Elon Musk's 'Scientific Method' (rollingstone.com)
From a new wide-ranging interview of Elon Musk: An unfortunate fact of human nature is that when people make up their mind about something, they tend not to change it -- even when confronted with facts to the contrary. "It's very unscientific," Musk says. "There's this thing called physics, which is this scientific method that's really quite effective for figuring out the truth." The scientific method is a phrase Musk uses often when asked how he came up with an idea, solved a problem or chose to start a business. Here's how he defines it for his purposes, in mostly his own words:
1. Ask a question.
2. Gather as much evidence as possible about it.
3. Develop axioms based on the evidence, and try to assign a probability of truth to each one.
4. Draw a conclusion based on cogency in order to determine: Are these axioms correct, are they relevant, do they necessarily lead to this conclusion, and with what probability?
5. Attempt to disprove the conclusion. Seek refutation from others to further help break your conclusion.
6. If nobody can invalidate your conclusion, then you're probably right, but you're not certainly right.
1. Ask a question.
2. Gather as much evidence as possible about it.
3. Develop axioms based on the evidence, and try to assign a probability of truth to each one.
4. Draw a conclusion based on cogency in order to determine: Are these axioms correct, are they relevant, do they necessarily lead to this conclusion, and with what probability?
5. Attempt to disprove the conclusion. Seek refutation from others to further help break your conclusion.
6. If nobody can invalidate your conclusion, then you're probably right, but you're not certainly right.
What bizarre set of questions, axioms and probability of truths would lead someone to conclude that drilling lots of tunnels without governmental oversight under major metropolitan areas is something that will reduce traffic, be good for the environment, etc?.
It reeks of a frosty pith.
We use "scientific consensus" now.
1. Ask a question.
2. Find a group of people who give the answer you want.
3. Misconstrue their statements to remove nuance and ambiguity
4. Package them all together into a "meta study"
5. Tell everyone "the science is settled".
>> 5. Attempt to disprove the conclusion. Seek refutation from others to further help break your conclusion.
#5's "refutation" seems to diminish with wealth and power. Ask anyone done in by a chorus of "yes men" afraid to challenge their meal ticket...
Yeah, that really isn't the scientific method.
"Attempt to disprove the conclusion."
This seems to be running fast and loose with the requirements of experimentation. One really needs to prove a hypothesis otherwise the effort is somewhat incomplete. I can't disprove God exists, but to make the assumption that the entity does exist for this reason is lazy and dishonest.
cheap / profitable transportation. . . ha!
Get billions of dollars in taxpayer funded subsidies for something that nobody in their right mind would invest their own money in.
Which is discouraged in education these days.
If you get all "evidence" from liberal mainstream media then you would know that global warming and evolution are proved, when they are really just shitty unscientific and unprovenable theories. The problem with the science today is that it is allways infected by liberal SJW political correctness.
This script is how all of us Trump supporters made up our minds. Dug up my old notes, here they are:
1. Ask a question:
Would Donald J Trump be an excellent president, better than any other?
2. Gather as much evidence as possible about it.
Scan bookmarked news sources for articles supporting this claim. Breitbart? Check. Fox News? Check. Infowars? Check.
3. Develop axioms based on the evidence, and try to assign a probability of truth to each one.
Businessman? 100% per Wikipedia. He sells buildings and stuff, ergo a businessman.
Smart? Yes 100%, per his Twitter feed he confirmed it himself. No better source than the source itself so don't need to look further.
Good looking with excellent hair? Yes, 100%. See entry for 'smart' above.
4. Draw a conclusion based on cogency in order to determine: Are these axioms correct, are they relevant, do they necessarily lead to this conclusion, and with what probability?
Donald J Trump would be an excellent president. All axioms were correct and relevant. Since all axioms were 100% the conclusion must be 100%.
5. Attempt to disprove the conclusion. Seek refutation from others to further help break your conclusion.
Scanned my Twitter feed. Nope, no dissenting views among those I follow. Nor on Facebook. Therefore p=0.00.
6. If nobody can invalidate your conclusion, then you're probably right, but you're not certainly right.
Nobody invalidated my views so my choice is 100% certainly right. Further, the opponent is evil and 100% wrong so my conclusion cannot be wrong.
It's similar to the scientific method philosophically.
But the scientific method is about _empirical_ inquiry. You need to conduct experiments with defined methodology and measure hard data for each of these "axioms".
Who, among Slashdot's esteemed editorial board, decided, the publication's audience needs a refresher on what scientific method is?
And who, subsequently, chose the Rolling Stone — whoever it is they are interviewing — as the best fount of this illumination?
In Soviet Washington the swamp drains you.
there's still so many Space Nutters given that science and math pretty trivially demonstrate that none of the sci-fi fantasies will ever happen?
Comment removed based on user account deletion
So, has he figured out why Amber Heard really dumped him?
But not in the world of people. Remember what matters mostly when you are interviewed for a job? Soft skills. The list of things that really matter in life, when living among human beings, is long and doesn't include any scientific method.
But there is one thing people can try to develop and inculcate, as they rise in the hierarchy. Try to find the subordinates who disagree with you, and try not to punish them for it and try to develop an atmosphere your subordinates feel comfortable in challenging you, seriously, even after they know what you deeply believe. If enough counter arguments flow into your thought process and mind set you are less likely to be led down the garden path by a bunch of yes men, sycophants brown nosing their way up the ladder.
sed -e 's/Chuck Norris/Rajnikant/g' joke > fact
@orlanz: "People need to get from point A to B."
Goods and services need to get from A to B. Given the cost on the environment, moving people from A to B is something we're going to look on as a a luxury.
Wow. Elon should start a church.
Church of Christ Science, or something like that.
He could be their leader.
I like it, but no one really has the time to do all the work themselves.
Yet Elon persists...
The problem with people like Musk is no common sense. Just because your very smart at one thing, doesn't mean your smart at everything. Musk fails at mass production, has no clue that creating a mass production product first gives you leverage for more unique products. Musk did Tesla ass backwards in business development. The Model 3 could very well end up a Delorean of EV cars.
He has rigor, peer review, refutational intent, and falsification or confirmation (i.e. proof of falsehood but never proof of truth). It's better than most people. And he is correct about most people's tendency to not change their thinking. The firmly a belief is held, the more one tends to accept all new information as it fits in with the existing belief... often snowballing one's confidence in a falsehood.
Elon Musk is a celebrity, not a scientist. So far, his wealth has depended on getting investors to give him money for unprofitable businesses. Ponzi had a similar business model he thought was successful. The real test of Elon Musk will be if we have an economic downturn and he can't simply sustain the cash flow by attracting more investments.
The converse of that is when they apply 'arts' to something that shouldn't. I saw a dental office offering 'Dental Arts'. Really? Who wants somebody to do a Francis Bacon inside of their mouth?
love is just extroverted narcissism
I was just talking to a few union folks (bus drivers and truck drivers) the other day about the fears they have with the driver-less vehicle revolution. In every case, their first reaction was that "the automation must be stopped"! As soon as I argued that there's just no historical case of people successfully putting those genies back into bottles once new technologies emerge, they quieted down and seemed to listen.
The thing is? I don't have any definite solutions for all the disruptive job loss it will create. But I do know that trying to fight the future is a losing strategy.
The worst case scenarios around all of this stuff seem to hinge on a few big corporations owning the entire thing, though. People don't want to live in a future where "Megacorp, Inc" owns every single self-driving car or truck and now collects 100% of all additional profits to be made by eliminating human labor as a cost of doing business.
Perhaps one way to prevent that is by passing legislation that says any business that operates with automation completely replacing human labor as part of the business model MUST be treated as a co-op? If a trucking company uses all driver-less trucks, require it be structured so each business wishing to use its services does so by buying into it with a fraction of ownership.
It may not stop the truck drivers from losing their jobs driving trucks ... but at least it ensures the extra wealth generated by taking human employees out of the equation gets spread around - benefiting dozens, hundreds or even thousands of businesses who use the services. That, in turn, means greater opportunities for a variety of other new jobs to pop up.
Really smart people are much better at rationalizing.
It's partially a problem that they are right a lot. Over 80%- maybe over 90%. When they are wrong- they just can't see it.
It sounds like Musk actively tries to avoid this. Good for him.
She was like chocolate when she drank... semi-sweet at first and then increasingly bitter.
Well, I don't know who did it but I suspect they may have applied the pseudo-scientific method the article outlines to decide that it was a good idea.
It sounds like Musk has just read Karl Popper's Wikipedia page. Doesn't he know that anyone with basic knowledge about the scientific method has most probably already read that? It just makes him sound stupid or like he's got so used to everyone around him kissing his arse and telling him he's so smart that he's forgotten how to self edit.
Better iterative science: ...etc
1 Predict
2 measure
3 compare
4 extend-prediction
5 new less-noisy measurement
6 clever compare
What Musk said is a pretty accurate sound bite method.
It takes a lot of words to actually lay out the scientific method. No one would have quoted the text below if Musk had taken the time to say them...
From school for dragons...
Home > How to Train Your Dragon > The Scientific Method > Scientific Method Steps
Scientific Method Steps
The âscientific methodâ(TM) merely refers to a broad framework for studying and learning more about the world around us in a scientific manner. It is not so much a series of absolute, unchangeable steps as a guideline to the method that must be used when trying to reach a scientifically acceptable theory about a subject matter. Therefore, it is not possible to provide a finite number of steps or an exact procedure for following the scientific method. However, the scientific method steps detailed below describe the main steps that scientists commonly take when conducting a scientific inquiry.
Steps of the Scientific Method
Make an Observation
Scientists are naturally curious about the world. While many people may pass by a curious phenomenon without sparing much thought for it, a scientific mind will take note of it as something worth further thought and investigation.
Form a Question
After making an interesting observation, a scientific mind itches to find out more about it. This is in fact a natural phenomenon. If you have ever wondered why or how something occurs, you have been listening to the scientist in you. In the scientific method, a question converts general wonder and interest to a channelled line of thinking and inquiry.
Form a Hypothesis
A hypothesis is an informed guess as to the possible answer of the question. The hypothesis may be formed as soon as the question is posed, or it may require a great deal of background research and inquiry. The purpose of the hypothesis is not to arrive at the perfect answer to the question but to provide a direction to further scientific investigation.
Conduct an Experiment
Once a hypothesis has been formed, it must be tested. This is done by conducting a carefully designed and controlled experiment. The experiment is one of the most important steps in the scientific method, as it is used to prove a hypothesis right or wrong, and to formulate scientific theories. In order to be accepted as scientific proof for a theory, an experiment must meet certain conditions â" it must be controlled, i.e. it must test a single variable by keeping all other variables under control. The experiment must also be reproducible so that it can be tested for errors.
Analyse the Data and Draw a Conclusion
As the experiment is conducted, it is important to note down the results. In any experiment, it is necessary to conduct several trials to ensure that the results are constant. The experimenter then analyses all the data and uses it to draw a conclusion regarding the strength of the hypothesis. If the data proves the hypothesis correct, the original question is answered. On the other hand, if the data disproves the hypothesis, the scientific inquiry continues by doing research to form a new hypothesis and then conducting an experiment to test it. This process goes on until a hypothesis can be proven correct by a scientific experiment.
The whole process is collaborative and is conducted in a clearly documented manner to help other scientists who are doing research in the same field. Throughout history, there are instances where scientists have stopped their research before completing all the steps of the scientific method, only to have the inquiry taken up and solved by another scientist interested in answering the same question.
She was like chocolate when she drank... semi-sweet at first and then increasingly bitter.
[citation needed] Did Elon learn anything from his BA in physics? There's this thing already called the scientific method that you're supposed to learn. Now he'll get credited with inventing that, like Al Gore did for inventing the internet. The interwebs are definitely living in a post-fact bubble.
Debate is a form of harassment. Do not question my truth.
Am I the only one who find's Musk's version of scientism worryingly naive?
Musk says other folks "engage in wishful thinking. They ignore counterarguments. They form conclusions based on what others are doing and aren't doing." But then in advocating for his version of the scientific method, he repeats these same errors. He talks about "truth", "probability", "axioms", "correctness", "objectivity", and what's "right," sliding casually between epistemology, naturalism, metaphysics, and ethics, but without pausing to define terms, examine what is given and what has presupposed, or consider the counterarguments those presuppositions have already silenced... surely a case of wishful thinking.
Musk identifies AI as the "biggest threat that humanity faces this century," AI is a product of "the scientific method" he advocates. So how to reconcile the poison and the cure? Shouldn't folks like Musk cogitate on that with a bit more critical depth, before spouting stuff like "It's really helpful for figuring out the tricky things," which sounds like a page from Zuckerberg's playbook, before Russia and Fake News turned up and proved that the world is a lot more complex than that.
Maybe my sense of ire is raised because I've been spending time reading Alfred Whitehead's books (Science and the Modern World, The Concept of Nature, Process and Reality). Whitehead was the co-author with Bertrand Russell of Principia Mathematica, so a major figure in logic. His discussion of the narrow-band thinking that takes place under the name of the scientific method is almost 100 years old, but still just as damning. In Whitehead's view, inductive and axiomatic methods are specialist modes of thought, useful in certain parts of science, but easily prone to fallacies and dogmatism in wider discussions. To pull up a random quote:
"In its use of [methods of induction] natural science has shown a curious mixture of rationalism and irrationalism. Its prevalent tone of thought has been ardently rationalistic within its own borders, and dogmatically irrational beyond those borders. In practice such an attitude tends to become a dogmatic denial that there are any factors in the world not fully expressible in terms of its own primary notions devoid of further generalization. Such a denial is the self-denial of thought."
Here's hoping that industry figures like Zuckerberg and Musk find a way to up their game - I don't expect them to be Whitehead readers, but can we at least move away from disappointing and disingenuous dogma?
A rising tide only lifts boats that aren't anchored to something. Those boats just get pulled under water. And most people are anchored by a whole lot of different things.
The theoretical argument about "efficiency" has never actually been tested because the people making the decisions aren't considering what is best for everyone, only themselves. In order to "outsource" you need capital and that has to be produced by taking some of the productivity of workers as profit. Then, instead of investing it in making those workers more productive, it is invested in China making Chinese workers more productive. This is good for the Chinese workers, good for the person who owns the capital, its not good for the people who created the capital. Especially if they are anchored by the value they place on their home, family and friends.
What's wrong here is the belief that money is the only value and that people don't care more about their neighbors than they do about people on the other side of the world.
By coincidence, I just watched this a couple of days ago:
Feynman: 'Greek' versus 'Babylonian' mathematics
This excerpt is possibly from the 1964 Messenger Lecture series at Cornell University, collectively titled The Character of Physical Law.
"The method of starting with the axioms is not efficient."
AKA "axioms are overrated".
Musk didn't invent Paypal, he founded a company that merged with the company that created Paypal. He ended up owning a lot of the stock and made a lot of money when Paypal was sold to Ebay.
If a large vacuum tunnel suddenly ruptured the in rushing air would strike passengers with a force of 15 lbs/sqin. On an average person's chest that would be 6,000 lbs of force, crushing them.
Running with Linux for over 20 years!
Elon Musk's scientific theory is:
1. Ask a question
2. Hire a lobbyist
3. Secure government funding
4. Hire another lobbyist
5. More taxpayer funding.
I always find it ironic when scientists say they don't believe there is a God when there is an overwhelming amount of evidence from physics that there is one.
"Question Everything." -- Elon Musk
In a world of the blind, the one-eyed man is king--and the two-eyed man is a heretic.
... obvious things now? Just because they are said by poster boys?
I do not believe in karma. "Funny"=-6. Do good and forbid evil. Yours, Oft-Offtopic Flamebaiting Troll.
Most conservatives have the underlying assumption (probably correct) that liberals are out to either control them, take their money, or otherwise harm their bottom line. If they cooperate with liberals on climate matters, then people with no connection to them (or far less connection) will be making decisions that impact them very personally, and they WILL take advantage of that far beyond what can be justified. Therefore, it's not a matter of science in politics, but a matter of trust.
> If we are trying something new how would we even have a probability of it being correct? Isn't that point of experimentation - to figure it out?
I've never tried knocking this computer off of this table. We can, however, make some reasonable predictions about what would likely happen based on some axioms:
Things knocked off tables tend to fall down
Tile floors are typically hard
Delicate electronics tend to break about 70% of the time.when they hit hard surfaces
Based on what we know, we can predict that there is about a 70% chance that this computer would break if I knocked it off this table.
Using other axioms, we can further say that we'll almost certainly break a computer if we knock 20 randomly chosen computers off of 20 randomly chosen tables.
That's why he said "gather all the evidence you can, develop axioms based on the evidence, then examine if the axioms are likely true, and how likely". That way, when you spend $20 million on an experiment you can use the experiment to test something that will probably work.
If it looks like a duck, swims like a duck, and quacks like a duck, then it probably is a duck.
If it only looks a bit like a duck and swims like a duck, it might well be a young swan. But if it matches all three characteristics *strongly*, it's PROBABLY a duck. (But not certainly).
You did a great job of quoting just part of one of the three definitions Merriam Webster's gives for "axiom".
> If you have to assign a probability of truth to it, then it's not an axiom. You accept it or not, but you don't accept it with some probability.
I see you studied Socratic reasoning; there have been a few advances in the last 2,400 years. Over 99% of all email sent is spam. 99.99% of spam is identified and automatically filtered using reasoning like this:
Emails mentioning V1agra are spam with 98% likelihood .gov servers are spam with 7% likelihood (93% likely not spam)
Emails mentioning "win" or "free" are spam with 92% likelihood
Emails mentioning the name of the product you sell are only 1% likely to be spam
Emails from registered
Emails that make it to this filter in the first place are 30% likely to be spam
From those rules, one can calculate the total probability using something called Bayes formula. It's also called "Bayesian probability". With the formula, you can determine the probability if it includes both the word "free" AND the name of your product. It works rather well. It's used all over the place these days. When you log into a porn site, most sites do that calculation to see if you're likely to be using a stolen shared password or just guessing passwords to see if you get one that works. It's based on each part of your IP address (whether you're on the same IP or same network as the last few times that user logged in) , your browser and installed plugins, and other bits of data the server can detect.
An advanced application of this that is in the headlines lately is self-driving cars. The car doesn't know for sure that what it is seeing is a stop sign. It sees something reddish (might be a stop sign), it seems to be on the side of the road (might be a sign), seems to be about 8 feet tall (might be a sign), is roughly round or perhaps an octagon (could be a stop sign), has some whitish marks on it ...
It figures the thing is probably a stop sign if it has several characteristics strongly similar to stop signs, and few characteristics that are strongly different from stop signs.
Clearly Elon simply mistakenly used the word "axiom" when he actually meant "hypothesis" -- and this is clear by the way he *used* the word.
Thought this was pretty funny -- I recently (re-) read "I Think You Must Be Joking", where Feynman described his problem-solving approach:
1. Write down problem
2. Think real hard
3. Write down solution
Do elementary schools even teach the "Scientific Method" any more? Or, is that now considered racist?
APK Hosts File Engine 9.0++ SR-7 32/64-bit https://www.google.com/search?hl=en&source=hp&biw=&bih=&q=%22APK+Hosts+File+Engine%22+and+%22start64%22&btnG=Google+Search&gbv=1/
Ads/script/malware rob speed/security/privacy/bandwidth.
Hosts add speed (via hardcodes/adblocks), security (vs. bad sites/malware/poisoned dns), reliability (vs. dns down), & anonymity (vs. dns requestlogs/trackers).
Less power/cpu/ram + IO use vs. DNS/routers/addons/antivirus + less security bugs/complexity & faster vs. addons/routers/remote dns!
Avoids DNSChangers in routers/IP settings & dns redirect (99++% of ISP DNS != patched vs. it) + DNS tracking & lighten DNS load & resolve faster from local RAM!
* Via what u NATIVELY have in a FASTER kernelmode IP stack (doing more w/ less).
APK
P.S. - Safe https://www.virustotal.com/en/file/e01211ca36aa02e923f20adee0a3c4f5d5187dc65bdf1c997b3da3c2b0745425/analysis/1433430542/ (self checking code vs. infection of it built-in)
This is what the scientific method has been for the last couple hundred years. It most certainly doesn't "belong" to someone who's just pointing that out, as any teacher does.
People love to recite the bit that warned about a "military industrial complex", which was a warning that has certainly come true particularly in the post-cold-war era of mergers between defense contractors, but they ignore the other half of his wise warning:
"Akin to, and largely responsible for the sweeping changes in our industrial-military posture, has been the technological revolution during recent decades.
In this revolution, research has become central; it also becomes more formalized, complex, and costly. A steadily increasing share is conducted for, by, or at the direction of, the Federal government. Today, the solitary inventor, tinkering in his shop, has been over shadowed by task forces of scientists in laboratories and testing fields. In the same fashion, the free university, historically the fountainhead of free ideas and scientific discovery, has experienced a revolution in the conduct of research. Partly because of the huge costs involved, a government contract becomes virtually a substitute for intellectual curiosity. For every old blackboard there are now hundreds of new electronic computers. The prospect of domination of the nation's scholars by Federal employment, project allocations, and the power of money is ever present and is gravely to be regarded. Yet, in holding scientific research and discovery in respect, as we should, we must also be alert to the equal and opposite danger that public policy could itself be come the captive of a scientific-technological elite." - Dwight D. Eisenhower, 1961
He was warning that science was becoming increasingly intertwined with government because of the necessary involvement of government funding and that just as in the defense contractor case, this could cause problems in both directions with government influencing the science and those funded scientists gaining policy influence - that policies could be driven under the slogan "SCIENCE SAYS!" just as other policies could be driven by the slogan "Boeing/Lockheed/the Pentagon SAYS!" and that these trends needed to be kept in check.
We have arrived at a point where huge slices of public policy are being influenced by cries of SCIENCE! and arguments about public policy in vast areas of society are being affected by people tossing about claims of science and the scientific method, often where science and the scientific method are not applicable, or are being applied improperly.
It is the task of statesmanship to mold, to balance, and to integrate these and other forces, new and old, within the principles of our democratic system-ever aiming toward the supreme goals of our free society.
that last line was also Eisenhower and should have been in italics inside the quotes but for some reason it pasted oddly
This is highly ironic, given Elon musk's completely baseless claims that humanity is facing an existential risk from the threat of AI superintelligence. Literally all that we are building today, even using deep learning, amounts to nothing more than fancy statistical regression.
These are not axioms, they are indicators.
If there is an axiom in your example, its the one that makes bayesian filtering work - with limits on timing and % effectiveness given that the opponent is a creative human and our dataset only grows when false negatives get through.
If you accept that it works, you don't assign a probability to the truth of the axiom. It's meaningless to do that. You do the experiment, you find out it does, and now the "probability of truth" is 1 because you proved it, at least under certain conditions.
If you were there when that particular invention was conceived, you might have a level of confidence about the result but you wouldnt be able to assign a probability of the "truth" of it. Ironic that you brought up spams email filtering because... you wouldn't have the training data to come up with any number.
When faced with something new, we cannot know the probability of it being true. We can only assign probabilities to something that we experience enough to create a model of it, like a fair coin toss.
So Elon Musk rehashes some old ideas of Karl Popper and this is revolutionary and newsworthy?