Science Facts Discovered in 2017

[jedishrfu]

A reference article to other articles on science facts that were discovered this year 2017. Some were theorized and finally proven, others were discovered totally by chance.

Which ones most impressed you?

http://www.sciencealert.com/23-science-facts-we-didn-t-know-at-the-start-of-2017

 

[mfb]

Two that didn't make into the list:

Humans suck at Chess, Go and https://www.uAlberta.ca/news-and-events/newsarticles/2017/march/poker-playing-ai-program-first-to-beat-pros-at-no-limit-texas-hold-em.
Computers also know how to do intelligence tests and they can identify plant species and exoplanets, https://www.technologyreview.com/the-download/609218/what-does-work-look-like-in-2026-new-statistics-shine-light-on-automations/, and some psychology, and probably everything else in a few decades.
Go was the last game with perfect information where humans were better. The new algorithms are not just a bit better, they play a different game. The new chess software didn't lose a single game out of 100 against the previously best chess algorithm (which had 1000 times the computing power available) - but won 1/3 of them. StarCraft is still an open challenge - but with work in progress. Computers are more and more moving to real-life problems. They typically start with a poor performance, but once they get somewhat close to human levels they quickly overtake us, and that pattern repeats everywhere. "But they can't do X" - yes, not yet.

We can beat plants in photosynthesis efficiency

Also interesting: We had a couple of studied suggesting humans evolved and spread over the planet earlier than thought. Homo sapiens earlier, out of Africa earlier, and into Asia earlier, earlier into the Americas, or maybe even much earlier.

 

[ISamson]

Wow. Really interesting list.
I was amazed by the possibilities with the mathematical time-travel possibilities, literally, in our own backyard.

I did not know that just where I come from, there are hundreds of thousand-year-old underground forests! This is so new and it could simply shake up the world of biology.

 

[Psinter]

Which ones most impressed you?

NASA Space Probes Have Detected a Human-Made Barrier Surrounding Earth
-https://www.sciencealert.com/nasa-space-probes-have-detected-a-human-made-barrier-shrouding-earth

Scientists Achieve Direct Counterfactual Quantum Communication For The First Tim
-https://sciencealert.com/scientists...tual-quantum-communication-for-the-first-time

This last one not only impressed, but I'm still baffled. Gotta leave it to quantum mechanics subjects to baffle people.

 

[mfb]

The time travel still needs negative energy densities as far as I understand - and probably ridiculous amounts or densities of it. Basically the same problem the Alcubierre drive has. That could be used for time travel as well. It is unclear if these negative energy densities can exist at all.

 

[AaronK]

Two that didn't make into the list:

Humans suck at Chess, Go and https://www.uAlberta.ca/news-and-events/newsarticles/2017/march/poker-playing-ai-program-first-to-beat-pros-at-no-limit-texas-hold-em.
Computers also know how to do intelligence tests and they can identify plant species and exoplanets, https://www.technologyreview.com/the-download/609218/what-does-work-look-like-in-2026-new-statistics-shine-light-on-automations/, and some psychology, and probably everything else in a few decades.
Go was the last game with perfect information where humans were better. The new algorithms are not just a bit better, they play a different game. The new chess software didn't lose a single game out of 100 against the previously best chess algorithm (which had 1000 times the computing power available) - but won 1/3 of them. StarCraft is still an open challenge - but with work in progress. Computers are more and more moving to real-life problems. They typically start with a poor performance, but once they get somewhat close to human levels they quickly overtake us, and that pattern repeats everywhere. "But they can't do X" - yes, not yet.

We can beat plants in photosynthesis efficiency

Also interesting: We had a couple of studied suggesting humans evolved and spread over the planet earlier than thought. Homo sapiens earlier, out of Africa earlier, and into Asia earlier, earlier into the Americas, or maybe even much earlier.

Indeed, 2017 was a great year for Machine Learning. Though if you talk to experts, you'll find a lot of disagreement about the future of AI, as well as the current state of research (i.e. Ali Rahimi's talk at NIPS this year). With respect to the development of a human-level (or beyond) general agent, there are respected experts on both sides--one side hates even mentioning the term 'AGI' and doubt that we will ever develop such a system (François Chollet for example), and the other side is of the mind that AGI is coming, and better to be in the developer's chair than not due to the implications of such a system (Shane Legg for example). To be fair, the former group are right that there exists no obvious path forward, but I personally tend to side with the latter, if only because I dislike the pessimistic attitudes and annoyed condescension of those that are certain discussions about the potential for AGI are not just useless, but completely unproductive.

If I'm being objective though, it's really anyone's guess. The only true direct path forward is through whole human brain emulation, and who knows how long that could take?

 

[mfb]

There was no obvious path to a strong Go program either. Or a strong Poker program. Or anything else in that list. Besides “we don’t have it yet so it cannot be easy” I haven’t seen good arguments from the first group, especially as we know a theoretical system that will perform at human levels - the full brain simulation.

 

[ZapperZ]

A reference article to other articles on science facts that were discovered this year 2017. Some were theorized and finally proven, others were discovered totally by chance.

Which ones most impressed you?

http://www.sciencealert.com/23-science-facts-we-didn-t-know-at-the-start-of-2017

I object to the use of the word "fact" to describe many of these things. It isn't a "fact" when it hasn't been discovered or verified. So how are these unknown facts before they were discovered/verified? The article doesn't seem to understand the difference between theory and fact.

Zz.

 

[Buzz Bloom]

The time travel still needs negative energy densities as far as I understand - and probably ridiculous amounts or densities of it.

I think that it is mostly generally accepted that mathematics is not a description of physical reality. Mathematics is a tool that enables the making of predictions, (sometimes probabilistic) that can be validated by experiments. Mathematics can also be used to create models of alternative plausible physical possibilities which experiments might be able to confirm that some possibilities are more likely than others.

The description of the time machine does not yet make any testable predictions since there are no descriptions of what the necessary material might actually be from which one might build such a device. The article also did not discuss how this kind of time machine would avoid the usual paradoxes caused by time travel.

Of the 23 listed "scientific discoveries", this one would be at the very bottom of my list by a very large distance. However, it might possibly be a good basis for a "hard science fiction" time travel story.

Regards,
Buzz

 

[bhobba]

12 - Carl Sagan was freakishly good at predicting the future - his disturbingly accurate description of a world where pseudoscience and scientific illiteracy reigns gave us all moment for pause.

He was - but why did it take so long to realize? The standard of rational, little alone scientific debate, amongst the general public is shockingly low. I have a friend with a biochemistry degree, a masters in computer science, and a masters in education that believes the moon landing a fake - and IMHO its getting worse. No - he doesn't teach anymore - he only did it for a year before realizing it was not for him - he is retired now.

Another example - professor Hattie at the University of Melbourne is the worlds leading expert in fact based education - well if not the leading expert one of them anyway. He has found out all sorts of things such as homework does no good, class sizes don't really matter - the list goes on. It has all been reported in the press, and even a number of TV specials about it done, where what he found was implemented with amazing results eg a school in the bottom 10% jumped to the top 25% in one year. But when they went out surveying what people thought made a good education nearly all believed the exact oppose of what the research showed.

Its crazy, utterly crazy - and as I said is getting worse.

Thanks
Bill

 

[bhobba]

I think that it is mostly generally accepted that mathematics is not a description of physical reality.

Guys like Penrose may disagree with you on that. If you don't know his view read some of his books such as the Emperors New Mind:
https://en.wikipedia.org/wiki/The_Emperor's_New_Mind

There are all sorts of views out there - some quite strange and wacky - but it also must be admitted strange and wacky is in the eye of the beholder.

Thanks
Bill

 

[Buzz Bloom]

Guys like Penrose may disagree with you on that.

Hi Bill:

The Wikipedia article on the Penrose book you cited mentions mathematics in a very limited way:

"the philosophy and limitations of mathematics"​

If you have a copy of of the book handy, could you please post a quote related to Penrose's disagreement with what I said in your quote from my post.

Regards,
Buzz

 

[bhobba]

If you have a copy of of the book handy, could you please post a quote related to Penrose's disagreement with what I said in your quote from my post.

I don't - but I also don't think it's something in a single quote.

What Penrose believes about math is well known. Reality lies in a platonic realm which mathematical truth literally resides. What we experience is a shadow of that realm. Penrose does not believe math is a description of reality in the language of math - like saying the apple lies on the table is a description of an apple lying on the table in English - he literally believes mathematics is the reality. Wacky hey? But when you look into it its very seductive - so seductive I believed in it myself - until I woke up to just how whacky it was and also not necessary after I saw a lecture by Gell-Mann:
https://www.ted.com/talks/murray_gell_mann_on_beauty_and_truth_in_physics

Thanks
Bill

 

[Buzz Bloom]

What Penrose believes about math is well known. Reality lies in a platonic realm which mathematical truth literally resides.

Hi Bill:

Thanks for the clarification. So Penrose is a Platonist. I did not think that they existed any more. Certainly they must represent a minority point of view.

Regards,
Buzz

 

[bhobba]

Thanks for the clarification. So Penrose is a Platonist. I did not think that they existed any more. Certainly they must represent a minority point of view.

I don't know - his view is very seductive - and, after all, Penrose is - well Penrose - a person not to be dismissed lightly.

Thanks
Bill

 

[Auto-Didact]

Not sure how this veered from the latest experimental discoveries into differing philosophies of mathematics.

On topic, I found the artificial "plastic bag" womb quite innovative. Kind of curious how long it will be until it evolves into full blown child factories like on the Kryptonian ship in Man Of Steel, or perhaps just synthetic autologous organ and animal meat farms.

 

[Auto-Didact]

I don't - but I also don't think it's something in a single quote.

What Penrose believes about math is well known. Reality lies in a platonic realm which mathematical truth literally resides. What we experience is a shadow of that realm. Penrose does not believe math is a description of reality in the language of math - like saying the apple lies on the table is a description of an apple lying on the table in English - he literally believes mathematics is the reality. Wacky hey? But when you look into it its very seductive - so seductive I believed in it myself - until I woke up to just how whacky it was and also not necessary after I saw a lecture by Gell-Mann:
https://www.ted.com/talks/murray_gell_mann_on_beauty_and_truth_in_physics

Thanks
Bill

Actually Penrose does not subscribe to the type of Platonism that you are describing there. Max Tegmark on the other hand does, he is a full blown mathematicist. Tegmark actually believes that literally everything is literally directly part of mathematics, i.e. only mathematical things exist, even literally unreal and fictional things are existing mathematical objects if they can be described by some form of mathematics.

Penrose is much milder in his Platonism, he believes there must be some kind of domain of mathematical existence, wherein mathematical objects 'reside', the reason for this being the incredulity argument w.r.t. construction, i.e. he believes that humans did not invent certain mathematical concepts, but discovered them. Moreover, Penrose believes that a special subset of such mathematical things are specifically part of the actual laws or principles of physics. Penrose' actual Platonism is therefore a realist position with regard to particular mathematical concepts as being part of the real world encoded through physics. In this way for example Penrose is a strong believer that the laws of physics are holomorphic and that therefore holomorphicity is a deep principle or property of physical reality.

Outside of this general philosophical stance, Penrose also professes an intra-theoretical prediction which happens to contain a more radical version of Platonism. This prediction is part of Penrose's own physical theory of consciousness, which is based on an alternative theory to quantum mechanics which incorporates gravitation, but which formally has not been fully constructed or named outside of being dubbed 'gravitized QM' (gQM) based on the name of the paper. One explicit prediction of gQM is the occurence of Diosi-Penrose objective reduction, there are at the moment over a dozen experiments running to find this effect. Based on gQM, is another more controversial theory, namely Penrose's own particular physical theory of consciousness, wherein he states that the mind is a physical object and that certain mental Platonic concepts are possibly directly characteristically encoded in the quantum description of spacetime structures, thereby describing mentality and explaining how and why the human mind can see, intuit and 'feel out' certain mathematical truths by pure virtue of its deep physical architecture. This predictive Platonic stance is purely theory dependent, in the sense that if the core idea behind his theory of consciousness (Diosi-Penrose objective reduction) is falsified, so is this prediction.

Source: I have all Penrose' books and have read them all at least twice, and have heard him make his position clear on this topic on multiple occasions.

Thanks for the clarification. So Penrose is a Platonist. I did not think that they existed any more. Certainly they must represent a minority point of view.

As Penrose himself points out, many if not all mathematicians either were or are Platonists to some degree with regard to at least some mathematical things. You'd be surprised at the number of Platonists once pressed. This is because most mathematicians, like most scientists, have not engaged in philosophical reflection regarding their own domain or even their own ideas.

This has both an advantage and disadvantage, namely more often than not once pressed explicitly they are revealed not to be subscribed to no philosophy but to the worst philosophy, while implicitly when not pressed the door luckily tends to remain open to philosophical stances outside the popular or historical ones. In contemporary mathematics, the popular philosophies are still formalism, logicism and intuitionism; none of these are probably the correct philosophy of mathematics and not explicitly adopting a known philosophy in the practice of mathematics may actually help in enabling in eventually discovering the correct philosophy of mathematics, and so even lead to the discovery of more mathematics.

 

[bhobba]

Source: I have all Penrose' books and have read them all at least twice, and have heard him make his position clear on this topic on multiple occasions.

I have read them as well, but not twice and its a while ago now.

I simply bow to your better recollection and understanding the nuances involved.

Still my point is there are all sorts of views out there some of which many would find strange.

Thanks
Bill

 

[collinsmark]

I agree with @ZapperZ on the misuse of the word "facts" in the original article.

On a different note, yet still in the spirit of this thread, here is this documentary that I found fascinating:

 

[Ygggdrasil]

A few from the biology side of things:
Humans and all other eukaryotes evolved from a newly discovered superphylum of microbes which have been named the Asgard archaea:
https://www.nature.com/articles/nature21031
https://www.theatlantic.com/science/archive/2017/01/our-origins-in-asgard/512645/

The genetics of complex traits is more complicated than we though:
http://www.sciencedirect.com/science/article/pii/S0092867417306293?via=ihub
https://www.theatlantic.com/science/archive/2017/06/its-like-all-connected-man/530532/

Scientists can successfully use gene therapy to correct neuromuscular disorders in children:
http://www.nejm.org/doi/full/10.1056/NEJMoa1706198
http://www.sciencemag.org/news/2017...pe-neuron-targeting-virus-saving-infant-lives

Genetic markers predict which patients will respond to a new cancer immunotherapy drug across a wide variety of tumor types:
http://science.sciencemag.org/content/357/6349/409
http://www.cancerresearchuk.org/abo...s-could-help-predict-success-of-immunotherapy

 

[scottdave]

Not sure how this veered from the latest experimental discoveries into differing philosophies of mathematics.

On topic, I found the artificial "plastic bag" womb quite innovative. Kind of curious how long it will be until it evolves into full blown child factories like on the Kryptonian ship in Man Of Steel, or perhaps just synthetic autologous organ and animal meat farms.

Reminds me like the book Brave New World.

 

[scottdave]

12 - Carl Sagan was freakishly good at predicting the future...

Another example - professor Hattie at the University of Melbourne is the worlds leading expert in fact based education - well if not the leading expert one of them anyway. He has found out all sorts of things such as homework does no good, class sizes don't really matter - the list goes on. It has all been reported in the press, and even a number of TV specials about it done, where what he found was implemented with amazing results eg a school in the bottom 10% jumped to the top 25% in one year. But when they went out surveying what people thought made a good education nearly all believed the exact oppose of what the research showed.

I'd like to read more about that. I can go Googling, but do you have any reference?

 

[bhobba]

I'd like to read more about that. I can go Googling, but do you have any reference?

http://www.abc.net.au/tv/programs/revolution-school/
http://www.abc.net.au/4corners/revolution-in-the-classroom/3815376

Here is the actual research:
http://www.abc.net.au/tv/programs/revolution-school/Summary_Survey_And_Research.pdf

Here is the truth and what Australians think is the truth:
Australians get fail mark on what works to improve schools A significant number of Australians wrongly believe that smaller class sizes, compulsory homework and private schooling all lead to better academic results for students. A national survey has found serious misconceptions about the most effective ways to raise Australian academic standards, which have fallen significantly in international rankings over the past decade. The survey was conducted in conjunction with the landmark fourpart ABC series, Making the Grade, which follows a year in the life of Kambrya College, a state secondary school in Melbourne’s outer south-eastern suburb of Berwick. In 2008 Kambrya’s Year 12 results put it in the bottom ten per cent of secondary schools in Victoria. Making the Grade follows the transformation of the school under the leadership of principal Michael Muscat, to the point where it is in the top 25% of schools. Muscat and his colleagues manage more than 1000 students, including those struggling to cope with school and home life. Making the Grade gives a raw and honest insight into the challenges facing these teenagers, while also showcasing what really works in classrooms to improve academic results. The series highlights the internationally renowned research of Professor John Hattie, and one of the world’s top ranked education institutions, the University of Melbourne’s Graduate School of Education. During 20 years of research analysing more than 70,000 studies involving a third of a billion students from around the world, Professor Hattie has established what is most effective to improve student learning.

He has found that teaching which involves goal-oriented, specific feedback to students, and positive teacher-student interaction, have the most impact on learning growth. Contrary to what the survey has revealed many Australians believe, Hattie’s research has established that smaller class sizes, state-of-the-art facilities and hours of homework have little or no impact on results:

“Reducing class size does enhance achievement, however, the magnitude of that effect is tiny,’’ says Professor Hattie. “And the reason that it’s so small is because teachers don’t change how they teach when they go from a class of thirty to fifteen.’’ The national survey found more than three quarters of Australians incorrectly think smaller class sizes have a positive impact on academic achievement. Less than 10% of people got this right."

Australians had other misconceptions:
1. Fifty two percent believe wearing a school uniform has a positive impact on students’ results, but Hattie’s research has found it has no impact at all
2. When asked if the academic achievement of secondary school students was better at single sex schools compared to co-ed schools, only a third correctly answered that it was not
3. More than two thirds of Australians incorrectly think that regular homework is essential for students to succeed at secondary school
4. When asked if the standard of teaching in private schools promoted greater academic growth among students compared to teaching in government schools, 43% wrongly answered yes
5. Only 34% correctly answered that there was no difference between private and public schools in terms of student’s academic growth
6. Nearly a third of Australians under-estimate the number of hours teachers work each week.

The survey also found two thirds of people think teaching is a worthy profession but ranked teachers behind doctors, lawyers, university lecturers, elite athletes and nurses in terms of perceived status in our society. More than two thirds of people think schools should place more emphasis on literacy and numeracy, but only 13% strongly agree that Australia should push secondary students harder to outperform Asian countries.

Just an example of the actual research, not peoples perceptions:

Myth:
Homework is a necessary evil
Reality:
Hattie’s meta-analysis has shown that the amount of homework a student does in primary school has no effect on student achievement or progress. He is not saying that there should be no homework, but if schools are going to set homework (which many parents expect) then the focus should be on the type of homework given. For example, children at primary level should be given fewer projects but could instead be set short activities to reinforce what they learned that day. Homework does have more effect on results for secondary school children, but generally students are given too much. A short time spent practicing what was taught that day can have the same effect as one or two hours of study. John Hattie says he rarely enforced his own four children doing homework, saying it was what happened in the classroom that mattered. “Homework in primary school has an effect of around zero. In high school it’s larger… Which is why we need to get it right. Not why we need to get rid of it … If you try and get rid of homework in primary schools many parents judge the quality of the school by the presence of homework. So, don’t get rid of it. Treat the zero as saying, “It’s probably not making much of a difference but let’s improve it”. Certainly I think we get over obsessed with homework. Five to ten minutes has the same effect of one hour to two hours. The worst thing you can do with homework is give kids projects. The best thing you can do is to reinforce something you’ve already learnt.”

Myth:
Reducing class size leads to better outcomes for students
Reality:
Reducing class size can enhance student achievement but generally the effect is only marginal. What really matters is that the teacher is effective and having an impact, no matter what size the class is. Hattie: “Well, the first thing is, reducing class size does enhance achievement. However, the magnitude of that effect is tiny. It’s about a 105th out of 130-odd effects out there and it’s just one of those enigmas, and the only question to ask is, ‘Why is that effect so small?’ And the reason, we’ve found out, that it’s so small is because teachers don’t change how they teach when they go from a class of thirty to fifteen.’’

Bottom line - teacher quality and how they teach is the key - all the rest is mostly not worth worrying about. Yet people think it's the other stuff that's important - it isn't. Now this is counter intuitive and you certainly can't blame people for not knowing it - I don't. What concerns me, and the comment in relation to Carl Sages observation, is all this has been on the media a lot, at least here in Australia. They should know it. But it just seems to go in one ear and out the other.

Thanks
Bill

 

[mfb]

Just going by what you quoted here, I think the author tried really hard to tell everyone else that they are wrong.

The national survey found more than three quarters of Australians incorrectly think smaller class sizes have a positive impact on academic achievement.

I wonder why they would think this?

Reducing class size does enhance achievement

And I would like to offer another point of view: Increasing the class size will reduce achievements. The study highlights the need for "specific feedback to students, and positive teacher-student interaction" - that gets more difficult with larger classes and more classes. You have less time for each student, and it gets harder to memorize details about every single student.

6. Nearly a third of Australians under-estimate the number of hours teachers work each week.

So more than 2/3 overestimated it? Or do we make arbitrary "that agrees" ranges to get whatever fraction we want?

The survey also found two thirds of people think teaching is a worthy profession but ranked teachers behind doctors, lawyers, university lecturers, elite athletes and nurses in terms of perceived status in our society.

There can be more than one worthy profession. You don't have to rank everything you call a worthy profession at rank 1.

And the reason, we’ve found out, that it’s so small is because teachers don’t change how they teach when they go from a class of thirty to fifteen.’’

Maybe they are simply not used to smaller classes and teaching would change after a year or two.

 

[bhobba]

Maybe they are simply not used to smaller classes and teaching would change after a year or two.

They have been trying smaller class sizes for years - utter flop. And he found what impact there was as tiny. To be exact - he said - “Well, the first thing is, reducing class size does enhance achievement. However, the magnitude of that effect is tiny. It’s about a 105th out of 130-odd effects out there and it’s just one of those enigmas, and the only question to ask is, ‘Why is that effect so small?’ And the reason, we’ve found out, that it’s so small is because teachers don’t change how they teach when they go from a class of thirty to fifteen.’’

So why not do those over 100 other things first?

Well yes Professor Hattie is highly likely to tell his side of the story - he has spend 20 year compiling the data.

What is the scientific method - try it and see - but of course trying it means just that - do not implement such things until you are 100% sure it works.

Also look at the schools where it was tried - it worked brilliantly. They however are a very small sample - it needs to be tried at more schools.

The point though was not the intellectual debate you are engaging in - it was this has been widely reported here in Aus and people still believed what they wanted to believe - not what research showed. If, like you, they engaged in intellectual debate about it - fine - even better than blindly accepting it - but to have little or no impact at all - that's sad.

But I think discussing this particular example of item 14 really needs its own thread - its just an example - not the main point of the thread.

Thanks
Bill

 

[mfb]

So why not do those over 100 other things first?

Maybe some are done, some cost money, and some are unpopular for other reasons.

Also look at the schools where it was tried - it worked brilliantly.

Just telling teachers you are trying something new and monitoring the progress can have a positive effect already.

I don't say he is wrong, but he should work on the way he presents his results, that might increase the acceptance of them.

 

[bhobba]

I don't say he is wrong, but he should work on the way he presents his results, that might increase the acceptance of them.

That is very possible. He is an academic and would have presented his results in academic journals. He has spoken to the media of course - that's why it has been reported - and quite frequently actually. That's why some schools trialed it - they were desperate and did it as a last resort - which itself speaks volumes. The thing is it worked. It should be opened up to even larger trials - and not just those so desperate they will try anything. But teachers will have to change their ways - changing how you do things is a lot harder than saying - give me smaller class sizes. So naturally it doesn't have the enthusiastic support of teachers - they always harp on easier things for them - like smaller class sized etc than how they teach. BTW this is not a criticism - its just natural human nature. I remember I had to change my ways when I went from a computer programmer in one government department to another. There walkthroughs where you are criticized by your peers was mandatory - it took some adjusting - and it wasn't exactly nice.

Thanks
Bill