Become a fan of Slashdot on Facebook

 



Forgot your password?
typodupeerror
×

Comment Re:Very informative article (Score 3, Insightful) 71

2) Observe that technological progress seems to be accelerating. So, the reach of any predictive models we have today will be even shorter when used tomorrow.

I think LeCun covers this quite well when he quotes, "the first part of a sigmoid looks a lot like an exponential." There's nothing that says the acceleration of technological progress has to continue as it has.

Comment Re:disclosure (Score 1) 448

Yes, but they get millions to conduct research. I doubt he took that $1.2 million home.

Hard to say, actually. My understanding is that his appointment was a "soft money" position, which means that he would have been entirely dependent on outside grants for all of his funding--including the salaries for himself and any staff or trainees. I wouldn't be surprised if $120K per year, less administrative overhead, doesn't even fully cover his own salary.

Comment Re:Yes, it's a conflict of interest. (Score 5, Informative) 448

Dr. Soon may even truly believe his science is valid, but the funding he receives creates a lopsided megaphone which unfairly skews the perception of the debate.

Having a conflict of interest is understandable; hiding a conflict of interest is problematic.

By the same token, all scientists who receive funding from the pharmaceutical industry or groups they influence, should be barred from publishing papers on vaccine safety.

Scientists who receive funding from, for example, the pharmaceutical industry are expected to fully and explicitly disclose potentially conflicting interests--and by golly, they do. It's taken quite seriously, actually. If you look at any article in a respectable medical journal today, you'll find a section of the manuscript that's explicitly headed with Conflicting interests: or something synonymous. It will appear on every article, even on the ones where it's followed by "None declared" or the like, just so that it's clear that the journal asked for and got an on-the-record response from the article's authors. It doesn't remove the potential bias associated with outside funding, but it at least makes the potential for bias transparent.

Lying about competing interests - even through omission - is looked on very poorly by serious, credible medical researchers. Interestingly, one of the many, many types of misconduct engaged in by Andrew Wakefield was his failure to disclose significant financial interests when he published his (now-retracted and thoroughly discredited) Lancet paper suggesting a link between the MMR vaccine and autism. (Wakefield wasn't receiving pharmaceutical money, of course; he collected almost half a million pounds from lawyers involved in an anti-MMR lawsuit.)

And while the practice of mandatory disclosure started with the medical journals, the expectation has gradually bled across into other fields as well, particularly among top-tier journals.

Comment Re:Why did he tape off his light switches? (Score 1) 248

If he's such a handy guy, he could've just wirenutted the wires in the boxes and put blank plates over them. simple and neat looking!

Well, "handy" is in the eye of the beholder. He's really just "software handy", which is a different thing, for people who don't like to manipulate physical objects. It's kind of like the difference between a real engineer and a "software engineer".

To be fair, it was probably intended to be a very temporary installation and test. (Though it wasn't fair for him to complain about problems caused by his own shortcuts, I suppose.) And he did try to install a switch replacement at one point, with unfortunate results that weren't exclusively his fault.

Comment Re:A tax on stupidity (Score 2) 480

Maybe, but... I'm trying to think of the things I could buy (or do) with $500 million that I couldn't also buy (or do) with "only" $40 million. I'm not coming up with a whole lot. Moon base, maybe?

I fear you may be grossly underestimating the cost of a Moon base. I mean, the Chinese Chang'e 2 lunar orbiter mission cost around $150 million; I can't find a price tag for their unmanned Chang'e 3 rover, but I can't imagine it would be cheaper. Forget putting a man on the Moon--either to live, or especially if you're concerned about getting him back again....

Back in the mid-2000s when you could buy a seat on a Soyuz launch for a week or two on the International Space Station, the quoted prices were between $20 and $40 million. So I guess you could do that with your $40 million win, but only if you were comfortable with (a) going by yourself, and (b) not having much left over afterward. With $500 million, you could afford to live very comfortably and make a return visit to the ISS every few years, just on the investment income.

Comment Re:Different market segments (Score 2) 422

Quite true. But keep in mind that this might not be free of cost (or effects) for those of us at the middle-to-high end. There's probably a bunch of "infrastructure" and overhead-type costs that are currently shared across different market segments.

For instance, Canon uses the same DIGIC signal and image processing chips across a bunch of different models: professional and consumer DSLRs as well as whole lines of point-and-shoots. If big pieces of the point-and-shoot market evaporate, then the cost of developing those chips and sustaining shorter production runs for them has to be carried by the DSLR market alone.

Aside from the direct economies of scale, there are going to be some more subtle business and technical reasons why losing a market segment will hurt the whole company. The fruits of R&D investment typically show up first at the high-end SLR market, and then trickle down through the prosumer and consumer SLRs, then make their way from the high to the low end of the point-and-shoot cameras. A smaller market overall means either less R&D or more expensive SLRs.

Heck, just having a wider product line means fewer boom-and-bust cycles when the company is between new models (with the accompanying bursts of new sales and marketing buzz.)

Comment Re:Moderation (Score 5, Funny) 200

"Is it true that you smoke eight to ten cigars a day?"
"That's true."
"Is it true that you drink five martinis a day?"
"That's true."
"Is it true that you still surround yourself with beautiful young women?"
"That's true."
"What does your doctor say about all of this?"
"My doctor is dead."

- An interview with George Burns

Comment Re:Last 2 planes? (Score 1) 293

The oldest flying 747 is also the fifth 747 produced. It was delivered in August 1970 and is still flying today.

This is true. As the linked article notes, a remarkable number of early 747s are still in service in Iran, either with the Iranian Air Force or as part of the fleet of Iran Air. These airframes date to before the 1979 Iranian Revolution. I didn't really want to open that can of worms, though--given the diplomatic situation between Iran and the United States, one wonders at the level of support that Boeing would feel it had to provide (or even would be allowed to provide, as a U.S. company).

I would wonder if there were any parallels to the situation in, say, Cuba, with its large population of Batista-era (pre-Castro, pre-Communist revolution) U.S.-built "classic" automobiles. Though you'll still see them on the roads of Havana, I wouldn't expect Ford to still have parts for, say, the 1957 Fairlane.

Comment Re:Last 2 planes? (Score 4, Informative) 293

So...$1.65 billion to buy the planes from Boeing, and how many millions per year to have Boeing keep a tooling line up for spare parts?

Since airlines were still ordering new 747-8s (the platform on which the new Air Force One(s) would be built) in 2014 - and might still continue to do so - this isn't exactly an obsolete aircraft. I mean, the first 747-8s weren't delivered to customers until 2011. There are still-flying 747-variant fuselages in commercial (passenger and freight) service that have been in the air since the late 1970s and early 1980s. Based on that history, it seems likely that Boeing will need to support its commercial customers through to at least 2045 or so.

Comment Re:Of course you can take things too far... (Score 1) 79

but I've always thought it was strange that people will spend days learning about, debugging, and fixing their car or computer...

Who are these "people"? Sure, there are some who will do this - and this being Slashdot, we've probably got a population enriched in them - but if you take a random person with a car or computer problem, they will just ask (respectively) their local mechanic or teenager--in exactly the same way that many people approach their doctors.

Yes, there exist people who will try to troubleshoot their cars, computers, or their own bodies. And there exist subsets within those groups who can are capable of doing so in a way that makes things better rather than worse. Most people aren't comfortable going beyond the most basic diagnostic steps in any of those three spheres, however. Putting gas in the car is their limit; checking the oil and topping up the washer fluid is something that happens at the dealership, and borrowing an OBD-II reader to interpret a Check Engine light is in the realm of black magic.

Worse, there are also people who spend good money on magic grounding straps for their cars, insist on clicking on VIRUS WARNING links in spam emails, and who actually believe stuff that homeopathic practitioners tell them.

Slashdot Top Deals

Garbage In -- Gospel Out.

Working...