Comment by searine | Hacker News Reader

Comment by searine | Hacker News Reader

I've been following this story pretty closing for several months now. So here's the opinion of an actual research scientist.

Q. Does the Em drive produce thrust?

A. Undetermined. Several tests from multiple independent labs have reported thrust, including in a vacuum. However several of the labs that reported positive thrust come from a sketchy background.

Q. Does it violate conservation of momentum?

A. Undetermined. There are several competing theories trying to explain the phenomenon, but we lack raw data. I think the last thing anyone wants to see here is "new physics".

Q. Roger Shawyer seems like a scummy kook?

A. Yes. This is why more testing is really important.

Q. What about more tests?

A. A few dozen people have taken it upon themselves to hack together emDrives and test them. The first couple so far were plagued by poor construction, RF interference, and poor experimental design. However, the hacking community is slowly getting its shit together.

Q. Pluto in 18 months?

A. If it works. My intuition screams "its cold fusion all over again", but there is still a chance.

> I think the last thing anyone wants to see here is "new physics".

The article's responds with the nonsensical "Some damage to our theories of physics is an acceptable payoff if we get a working space drive."

EDIT: Let me clarify why it's a nonsensical statement:

First, I think it meant to say "cost", not "payoff". Second, there would be no cost. None of the things we were able to do with our existing theories would be lost. We'd only be pushed to refine the theories (or in rare cases come up with an entirely new theories) that would be at least as accurate as the prior theories and cover phenomenon the old didn't. If anything, we'd be able to go back and do better at all the things to which we'd applied the old theories.

It's like saying "Some damage to Newtonian physics is an acceptable cost if we get a working understanding of relativity and gravity." Einstein did no "damage" to Newtonian physics. He made it better.

Well, it’s improbable, not nonsensical. If someone goes and implements a technology that trounces our working theories, we certainly need to update those theories. It’s just that that’s exceedingly unlikely to happen.

Photoelectric cells were invented before we understood the photoelectric effect, alcohol and preserves were used to store drinks and foodstuffs before we understood microbes, forging and annealing used before anyone had a theory on metallic crystal formation, and incandescence was used before anyone understood electromagnetic radiation; to name but a few examples. Why would it be surprising that someone invented a device which functions in a manner which is not understood?

Oh! This is one of my favorite questions about scientific progress.

People seem to regard scientific progress as this kind of wave function, where some theory comes along and then some other theory or discovery disproves the previous one and so on and so forth.

But if you want to think of scientific progress that way, then you have to think of it as a dampened sine wave (https://en.wikipedia.org/wiki/Damped_sine_wave) -- that is, discoveries and theories tend to refine previous discoveries and theories, and scientific progress over the millenia has been rapidly zeroing in on the truths of our reality.

What is left at this point is mostly things at very large scales -- cosmological stuff like dark matter, dark energy -- and very small scales -- stuff like sub-subatomic particles and the Higgs boson and the nature of mass and so on -- and very complex things like biology and climate science.

What isn't left at this point is a revision of the laws of conservation. It is exceedingly unlikely that anything will be discovered at any point in the future which will do anything other than add perhaps the very smallest of exceptions to those laws, and even then, I doubt very much that will happen.

So, essentially, the age of physics discoveries in garages is mostly over. There will be many more discoveries in physics, but they now require the collective efforts of entire nations.

I'm sorry, but your reasoning completely escapes me. The only things that are left are very big, and very small? In addition to this being an incredibly vague catchall (which misses all the possible discoveries of 'black swans', that is phenomena and occurrences we have never seen), your statement more importantly misses the fact that humans don't even understand what something as common and important as light (or electromagnetic radiation in general) is. Is light really big or really small? We've certainly characterized light's behavior (to some degree), though we still have basically no understanding of what it actually is (though I may be the only one who finds the circular definitions of electromagnetic fields unsatisfying). The same is true of things like matter and gravity. Once we have better understandings of what these are, I may grant that further fundamental insights will be difficult to come by; but we're not there yet.

You might be correct in saying that humans have characterized particles, objects, and phenomena that are closest to their scale and location, but we have done a poor job of understanding their fundamental nature. We also don't understand and have not even characterized things we've never observed (the 'black swan' problem).

I agree. I also take issue with

> rapidly zeroing in on the truths of our reality

The truths? No, humans are as likely to arrive at the truths as ants an understanding of a microwave oven.

A fair reading of that part would assume that I meant truths like, "the cosmos is expanding", not philosophical truths.

I think it was implied that "truths" meant scientific truths, which are well defined as anything that can be verified beyond statistical doubt with experimental evidence.

> We've certainly characterized light's behavior (to some degree), though we still have basically no understanding of what it actually is.

As a physicist I really can't imagine what answer would satisfy you and could be falsifiable the same time. It is what it is, all we can do is to characterize its behavior.

One could say the same of a stone, a plant, a planet, or a star, yet we do not simply characterize any of these, we endeavor to understand its origins and constituent parts. In the case of the star, we could have analysed its spectrum, size, trajectory (through space), and surface characteristics, then called it a day; yet we continually try to understand the elements which make up the star, what their state of matter is, how they are reacting, and many other aspects of the star. I disagree with you when you say (of light) that "all we can do is to characterize its behavior".

Well maybe that's not all we can do. Maybe the abstraction of treating light and the rest of the Universe somewhat separately will break (of course we describe the coupling too). Maybe the abstraction of treating light as a non-dividable entity in the Universe will break. It would only create other axioms and/or other non-dividable entities, at what point would you be satisfied? It could very well be that you can't divide light any further and I would be satisfied with it if we can describe its behavior fully. If one says that light is a broken abstraction and you can describe the Universe using only "string" then you would ask "Yeah, but what strings are?".

I'm not saying that asking what light is is not an important question. But there is a possibility that there is simply no answer and yet we can have fully working physics. At least nobody will question your theories just because you don't tell what exactly your elementary particles are. They are elementary, we just have them, maybe light is not one of them.

I think that your last point gets to the heart of the question, and you are right that we may never achieve a deeper understanding of light, matter, or gravity; but that does not mean that a deeper understanding is not desirable, or that we should stop asking the questions. Further, we will never know for certain when we have reached the bounds of our capacity to understand, so I will continue to ask "Yeah, but what strings are?". I always like to keep in mind that 'atom' means 'indivisible' in Greek, and yet we found that bound of knowledge to be surmountable (with debatable consequences).

I would also like to be clear that I am not criticizing anyone for making insufficient progress; it is just that there is a lot out there left for us to understand, and not all of it is marginal stuff that's "very big or very small".

> we still have basically no understanding of what [light] actually is.

Finding the true nature and, whether it exists or not, is the domain of philosophy.

Science builds models. And we do have at this point extremely good models of electromagnetism which can correctly predict results of all experiments done so far.

Whether these models reveal some fundamental truth or not is a matter of opinion (or perhaps religion). It has nothing to do with refining the models or finding experiments that might prove them wrong.

I think you misunderstand my point. Model-building is only one aspect of science, and happens to be the most prevalent in some (though not all) of today's mathematical and statistical based studies of physics (which are often criticized as an exercise in curve-fitting). One example is how string theorists are making a valiant (though perhaps errant) effort to understand matter; they are not simply modelling well characterized behaviors. There are many physicists and other scientists who are continually trying to understand their subject matter (pun intentional) at a deeper level, such as those who discovered and described DNA.

I am not interested in whether these scientists reveal "some fundamental truth or not", but we cannot say that we understand all there is to know about electromagnetic phenomena because we can make a few simple predictions. It would be tantamount to saying that we understand the sun because we know where it is relative to us, and how bright it is, and can predict both of these parameters for a while. Understanding that it is a star, how it was formed, how it will cease to be, that it is made up of a number of elements, and converts large amounts of matter to energy are all very important discoveries, and we must continue to seek this kind of advance in the frontiers of our understanding in electromagnetic phenomena, matter, gravity, and many other areas.

So, I originally argued the point that progress in the physical sciences generally doesn't involve utterly canceling out previous discoveries. The link to Asimov's article elsethread illustrates this better than I could hope to. Relativity for instance did not find that Newtonian mechanics was wrong, it simply revised it under pathological cases. The Bohr model of the atom isn't completely wrong, and it's still useful in chemistry, it's just that as physicists have learned more they've found details where it's not completely right either. This kind of constant revision is a theme throughout the history of scientific progress.

So that leaves me a little bit stumped at what point you're trying to make here. Are you disagreeing with that view of science? Are you trying to say, "well, that's how it's been for hundreds of years, but at some point in the future I believe we'll all discover that physics is completely and utterly and embarrassingly wrong in every way"?

I've gone back and reread your reply to me at least half a dozen times. You seem to be disagreeing with me, but none of the examples you brought up really actually disagreed with me. At best, I chose some poor metaphors and that led you to misunderstand me, but then I still can't figure out what it is that you're actually trying to say.

I guess you can say that ariving at Maxwell equations or QED is an exercise in curve-fitting in some sense. In the end, it is the simplest set of rules that matches a set of observations.

Similarly, the current model of the sun is the simplest model that explains some observations (neutrino flux, the fact that some stars seem similar spectroscopically, our knowledge from other branches of science, etc.).

I see no fundamental difference between these two.

"we cannot say that we understand all there is to know about electromagnetic phenomena because we can make a few simple predictions"

This is quite an understatement. QED, which is the underlying theory of EM, is the most accurate scientific theory ever. Full stop. It's theoretical predictions match experimentally measured observables out to ten decimal places. By your standards we don't really understand any physical phenomena. To bring up your analogy with the sun: our understanding of QED is like saying that we know the exact chemical composition of the sun to an accuracy of one part in ten billion. I think we would be justified in stating that we understand the sun in those circumstances.

On a related note, QED does provide a lot of context for what "light"(photons) is(are). It's the gauge boson (read: mediating particle) of the electromagnetic force. In my mind it doesn't get much more elegant than that.

Actually, there is still the possibility that conservation/symmetry is also just an approximation of the actual fundamental laws. Lee Smolin argues that way in his book Time Reborn: https://santitafarella.wordpress.com/2013/06/04/lee-smolins-...

Relevant essay by Asimov: The Relativity of Wrong

http://chem.tufts.edu/answersinscience/relativityofwrong.htm

Thank you! I had read this in the past, and it was one of two major influences on my thinking about scientific progress, but I couldn't remember exactly what it was I had read.

Asimov naturally describes it much better than I could.

http://amasci.com/weird/end.html

Yeah, that specific article was on my mind when I wrote my comment. But I didn't say there were no more discoveries left to be made; I said they weren't happening in people's garages anymore, which is true in the majority.

Inventing a device which is not well understood and inventing a device which contradicts things that are thought to be well understood are two different things. The photoelectric effect was pretty big news though and it did break existing laws that time so it doesn't really fit here.

I could argue that all the inventions I mentioned violated one existing theory or another at the time of their creation, though the level of scientific reasoning behind each is different. Remember that at one time, many believed that heat was an element released from logs by fire, which is a theory clearly violated by the incandescent light bulb, and other earlier examples of incandescence (such as that of other heated metals).

Regardless, how is it relevant whether a large group of people made a mistake (by coming up with an incorrect theory) before an invention that had no basis in existing theory was made? I do not agree with the premise of the post I responded to, as a new drive does not necessarily contravene existing theories (since we do not understand its principle of operation).

I agree that contravening an existing theory is a bit of an odd course for an inventor to take, and may make it difficult to get angel funding, but that's about the only difference I can see.

I think it would be really exciting if we found a flaw in our model of physics.

Not sure what is to fear from new physics. Sounds like something new and exciting to work on. Aren't physicists tired of perpetually iterating over string theory and making no new discoveries?

Thanks, nice summary.

> I think the last thing anyone wants to see here is "new physics".

I really want to see new physics, because I want to see us get off this rock and far away. I think most people want to see new physics. I think that's why this topic is so popular.

That said, I can understand physicists not wanting to see new physics.

Wait. In a way, that sentence is crazy. In a way, physicists should want to see new physics more than anybody.

Where does the "not wanting to see new physics" sentiment come from? Just not wanting to be wrong?

I read it as: we want a better explanation, than just: "new physics!"

Maybe not wanting to see new physics before seeing rigorous, reproducible experimental results? Kind of a cart before the horse thing? Just guessing.

Laymen tend to underestimate the very vast body of research which supports current physical models. "New physics" would imply that something has been fundamentally wrong with physics for a long time, and that is unlikely.

I get that.

I'd love to see new physics! Every physicist I know loves new physics.

But if you claim to violate the conversation of momentum, everyone is going to be skeptical and not believe it. We have hundreds of years of experiments and observations that support the law and have looked in many places from small to large and not seen violations. This is what makes people not believe it and thus physicists are going to assume you are wrong and not physics. It's not so much a negative concern, but more of a practical one. But should it be tested in a rigorous manner and reproduced, then physicists will believe it. It's just it hasn't and thus we are going to stick with the established very well understood and tested theory. Like are you sure you have properly counted all the initial momentum? Are you 100% you are not introducing energy to the system? The possible explanations is long and a good experiment will eliminate as many of these as possible.

As an example, look at the OPERA experiment which found to have superluminal signals. New physics? Nope, just a faulty piece of equipment. BICEP showing gravitational waves? Nope, just poor data interpretation and neglecting interstellar dust. So if you want to claim new physics, be damned sure you have ironclad proof.

> Every physicist I know loves new physics.

If physicists don't want new physics, they should stop doing research!

Only as long as the new physics is in your little sphere of influence so you can keep the grant money gravy train flowing. Presumably.

That makes sense, but that's not how I interpret "nobody wants to see new physics" in the ancestor comment. (But who knows.)

I mean, wouldn't it be really frustrating to find out the law of conservation of momentum is not universal? Exciting, but also frustrating? Maybe even embarassing for physics? It would be like someone proving P=NP.

99.99% of the time when "new physics" is invoked, it amounts to crookery or kookery. Nobody wants to see that, because it takes energy to disprove even the obviously flawed claims, and the energy is taken away from more promising endeavors, that are usually less appealing to the public.

"Free Energy" is very appealing, so is "10x better batteries", "10x more flash storage", "momentum-defying space thruster", and so on. These modern-day fairy tales are gaining press simply because people want them to be true, not because they actually show a path for progress.

Similarly, a promise of a "miracle cure" is very appealing to the public, and so it may seem cynical to downplay any such claim right away, but it has absolutely nothing to do with actual advances in medicine.

EmDrive and Cold Fusion are the perfect complementary technologies. Next someone should claim to invent a cooling system that doesn't reject heat, and we can start building Stargate Puddle Jumpers.

Reply on news.ycombinator.com