4 ms·

>Feel free to share a link to some of his work that you find relevant https://www.trevormarshall.com/ https://www.trevormarshall.com/ should be a good _startin

>Feel free to share a link to some of his work that you find relevant https://www.trevormarshall.com/ https://www.trevormarshall.com/ should be a good _starting_ point. (Side note: don't you think experimental evidence should override all speculation of how things should work or not work? I believe that Prof TM has done some of those experiments, and referencing a fair amount of literature on the subject)

arghwhat 2y ago
> https://www.trevormarshall.com/ https://www.trevormarshall.com/ should be a good _starting_ point. That's not a starting point, that's just his web page. Works that were relevant would be e.g. specific papers on the topic that gained traction. > Side note: don't you think experimental evidence should override all speculation of how things should work or not work? Well, that depends. Once you have sufficient experiments from several parties that hold up to scrutiny and meta-analysis, then you have the basis for building a theory to describe it, and build further experiments to attempt to disprove it or its competing theories. At some point, the results and underlying theory becomes agreed upon knowledge - even if temporary. Before that, experiments are interesting anecdotes to pique interest and build further research on. Most results, however, fail to get that far as most experiments are based on flawed test methodologies with conclusions not following the data, often driven unintentionally by a selection bias for a theory one set out to prove. Science is chock full of non-reproducible work and wrong conclusions (remember the recent superconductor scandal? most of these never make the news), which is why the power of science (like all of human knowledge) must always be in numbers. Yes, this does sometimes lead to understudied subjects lacking conclusions as it requires extensive interest and funding from multiple parties. For example, the focus on "alternative medicine" has been to debunk its claims of resolving medical conditions (the majority being quackery predating even basic understanding of biology), but as a result alternative chronic pain management regimes within these fields which some individuals find temporary relief within lack any useful data - an area modern medicine handles poorly. But this cannot be used as argument to relax the requirement of numbers. So yeah, "depends". > Prof TM I can't help but read this as Prof™.

arghwhat 2y ago

> https://www.trevormarshall.com/ https://www.trevormarshall.com/ should be a good _starting_ point. That's not a starting point, that's just his web page. Works that were relevant would be e.g. specific papers on the topic that gained traction. > Side note: don't you think experimental evidence should override all speculation of how things should work or not work? Well, that depends. Once you have sufficient experiments from several parties that hold up to scrutiny and meta-analysis, then you have the basis for building a theory to describe it, and build further experiments to attempt to disprove it or its competing theories. At some point, the results and underlying theory becomes agreed upon knowledge - even if temporary. Before that, experiments are interesting anecdotes to pique interest and build further research on. Most results, however, fail to get that far as most experiments are based on flawed test methodologies with conclusions not following the data, often driven unintentionally by a selection bias for a theory one set out to prove. Science is chock full of non-reproducible work and wrong conclusions (remember the recent superconductor scandal? most of these never make the news), which is why the power of science (like all of human knowledge) must always be in numbers. Yes, this does sometimes lead to understudied subjects lacking conclusions as it requires extensive interest and funding from multiple parties. For example, the focus on "alternative medicine" has been to debunk its claims of resolving medical conditions (the majority being quackery predating even basic understanding of biology), but as a result alternative chronic pain management regimes within these fields which some individuals find temporary relief within lack any useful data - an area modern medicine handles poorly. But this cannot be used as argument to relax the requirement of numbers. So yeah, "depends". > Prof TM I can't help but read this as Prof™.