4 ms·

Evidence is not the same as proof. Evidence can often be interpreted in different directions. As more and more evidence accumulates about some topic, we can us

Evidence is not the same as proof. Evidence can often be interpreted in different directions. As more and more evidence accumulates about some topic, we can use a Bayesian process to continously update our priors, but it is common to disagree about what constitutes enough evidence to conclude. This is partly due to us having different priors, but also often due to our tendency to not trust evidence that is in conflict to our priors. For both of these reasons, sufficient evidence for a new proposition for it to be logically more likely than the pre-existing consensus typically exist for quite a while before it becomes the new consensus. In many cases, what is required for the paradigm shift to eventually happen, is not so much more evidence, but rather for the old guard to retire. In the interrim, it can be 'obnoxiously obvious' for some that the evidence is already there.

pvg 4y ago
If someone says something is 'obnoxiously obvious', nobody needs to retire for them to explain what they mean by that.
- trashtester 4y ago
  In the late 80s, I (a teenager at the time) was discussing the topic of airplanes being flown by computers. I was convinced that computers would be able to fly airplanes within my lifetime, and would even replace pilots in figher planes at some point. For me, this was 'obnoxiously obvious'. The first time, I discussed it with an engineer/hobbyist pilot. He claimed that flying an airplane was an impossible task for a computer. His argument was from chaos theory; a computer would never be able to calculate all the variables involved in turbulent airflow, so they would not be able to fly an airplane. Clearly, he ignored that humans do not do those calculations either, but I was unable to make my point. The second time, it was a former figher pilot, then airline pilot. He did not in principle object to a computer flying an airplane, but he completely dismissed the possibility of a computer flying a fighter plane into combat. Both are now retired. I still think it is obnoxiously obvious that computers can fly planes, and that within my lifetime, even fighters will be unmanned. Clearly, there is more evidence now, but I think that today, the evidence from the 80s would have been considered sufficient by most people my age or younger. (Now, I'm almost 50, and maybe this time I'm the one being obnoxious, as indicated by my previous post being downvoted.)
  - pvg 4y ago
    There was plenty of evidence for this in the late 80s, from cruise missiles to the space shuttle and many other systems. It's just that 80s arguments didn't have the benefit of the internet and the web in everyone's pocket. The thing we're dealing with here is the extraordinary-sounding claim that it was just about biochemically impossible for pre-historic hunter-gatherer societies to have carved some bigass stones. The person making it seems to have the boon of internet access to help them buttress their argument.
    - trashtester 4y ago
      > There was plenty of evidence for this in the late 80s, from cruise missiles to the space shuttle and many other systems. It's just that 80s arguments didn't have the benefit of the internet and the web in everyone's pocket. I would agree, but that was not even the evidence that I (a 13 year old kid at the time) was using to draw my conclusion. In my opinion then (and still) the main argument was that the brain is not magical, and provided that Moore's law would continue, computers would eventually be able to do most of the things a human brain is doing. While it might take some time for computers to do all things as well as the brain, computers would have advantages in terms of cost, space requirements and not being restricted to 9g (for fighters). I think the 40 year engineer and 50 year old former fighter pilot had a model of the mind that were dualist, ie there was some ghost in the machine that allowed humans to do things that computer never will be able to. Many famous thinkers have had similar ideas, such as Gödel, von Neumann and Penrose, all extremely intelligent. Now, the consensus has shifted, but Penrose has not changed his mind. Recent developments in AI may be seen as pretty hard evidence that computers one day will be able to do what the brain is doing, but Penrose doesn't think the evidence is strong enough, and still looks for his ghost in the machine in quantum effects within neurons. As for the original discussion, I actually don't have a strong opinion. But when someone claims that something is 'obnoxiously obvious', I presume good faith, and that evidence has already been presented, but not yet accepted by the community. Maybe I was unfair in my first comment in this thread. Generally, strong claims do require strong evidence. But it IS pretty common for the Old Guard to require the evidence for a new idea to meet much higher standards than the evidence for the old idea, and if so, it is typically the case that one just has to wait and see what will become the consensus over time (often more than 1 generation).

pvg 4y ago

If someone says something is 'obnoxiously obvious', nobody needs to retire for them to explain what they mean by that.

trashtester 4y ago
In the late 80s, I (a teenager at the time) was discussing the topic of airplanes being flown by computers. I was convinced that computers would be able to fly airplanes within my lifetime, and would even replace pilots in figher planes at some point. For me, this was 'obnoxiously obvious'. The first time, I discussed it with an engineer/hobbyist pilot. He claimed that flying an airplane was an impossible task for a computer. His argument was from chaos theory; a computer would never be able to calculate all the variables involved in turbulent airflow, so they would not be able to fly an airplane. Clearly, he ignored that humans do not do those calculations either, but I was unable to make my point. The second time, it was a former figher pilot, then airline pilot. He did not in principle object to a computer flying an airplane, but he completely dismissed the possibility of a computer flying a fighter plane into combat. Both are now retired. I still think it is obnoxiously obvious that computers can fly planes, and that within my lifetime, even fighters will be unmanned. Clearly, there is more evidence now, but I think that today, the evidence from the 80s would have been considered sufficient by most people my age or younger. (Now, I'm almost 50, and maybe this time I'm the one being obnoxious, as indicated by my previous post being downvoted.)
- pvg 4y ago
  There was plenty of evidence for this in the late 80s, from cruise missiles to the space shuttle and many other systems. It's just that 80s arguments didn't have the benefit of the internet and the web in everyone's pocket. The thing we're dealing with here is the extraordinary-sounding claim that it was just about biochemically impossible for pre-historic hunter-gatherer societies to have carved some bigass stones. The person making it seems to have the boon of internet access to help them buttress their argument.
  - trashtester 4y ago
    > There was plenty of evidence for this in the late 80s, from cruise missiles to the space shuttle and many other systems. It's just that 80s arguments didn't have the benefit of the internet and the web in everyone's pocket. I would agree, but that was not even the evidence that I (a 13 year old kid at the time) was using to draw my conclusion. In my opinion then (and still) the main argument was that the brain is not magical, and provided that Moore's law would continue, computers would eventually be able to do most of the things a human brain is doing. While it might take some time for computers to do all things as well as the brain, computers would have advantages in terms of cost, space requirements and not being restricted to 9g (for fighters). I think the 40 year engineer and 50 year old former fighter pilot had a model of the mind that were dualist, ie there was some ghost in the machine that allowed humans to do things that computer never will be able to. Many famous thinkers have had similar ideas, such as Gödel, von Neumann and Penrose, all extremely intelligent. Now, the consensus has shifted, but Penrose has not changed his mind. Recent developments in AI may be seen as pretty hard evidence that computers one day will be able to do what the brain is doing, but Penrose doesn't think the evidence is strong enough, and still looks for his ghost in the machine in quantum effects within neurons. As for the original discussion, I actually don't have a strong opinion. But when someone claims that something is 'obnoxiously obvious', I presume good faith, and that evidence has already been presented, but not yet accepted by the community. Maybe I was unfair in my first comment in this thread. Generally, strong claims do require strong evidence. But it IS pretty common for the Old Guard to require the evidence for a new idea to meet much higher standards than the evidence for the old idea, and if so, it is typically the case that one just has to wait and see what will become the consensus over time (often more than 1 generation).