How would you describe briefly your own mental modeling of chemistry?

[sbrothy]

Yes. I was just about to say that Mr. Distler wasn't the only one involved and that there was another side trying to make it work.

 

[sbrothy]

Also, this discussion is, in fact, veering into "Beyond Standard Model" land....

EDIT: The hijack was truly unintentional. Sorry.

 

[pellis]

@Mayhem @TensorCalculus @DaveE @berkeman @Borek @mjc123 @sbrothy @erobz @renormalize

To try to bring this discussion back on course, just briefly, I would appreciate any comments on what you think my answer (#36, Tuesday 3:07pm) omits or gets wrong.

I take mentions of a few points, especially kinetics (Mayhem). A few more may help me refocus my summary.

With thanks to all who've contributed so far - and anyone else tempted to do so in the future.

 

[Muu9]

There are a few metaphors/mental models/mnemonics I like to use for general chemistry, like thinking of chemical energy minimization as being similar to gravitational energy minimization. Another example is how electronegative atoms are "greedy", so they "hog" the electrons, leading to their region of the molecule being more negatively charged. It reminds me of when I was much younger and thought of arithmetic with positive and negative numbers as good guys vs bad guys...

But in my opinion, this isn't what chemistry is "about" any more than math is about the alligator always eating the bigger number.

 

[sbrothy]

I can relate to the visual understanding. My brain works the same way. In fact, if I'm unsure how to spell a word I can usually write different versions of it down and my brain will somehow recognize the correct one.

I assume you're mostly concerned with organic chemistry. Not that it makes a lot of difference for a fundamental understanding. But I think we can all agree that the carbon element is special as basis for life as we know it, and for it's willingness to interact in so many ways with so many other elements.

Electron bonding and the electron shell picture is still the way I picture chemical syntheses, then again my chemistry lessons were more than 20 years ago! The "shells" have become much more diffuse since Niels Bohr's simplified picture was overtaken by the quantum one - eh now I'm going there anyway - but I'm not sure it really means something for everyday chemistry. It's a little like Newton is adequate for most of the day to day physics needed, relativistic effects only becoming necessary in extreme cases.

Thermodynamics was mentioned (and of course no matter what you study there's no way around it). First time I experienced an endothermic reaction was really fascinating. As are catalysts.

There are so many rules and exceptions to them as only nature can do. And that is before we get into quantum chemistry! Hypervalence, electron deficiency, stereochemistry... and spectroscopy is so useful it almost reminds me of the babel fish from Hitchhiker's Guide to the Galaxy.

Richard Feynman, in his book What Do You Care What Other People Think?": Further Adventures of a Curious Character explains that he is a synesthete, (There's a small article which mentions it here: The Brain of a Synesthete). I'm sure it can be a gift as well as a curse, either useless or useful - perhaps downright debilitating!

I don't know if such a thing is possible with elements and molecules but it could be illuminating to examine how many gifted scientist benefitted from this syndrome.

EDIT: Yeah, and that's another thing. I doubt gravity will ever be relevant for everyday chemistry (possible on the nanoscale and lower) with electromagnetism being so much stronger.

 

[pellis]

There are a few metaphors/mental models/mnemonics I like to use for general chemistry, like thinking of chemical energy minimization as being similar to gravitational energy minimization. Another example is how electronegative atoms are "greedy", so they "hog" the electrons, leading to their region of the molecule being more negatively charged. It reminds me of when I was much younger and thought of arithmetic with positive and negative numbers as good guys vs bad guys...

But in my opinion, this isn't what chemistry is "about" any more than math is about the alligator always eating the bigger number.

@Muu9 Agree about electronegativity. As for energy minimisation, Peter W Atkins (well known text writer at Oxford U) reckons all energy is PE or KE, with one exception - if I remember correctly, he can't decide how to consider photons... But energy minimisation is central to quantum chemistry, at least for ground state properties.

 

[pellis]

I assume you're mostly concerned with organic chemistry.

Richard Feynman, in his book What Do You Care What Other People Think?": Further Adventures of a Curious Character explains that he is a synesthete, (There's a small article which mentions it here: The Brain of a Synesthete). I'm sure it can be a gift as well as a curse, either useless or useful - perhaps downright debilitating!

EDIT: Yeah, and that's another thing. I doubt gravity will ever be relevant for everyday chemistry (possible on the nanoscale and lower) with electromagnetism being so much stronger.

Organic chemistry is only a focus as it is the gap in my studies that limits my understanding of the bigger issues relating to the origin of life, but I think it's just as subject to the same points as the ones my summary covers.

I only mentioned gravity as I'd mentioned the em field, but in retrospect it really isn't directly relevant except when it pulls enough matter together to create metastable environments in which chemistry can happen; more relevant are the ElectroWeak and Strong Nuclear Force, which do impact chemistry via radioactivity/isotope decay etc. I'll have to think about whether to insert some references to them (as one other contributor here did earliuer).

Thanks for the ref to The Brain of a Synesthete. I used to keep a list of words in the colours I felt they were best represented by "Philip" was brown, "Maupertuisian" was pale blue, and "sushumna" was red - there were others but It was decades ago. Then there's the famour Stroop test https://en.wikipedia.org/wiki/Stroop_effect

 

[sbrothy]

As an aside I tried to search for "scientists with synesthesia". Wiki has a short list but surprisingly it looks like it's mostly artists. But if notes translates to colors I guess that makes sense. I can't really find any other scientists than Feynman.

Sorry. Carry on.

 

[pellis]

AI answers are prohibited on Physics Forums. Please cite actual published references that evaluate and discuss Lisi's proposal.

AGAINST: Distler & Garibaldi (2009, peer-reviewed), SABINE HOSSENFELDER https://backreaction.blogspot.com/2007/11/theoretically-simple-exception-of.html
ATTEMPTS TO ENGAGE: “The Plebanski action extended to a unification of gravity and Yang–Mills theory” by Lee Smolin (Physical Review D, 2007). AND Unification of gravity, gauge fields,and Higgs bosons, A. Garrett Lisia, Lee Smolin, and Simone Speziale, Journal of Physics A: Mathematical and Theoretical, 2010. (ATTACHED)

@renormalize @sbrothy @TensorCalculus In order to find those references, aside from my own files I could have found them in https://en.wikipedia.org/wiki/An_Exceptionally_Simple_Theory_of_Everything, and you likely wouldn't have objected.

Does the fact that I used the newly released ChatGPT5.0 to find the references actually contravene PF's rules? It was certainly more convenient and provided additional context that was reassuring (not copied into the refs listed).

I can understand that purists and those who see it as merely a token manipulator will be doubtful of its use. But I think we're all going to find, quite soon, that AI assistance will become more widely accepted, and is probably being used far more already than some people realise.

Given that humans also make mistakes, does openly-acknowledged use of AI assistance really make any difference, if used judiciously and checked carefully?

Note also that over a year ago (before learning that AI assisted answers were not allowed there either) I posted on Physcis Stack Exchange, as an experiment, an answer to the highly technical question:
https://physics.stackexchange.com/q...rs-under-vielbein-redefinitions/812257#812257
...which was upvoted and accepted, has been viewed over 100 times and never challenged beyond a request for clarification (also answered). It's more advanced theoretical./mathematical physics than I am familar with but I checked it as carefully as I could before posting it, having felt free to do so as no-one else had even tried to answer the question since it had been posted 5 days earlier.

 

[sbrothy]

I suspect it would have perfectly acceptable (I'm not a modertor) if you had also linked to the arXiv paper. But I may be wrong. I'm myself a little new to this AI stuff and I don't use it much.

 

[pellis]

I now use it even more than Wikipedia, and to be completey honest, and accepting its faults (very poor graphics, occasional freezing,, occasional loss of material, poor transfer of math formatting into Word...) nevertheless it makes a very useful tutor - unlike Wikipedia or Google searches, one can drill down on any point one doesn't understand and usually it will clarify.

 

[TensorCalculus]

@renormalize @sbrothy @TensorCalculus In order to find those references, aside from my own files I could have found them in https://en.wikipedia.org/wiki/An_Exceptionally_Simple_Theory_of_Everything, and you likely wouldn't have objected.

Does the fact that I used the newly released ChatGPT5.0 to find the references actually contravene PF's rules? It was certainly more convenient and provided additional context that was reassuring (not copied into the refs listed).

I can understand that purists and those who see it as merely a token manipulator will be doubtful of its use. But I think we're all going to find, quite soon, that AI assistance will become more widely accepted, and is probably being used far more already than some people realise.

Given that humans also make mistakes, does openly-acknowledged use of AI assistance really make any difference, if used judiciously and checked carefully?

Note also that over a year ago (before learning that AI assisted answers were not allowed there either) I posted on Physcis Stack Exchange, as an experiment, an answer to the highly technical question:
https://physics.stackexchange.com/q...rs-under-vielbein-redefinitions/812257#812257
...which was upvoted and accepted, has been viewed over 100 times and never challenged beyond a request for clarification (also answered). It's more advanced theoretical./mathematical physics than I am familar with but I checked it as carefully as I could before posting it, having felt free to do so as no-one else had even tried to answer the question since it had been posted 5 days earlier.

I mean, you did credit ChatGPT which is a good start: but ChatGPT can also be a source of a lot of error, particularly with physics, which is why AI outputs are generally not accepted on PF. Maybe in this case it was fine: but in another case, it might not be. Call it deontological if you will but if a rule applies in one circumstance on PF, it should apply in all circumstances. If someone replied to a question in the advanced physics homework help section with ChatGPT, you would also be a bit skeptical, would you not?
You could be fed misinformation and not even know it: I've seen people try and use ChatGPT for middle-school level physics homework and get half of it wrong. Just because it got something right once, doesn't mean it always gets it right...
The forum guidelines say:

• ChatGPT and AI-generated text
  • Posting AI-generated text without attribution is categorically disallowed and will lead to a warning and an eventual permanent ban with continued use.
  • Answering a science or math question with AI-generated text, even with attribution, is not allowed. AI-generated text apps like ChatGPT are not valid sources.
  • Threads about the technology and cultural impact behind AI like ChatGPT are allowed
  • Usage of AI-generated text output in entertainment threads in General Discussion with attribution is allowed

Finding references... don't know if that's fine or not... but IMO it's better to find sources yourself than to get an AI to find them for you. That way you have to apply your brain a bit more.
(Not that I'm a moderator, in fact I'm pretty new here myself, so maybe I shouldn't be doing this... but... yeah)

I now use it even more than Wikipedia, and to be completey honest, and accepting its faults (very poor graphics, occasional freezing,, occasional loss of material, poor transfer of math formatting into Word...) nevertheless it makes a very useful tutor - unlike Wikipedia or Google searches, one can drill down on any point one doesn't understand and usually it will clarify.

Of course ChatGPT has it's benefits, but sometimes, like with learning, it's better to do it the hard way... I often feel like ChatGPT spoon-feeding you everything prevents critical thinking and dulls the "struggle" to learn something...
I do use it from time to time though, and I used to use it a lot, I'll admit that. But not for physics, or maths, or Chemistry. Not to teach me.

 

[renormalize]

Does the fact that I used the newly released ChatGPT5.0 to find the references actually contravene PF's rules? It was certainly more convenient and provided additional context that was reassuring (not copied into the refs listed).

From the Physics Forum rules:

• ChatGPT and AI-generated text
  • Posting AI-generated text without attribution is categorically disallowed and will lead to a warning and an eventual permanent ban with continued use.
  • Answering a science or math question with AI-generated text, even with attribution, is not allowed. AI-generated text apps like ChatGPT are not valid sources.
  • Threads about the technology and cultural impact behind AI like ChatGPT are allowed
  • Usage of AI-generated text output in entertainment threads in General Discussion with attribution is allowed.

In my view, your post #47 clearly violated the second bullet point. So by all means, use your own search skills, AI or even tarot cards if you like, to find credible references that support your point. Then post those references and describe their contents in your own words.

 

[sbrothy]

I suspect it would have perfectly acceptable (I'm not a modertor) if you had also linked to the arXiv paper. But I may be wrong. I'm myself a little new to this AI stuff and I don't use it much.

From the Physics Forum rules:

• ChatGPT and AI-generated text
  • Posting AI-generated text without attribution is categorically disallowed and will lead to a warning and an eventual permanent ban with continued use.
  • Answering a science or math question with AI-generated text, even with attribution, is not allowed. AI-generated text apps like ChatGPT are not valid sources.
  • Threads about the technology and cultural impact behind AI like ChatGPT are allowed
  • Usage of AI-generated text output in entertainment threads in General Discussion with attribution is allowed.

In my view, your post #47 clearly violated the second bullet point. So by all means, use your own search skills, AI or even tarot cards if you like, to find credible references that support your point. Then post those references and describe their contents in your own words.

Oh yeah, and by the way: Linking to Wikipedia is heavily frowned upon.

So heavily in fact, that it triggers an actual browbeating. And watch out for @berkeman as he’s just been promoted to admin. He might just be lurking around, waiting for an opportunity like that.

 

[renormalize]

Oh yeah, and by the way: Linking to Wikipedia is heavily frowned upon.

Actually, I routinely post links to Wikipedia without issue. Have I just been lucky?

 

[TensorCalculus]

Oh yeah, and by the way: Linking to Wikipedia is heavily frowned upon.

So heavily in fact, that it triggers an actual browbeating. And watch out for @berkeman as he’s just been promoted to admin. He might just be lurking around, waiting for an opportunity like that.

Welp, thanks for telling everyone: now I know not to do that (not that I've ever needed to reference to Wikipedia before)

 

[berkeman]

Oh yeah, and by the way: Linking to Wikipedia is heavily frowned upon.

Not really, depending on the situation. For non-controversial subjects, Wikipedia is often a good initial reference (and most of the Wikipedia articles have links to peer-reviewed articles about the subject). For controversial technical subjects, Wikipedia alone will probably not be enough to be making any assertions here.

 

[sbrothy]

Yeah well I was exaggerating a little for humor (also why I didn’t qoute the rules. That is: if I can make you believe I’ve got them memorized). I’ve been known to get away with it too. It depends on what you link to like @berkeman basically said.

 

[sbrothy]

Although I'm pretty sure any link to viXra will get you a virtual beating.

EDIT: Possibly with the lead pipe without soft pellets.

 

[jedishrfu]

Since this thread has run its course, its a good time to close it.

Thank you all for contributing to it.

Jedi