* mass produced automobiles for the common person

And so on.

In the first 50 years of my life the inventions that have changed the face of the world comparably are:

* Cell phones
* the Internet

Not small things, but there are only two of them. Three if you count "ubiquitous computing" as separate from the 'Net.

By a simple count alone the pace of major, socially-changing innovation is a factor of three lower than it was a hundred years ago.

Comment Re:bound state QED and QCD (Score 1) 171

This is my suspicion as well, specifically about the charge distribution. I think a 4% effect could easily be explained using a model with distributed charge.

It has been my experience that posts where the poster annouces their "suspicions" are almost alway gibberish, displaying a profound ignorance of the most basic elements of the subject they have suspicions about. It really is a useful litmus test, to the extent that I think /code should be modified to automatically down-mod any post that contains phrases like "I suspect that" and "my suspicion".

It's not that there's anything wrong with expressing doubts. It's that this specific way of putting it seems to be used almost exclusively by people who don't know the first thing about what they are talking about.

With regard to the question: the structure function of the proton, which describes its charge distribution, is precisely what these experiments are all about. The "radius" of the proton is in fact a parameter of the structure function, and the curious aspect of these results is that the discrepancy cannot be "easily" explained by naive adjustments to it.

It's worth noting, however, that the structure function has more-or-less exponential tails in most models, and the muon orbitals have much smaller characterisitc radii than the electron orbitals, so they are much more sensitive to the precise manner in which the structure function falls off with radius.

For comparison: consider low Earth orbit satellites. If the Earth's atmosphere falls off just a little bit more slowly than expected, it will affect the orbits of such satellites much more than ones even a few hundred kilometers higher up.

It is very likely that there is some extremely subtle effect that is being neglected or approximated inaccurately in our calculations of the structure function that is the root cause of the discrepancy seen here, but whatever it is, you can be sure that the explanation isn't "easy", regardless of what suspicions you may have.

Comment Re:André Gorz (Score 3, Informative) 586

Rises in productivity due to automation are incompatible with a culture that values 'work' on a moral basis, and associates it with a persons identity.

This is the critical thing, in much the same way that decoupling wealth and power from land ownership during the Industrial Revolution was incompatible with a culture that valued landed estates on a moral basis, and associated them with a person's identity (at least for the gentry, who were after all the only people who counted as "people", back in the day.)

It took something closer to centuries than decades for a relatively small and educated class to come to terms with that (my Scotish friends tell me England is still struggling with it.)

Today, we have a system of distribution of benefits from social producitivity [*] that depends on "work", while automation is rapidly eliminating jobs while maintaining productivity (and therefore profits for owners.)

It is of course completely indeterminate how this is going to end, but we can be pretty sure that a hundred years from now the status quo of the past century in which paid corporate employment has been the common basis for the distribution of wealth, won't be the norm, and more than the leasehold farming and villiage life that was the norm in England in 1750 much resembled the average English life in 1850 (Male Employment in Agriculture/Industry = 1760: 52.8%/23.8%; 1840: 28.6%/47.3%).

[*] if you don't think social goods like the rule of law in general and the Companies Act in particular are absolutely necessary, though admittedly not sufficient, for "private" corporations to exist, much less thrive, you might be a libertarian lunatic

Comment Re:Now THERE's a reversal. (Score 1) 251

Actually, one of the biggest recent meta-studies to come out on climate science showed that warming over the last 20 years has been very close to the average consensus forecasts over the last 20 years.

Then why does every single story on new work in this area appear beneath a headline announcing how badly previous work underestimated the effects?

The answer, of course, is that the public debate about AGW has nothing to do with the science of AGW.

The sole policy prescription the pro-AGW side have is, "Reduce CO2 emissions by any means necessary except investment in nuclear power, even if a back-of-the-envelope calculation shows that it is infeasible to the point of causing hundreds of millions of premature deaths."

The anti-AGW side say, "Yew kin take ma coal when yew pry it outta ma cold dead hands." As near as I can tell, they have a deep emotional connection to soot.

Neither side is interested in rational policy responses to the problem (Not quite true: there are a few lonely voices on the pro-AGW side that have attempted to have a science-based policy debate, but they are marginalized by the mainstream IPCC process and the sensationalist, anti-scientific press. Just ask Christopher Landsea.)

Comment Re:What about the other way around? (Score 1) 332

How about whether high IQ folks are more likely to smoke pot or dumb ones?

The data suggest lower IQ males are more likely to start smoking pot: (and kudos to the guy who responded to that comment by suggesting pot smoking turns women into men!)

Comment Re:And .... (Score 1) 332

And most people will focus on the person in the above stories that confirm their bias.

I'd love to see these sorts of study first released with the blind terminology, so the claim would be, "X causes Y". Blind the demographic variables too. Let the reviewer evaluate THAT and see how much of the actual review is based on the quality of the data and the analysis rather than the bias of the reviewers.

This study was obviously pretty questionable, even on a cursory look. As I said at the time of the original article (

"A randomized controlled trial is by far the best means of proving causality, but a strong dose-response curve is a good secondary indicator. In this case, the data don't seem to support the contention of the abstract very well....

There are a couple of striking things: the percentage of males jumps markedly as the regularity of cannabis use goes up, and the initial IQ drops. So this study shows that young men with slighlty lower than average IQ are more likely to engage in regular cannabis use, and this may or may not result in a further decrease in their IQ over time."

This was apparent after about five minutes of looking at the paper (admittedly, I'm reasonably experienced at this, but you'd expect the people who reviewed the paper to be as well.) That it got passed by reviewers and published in PNAS is a travesty.

Comment Re:Noise (Score 1) 457

Cube farms are cheaper than offices,
Open office layouts are cheaper than cubes.

The difference in developer productivity is at least a factor of two between cubes and offices. It may be as high as a factor of ten in some cases. So you waste a lot of money on salary--which is the dominant expense in development shops by a very large factor--by having cubes rather than offices.

So the answer is clearly anything but money. If it was just about money (the least money for a given work product) small teams of developers in good conditions would be strongly favoured.

Businesses are not run to maximize productivity, they are run to maximize manager's feelings of power and control.

Comment Re:Could they redirect only a certain hotness? (Score 4, Informative) 82


They first have to select specific wavelengths and then--it sounds like--frequency-shift them.

To call this "heat" is a deliberately misleading statement designed to elicit precisely the question you are asking, as that will attract much more of our most limited resource--attention--to this otherwise interesting but essentially esoteric work.

"Heat" in ordinary parlance is constituted by vibrational modes that obey the principle of equipartition, which this "heat" manifestly does not.

Comment Re:Good Advice (Score 1) 316

Any one who says the data doesn't support the vaccine is an idiot or a liar. That data is pretty well documented.

Nothing like adding a nice little anti-scientific ad hominem to your day.

The data are pretty well restricted to deaths from "flu and pneumonia", so no one actually has a clue what the death rate from influenza is, in the normal course of events. We have slightly better data on outbreak years, but from an overall public health perspective you'd think the CDC would actually have a reasonably good estimate of the total number of deaths prevented each year before recommending a universal vaccine that has about a 1 in a milliion rate of serious neurological complications (much worse in some years... now convince me that next year isn't going to be one of those years...)

There is an argument to be made for flu vaccines, but the case is no where near as open-and-shut as it is for smallpox and the like, and fleeing the realm of science to hurl insults is not a good way to resolve the issue.

Comment Re:24,000 Americans die each year (Score 2) 316

Influenza and Pneumonia: 53,692

Right, so how many died of the flu?

0? 53,692?

This is the dirty little open secret of the anti-flu business: no one knows how many people die of the flu. The number is certainly not zero, but it is equally certainly not the full tally of "flu plus other things that present similar symptoms that are not the flu".

So the question remains: "Why doesn't the CDC keep tabs on overall deaths from the flu?" and the answer is: "It is not economic to do a proper diagnosis of every fatality from 'flu and penumonia'."

The real question is: why so much hype around flu shots, whose effectiveness varies from year to year but is never over 75% (and sometimes is considerably less) when every now and then a problematic batch produces a risk of about 1 in a million of serious neurological consequnces? The risk of death from "flu and pneumonia" is about 1 in a million amongst healthy adults, so the risk of death from flu is lower than that.

There is an argument to be made for flu shots amongst health care workers, the elderly, and possibly the very young (risk goes up dramatically below the age of 5), but the case is much harder to make--except possibly for herd immunity--amongst healthy adults.

When I worked in a hospital I got my flu shot. Now that I don't, I don't.

Comment Re:One question (Score 3, Informative) 453

If divorce rates are increasing (which I suspect they are),

There's this amazing thing called a "search engine" that you can type "queries" into. If you'd spent two minutes doing that you'd have found that what you "suspect" is wrong. Divorce rates have been flat for decades, and may even be decreasing a bit (can be tricky to tell because divorce rates drop in poor economies.)

Comment Re:The article itself comes with some misconceptio (Score 1) 265

Similarly, if we found out precisely how the brain gives rise to consciousness, that wouldn't mean thinking per se didn't exist.

The "phenomena are not real" crowd have two basic moves, which contradict each other.

The first is, "We can reduce phenomenon X to cause Y, therefore phenomenon X doesn't 'really' exist" (because they for some reason believe that only their atomic terms to which they want to reduce everything 'really' exist--no explanation for this surreal prejudice is ever given)

The second is, "We cannot reduce phenomenon X to cause Y, therefore phenomenon X doesn't 'really' exist" (because they have assumed ab initio that only their atomic terms 'really' exist, and therefore anything that cannot be reduced to them must not 'really' exist.)

Self-contradictory 'arguments' against free will, which attempt to get people to use their free will to voluntarily change their minds regarding the obviously existing phenomenon of free will are of the latter type.

The utility of Kuhn's analysis is that he at least vaguely recognized that the sciences have two basic moves available: to make arguments reducing phenomena to known causes that flow from existing atomic terms; and to introduce new atomic terms (wave functions, tectonic plates, genes, subconsious motivations, germs.) He vastly over-stated the differences between these two moves in day-to-day science. They both happen all the time, and the distiction between "normal" and "revolutionary" science is fairly blury in practice.

Comment Re:There _is_ a shortage (Score 1) 232

If you study, on the other hand, semiconductor physics, friction, or material physics you'll find half a dozen offers for well paid positions in industry research labs in no time.

Don't. Make. Me. Laugh.

"Semiconductor physics" is a huge field, and "materials science" is even broader, and only a few tiny niches are hot at any moment, and if you don't luck into one of those you may as well have spent a few years putting a more precise limit on a particular branch of the decay of a non-existent particle (which is what I did for my PhD.)

When I was a student there were two particularly big things in semiconductor physics: molecular beam epitaxi and gallium arsenide. By the time I graduated both were passe', although I knew one student who just squeaked in under the wire and got a good job on the basis of his graduate work in MBE.

Tribology (friction) certainly has some appeal currently, particularly in medical devices and implants, but that's today. I would never advise a student to go into a particular field simply based on the current prospects for jobs, because I know too well that the jobs may not be there in three or five (or ten) years.

From another perspective, there is no shortage of scientists: there is a shortage of scientists who fit the unbelievably narrow specificatoins that hiring managers put on open positions (the so-called "purple squirrel" phenomenon.)

On top of this, there is this relentless chorus of educators and policy-makers continually screaming that we need more educated people in general and more people with science and engineering degrees in particular, preferably at the PhD level--because a Masters is pretty much an admission of failure in the sciences--while the hard reality is that threads like this are full of two pieces of generally sound advice for people with PhDs looking for jobs outside of the hundred-to-one odds of tenure-track appointments:

1) apply to universities in support capacities, where smart people are valued even if their degree isn't in the immediate field of application

2) lie about having a PhD.

Those are good pieces of advice unless you present yourself very, very carefully. I've been fortunate to be able to leaverage my PhD into some amazingly cool stuff and also fortunate that I've got the right attitude and skills to run my own business, which I did for about a decade. But I broke into the commerical software market during the dot-com boom where they'd hire a dog if had some coding experience. Unless you've had comparable good fortune, the thing that the OECD and others are all telling you should do--get a PhD so you can get a good job--is actually a significant impediment to getting a good job.

Almost everyone here knows this, but almost no one going into a PhD program realizes it because of the sytematic hype-machine promoting higher education as the road to riches.

Slashdot Top Deals

Just go with the flow control, roll with the crunches, and, when you get a prompt, type like hell.