Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[wierd]

I was at work at the time, and Google News decided to give me a faux news article instead of something more reputable. My phone was trying to die terribly, so I ran with it.  The news was that there is a suspiciously interesting anomaly in photon production in the detectors. Other than the bias inherent in the faux news network, they didnt really screw up the story.

I really wish I could tell google to not give me faux news sourced articles, and give more weight to more reputable sources, but then Faux news would complain about being marginalized for being a filthy rag press, and lawyers would probably get involved.

[Starver]

I really wish I could tell google to not give me faux news sourced articles, and give more weight to more reputable sources, but then Faux news would complain about being marginalized for being a filthy rag press, and lawyers would probably get involved.

Not that I've tried it, but can't one add something like "-site:foxnews.com" to the appropriate search?  My GoogleFu is a little rusty, but the mechanisms are probably in place, if not linked together.

ETA (ETA2: too late for the immediate reply, it seems)...as for the announcement, science is the most interesting when things appear to be different from established thought (either the established thought or the strange result being shown to be erroneous).  No matter how ancient or recent the 'established thought' is.  Science!  That's how it rolls, dudes...  Enjoy the trip!

[wierd]

I may have to see if that works with a general  aggregation for the daily news. I would like to see all the results, especially those in the science section, NOT be sourced from known terrible sources.

[martinuzz]

American researchers published the results of an extensive study into indentical twins in PNAS magazine (sorry didn't bother to find a link because I'm pretty sure I'd just link to a paywall, as with all prominent scientific publishers:

Smoking weed during adolescence does not hurt IQ, as some earlier studies suggested.

The researchers explain those other studies by saying that a statistical error has been made; people from lower income classes have a lower average IQ, and on avervage, smoke more weed. So even though this could seem to point at a correlation between weed and lower IQ, there is no causality.

The twin studies effectively debunked that.
The researchers followed 2 groups totalling over 3000 indentical twins. Twins per definition have the same social and genetic background, so any difference in IQ cannot stem from those factors.
The IQ of the participants was measured twice: once at age 10, and then again at age 18. The results showed no significant difference in IQ between a twin that smoked weed during adolescence, and one that did not.

The researchers do warn however that their report should not be used to stimulate adolescents to smoke weed; Even though it will not lower IQ, it will not raise it either, and going to school stoned does make it harder to pay attention and remember the lessons.

[Starver]

Twins per definition have the same social and genetic background, so any difference in IQ cannot stem from those factors.

The most interesting Twin Studies has 'twins separated at birth' studied so that there was an identical genetic background but a different social background, as they're brought up in different families.

But it's harder and harder to carry these out, these days, given we don't tend to do this quite so flippantly separately adopt twins; even if we might now at least more easily track/re-acquire the pairs, rather than them getting lost in the system until they both end up retiring to the same Florida retirement home in 60 years time and noticing they tend to buy the same wacky type of tie and, come to think of it, look so much alike that they've already been unintentionally confusing friends and family over many years, and that's why they were both temporarily sectioned in the '80s, ironically in different wings of the same hospital...

Darn the new social conscience, eh?  Look what information we're losing!

The results showed no significant difference in IQ between a twin that smoked weed during adolescence, and one that did not.

In a same-family situation, one wonders whether a cause for the twin-that-smoked-weed was discovered (or else, a reason why the twin-that-did-not-smoke-weed did not).

Are you more likely to regress into the arms of the spliff if everyone, especially your twin-sibling, keeps reminding you that you're the younger twin?  Or, of the two twins, you're the one with the silliest name (your brother is Bertone, and you're Berttwo).

Science demands to know!  No, scratch that, I demand to know...  Science (and my long-lost secret twin brother who is almost definitely a billionaire with his own metal suit, or so I extrapolate from all the available evidence) probably already knows but they refuse to tell me!

[martinuzz]

In other news, British scientists claim to have proven that insect brains are, surprisingly, large enough to see in 3D.
To accomplish this, the gave a praying mantis small 3D glasses, and tested it's reaction and accuracy versus moving dots in 2D and 3D animations.
The insects preferred striking at the 3D animations, and almost completely disregarded the 2D animations.
Animal brains create 3D vision by comparing the slightly different images from the right and left eye. Up until now, scientists doubted that insect brains would be large enough to accomplish this.

Instead of red-green 3D glasses, the experiment required the scientists to find a suitable alternative, since insects are incapable of seeing the colour red.
The glasses used instead use green-blue, which after much testing seemed to work best for the mantis.

Spoiler (click to show/hide)

[Gentlefish]

I think with twins, they share slightly different life experiences. They both have the capacity to do X, say, but only one experiences catalyst Y that makes them do it.

[Telgin]

In other news, British scientists claim to have proven that insect brains are, surprisingly, large enough to see in 3D.
To accomplish this, the gave a praying mantis small 3D glasses, and tested it's reaction and accuracy versus moving dots in 2D and 3D animations.
The insects preferred striking at the 3D animations, and almost completely disregarded the 2D animations.
Animal brains create 3D vision by comparing the slightly different images from the right and left eye. Up until now, scientists doubted that insect brains would be large enough to accomplish this.

Instead of red-green 3D glasses, the experiment required the scientists to find a suitable alternative, since insects are incapable of seeing the colour red.
The glasses used instead use green-blue, which after much testing seemed to work best for the mantis.

Spoiler (click to show/hide)

That's pretty cool, but I'm a little surprised that that was even in doubt.  The math behind 3D point triangulation is fairly simple, so I guess the question was if they had enough brain power to do object correlation between two images.  I'm guessing that the brain works on highly downsampled information for that, so it probably doesn't require all that much power to do, comparatively.

[Levi]

A google AI has beat a professional at playing Go.

This is kind of a big deal, as plenty of people have said this would never happen.  In March the AI will challenge Lee Sedol, who is one of the best players out there apparently.  (Ranked #2 internationally)

Crazy to think what kind of AI we might see 10 years from now.

[Starver]

A google AI has beat a professional at playing Go.

This is kind of a big deal, as plenty of people have said this would never happen.  In March the AI will challenge Lee Sedol, who is one of the best players out there apparently.  (Ranked #2 internationally)

Crazy to think what kind of AI we might see 10 years from now.

obXKCD

(Looks like one (top-rated) human had a bad day, though, from reading elsewhere.)

[Telgin]

I find it pretty interesting that XKCD would say that there are games where AI may never beat humans.  After all, you can always emulate how a human thinks and just add more processing power and memory, in theory.  In practice, brute forcing problems with such a large state space does get wildly impractical fast, and I can imagine that it's quite challenging to program algorithms that have to effectively guess at the next best move for a lot of these games.

I was also surprised to see that there's a game called snakes and ladders.  I thought the SpongeBob eels and escalators sounded too silly to be based on something real. 

[i2amroy]

I find it pretty interesting that XKCD would say that there are games where AI may never beat humans.

Well, snakes and ladders is a totally luck based game, Mao is a game that's key feature involves making up new rules to the game as you go and then figuring them out, Seven minutes in heaven is a kissing game (though I guess a sexbot could maybe beat us there), and calvinball has literally no rules other than "whatever we make up". I think they're pretty safe in saying that computers may never outplay humans in those games (till the advent of strong AI, though even then they still may not beat the top humans in Snakes and ladders ).

[Egan_BW]

I find it pretty interesting that XKCD would say that there are games where AI may never beat humans.  After all, you can always emulate how a human thinks and just add more processing power and memory, in theory.  In practice, brute forcing problems with such a large state space does get wildly impractical fast, and I can imagine that it's quite challenging to program algorithms that have to effectively guess at the next best move for a lot of these games.

I was also surprised to see that there's a game called snakes and ladders.  I thought the SpongeBob eels and escalators sounded too silly to be based on something real. 

But in that case, that would be an emulated human beating a human, not a computer.

[Putnam]

I don't think that's a reasonable dichotomy.

[inteuniso]

I don't know if I've rambled on in here before, but with[1] multiple[2] studies[3] utilizing carboxylic acid to make graphene oxide out of graphite, and the impasse that humanity has somehow reached with an insatiable thirst for graphene coupled with an apparent lack of ability to mass-produce it, I will reiterate the obvious.

Cannabinoids are 2-carboxylic acids. If you mix carboxylic acid with graphite, you get graphene. Hemp is amazingly cheap to grow as well as advantageous seeing as cannabis improves soil quality. You mix cannabis/hemp with graphite (don't even need to process the cannabinoids out of the plant material, seeing as hemp nanosheets are on par if not better than graphene) and you have a graphene mixture that's roughly better than anything that's been made in a lab.

TL;DR The War on Drugs is dum