I've heard the argument that making it easier to quickly understand something reduces long term understanding. That's probably true if it reduces the time spent thinking about the subject, but I hope that by making the visualization process easier there is more time for reinforcement through example problems.
This is awesome work! Is the source code / build files available? Is there a specified license under which this work is published? I would love to link to this from (and incorporate some of it in) a STEM outreach project of mine: spinwearables.com/book which also tries to incorporate exolorables in the educational materials.
The code is open source and hosted on github. It's free to use under the GNU General Public License v3.0. If you have any questions about implementing the code feel free to ask me.
Fwiw, apropos the friction page[0], yesterdayish I saw a recent Friction video[1], which told a causation story of transient bonding, which prompted me to dig up a talk[2] by Mark Robbins (Johns Hopkins, but recently deceased). I saw a variant of it a year or few back at Harvard, and IIRC, he suggests Phys101-flavor Amontons'-laws friction was dominated by interfacial debris mechanics. Rather than by boding or direct interference. He told a fun story of a historical experiment replica, which gave unexpected results, until taken to a K-12 outreach event, where finger grease was helpfully added. Amontons' laws of large objects sliding on pig fat.
There's often a large gap between education content and research understanding. The pipeline flows neither smooth nor fast. Even simple things, like the color of the Sun, can be pervasively wrong. Let alone hard things, like friction. And there's almost no use of deep research understanding to craft novel learning progressions. If anyone has heard of efforts with this kind of focus, I'd appreciate hearing of them - thanks!
Addendum: re "Examples of visible thermal radiation: [...] the yellow part of fire (not the blue)", my understanding is this is another common misconception, and that candle yellow, while having a spectra shaped similar to blackbody radiation, is actually the mess of soot C-C bond emission spectra. It's way too cool to be blackbody - you'd have to be able to put a bit of metal in the flame, and have it too glow as bright and white-rather-than-red. Err, well, just since we're here, re "thermal emission [...] hottest to coldest: blue, white, yellow,", the Planck curve doesn't go through yellow: blue-white-red. It's a quick way to see if software is getting its chromaticity math wrong, though that happily seems less common now than it was a few years ago.
It's always funny/annoying when a teacher confidently presents misinformation. I still remember my biology teacher told me deoxygenated blood was actually blue. (it's dark red)
I'll review those slides you linked. Teaching friction is "rough"...
Wikipedia is still going with black body radiation as the source of the yellow color of fire. Maybe I should find better sources, although normal wikipedia is great for physics. Does anyone have any sources where I could learn more about the color of flame?
> Wikipedia is still going with black body radiation as the source of the yellow color of fire.
Wow. The flame section is pretty terrible and most claims aren't cited (I'm pretty sure many of them are also wrong, which hopefully makes citations more difficult to come by).
GP's point about heating metal to an equivalent temperature (and thus color) is correct AFAIK. Interestingly enough Wikipedia actually has an annotated experimental spectrum of a blue flame elsewhere. (https://en.wikipedia.org/wiki/File:Spectrum_of_blue_flame.pn...)
> Maybe I should find better sources, although normal wikipedia is great for physics.
It's indispensable as a time saving overview and index to primary sources. As with any encyclopedia (any non-expert body of text really) you can't trust every last detail to be correct.
Yeah, deoxygenated blue blood is a famously common misconception. That horrible and pervasive graphic, reinforced by veins looking blue through skin. Science education content fails both teachers and students, far worse than is often appreciated. I've wondered if it might be pedagogically useful to tell students that, or not.
I wish I had a better friction talk or paper for you. I don't recall seeing that video, but the version of the talk I saw, I recall as benefiting from the audience discussion. Sorry, it's what I had at hand. At some point, perhaps already, we'll get a nice survey paper, which can then be repackaged for teachers and students.
Wikipedia seems to not do well with widespread misconceptions. Perhaps if it had an additional layer of Backstory/WritersNotebook, in addition to Article and Talk. Or at least a similar convention on using Talk. As it its, past talk gets buried, and people come along and tweak things to better match their misconceptions, or some buggy source, so correctness isn't stable.
The google snippet of [1] looks hopeful... but isn't available on sci-hub - perhaps email the author? [2] is eh. I didn't quickly find nice researchy sources. For soot details (eg, I remember it as C-C bonds, but maybe not), there's ongoing work on using soot luminescence as a light source, which might well have good details buried somewhere, but I didn't quickly find them.
If anyone finds anything nice, I too would be interested.
Hmm, interesting. And a "Misconceptions" section might also help immunize readers too. I wonder if the WP community would find that acceptable? K-12 teaching has been increasingly dealing explicitly with misconceptions, so at some point it might become unremarkable. I'm afraid a sentence or paragraph in isolation, would eventually be edited out.
Checking my personal touchstones, WP currently has the color of the Sun correct, yay. Though there's a bogus " spectral class. As such, it is informally and not completely accurately referred to as a yellow dwarf (its light is closer to white than yellow).", misunderstanding what spectral class "yellow" means, and then creatively flailing to reconcile that. On Stellar_classification, the Vega-relative chromaticity background colors look wrong, which perhaps doesn't help.
One challenge is handling incorrect graphics like [1] (that Planck curve isn't going through white, either from broken code, which used to be common, or there used to be some incorrect math on the web). The graphic is used in many articles. But there have been several such over the years, correct and incorrect. So adding comments to the graphic won't help long-term. I suppose a Misconceptions section could be added to the Chromaticity article, in the hope that the next person to swap in a pretty but incorrect chromaticity diagram, might notice that, as they haven't noticed the associated Talk. Might work? Or, hmm, setting up some later reader to recognize the problem, and giving them the confidence to fix it? Focus on facilitating repair, rather than on prevention? Intriguing.
Correctness seems socially hard. I saw an high-profile OER intro astronomy text, that was obviously written thinking the Sun was yellow. All the graphics have it yellow. But text sections dealing with color, were seemingly edited just enough, that through ambiguity and incoherence, the authors could say "we didn't say that - we don't have it wrong", even if there's no chance students were surviving that minefield intact. As was pointed out to them, on their nice public errata ticket system, and closed as WONTFIX. An important part of why science works, is people fearing being embarrassed in front of their colleagues, but for reasons I don't understand, problems like this don't seem to trigger that, so even long-term, they don't get fixed. If one could figure out how to dial up the embarrassment, that might have a broad non-trival impact on the quality of science education content.
I appreciate that use of KaTeX allows for faster Math rendering. Since developing intuition about problems can be increased by seeing multiple examples, can you leverage this speed to generate a specific example from a template and some sample values for a variable each time you interact with a student ?
It seems part of the trick is generating good collections of interacting values that are easy to present such as all rational numbers or all integers.
Aren’t these related to a Hamiltonian circuit of a elliptical values ?
I have a mix of premade examples that try to help students work around a common mistake, and random problems with generated solutions in KaTex math notation. Although, my experience is that the randomized problems don't seem to be as valuable as the hand crafted ones.
I don't think I fully understand the "Hamiltonian circuit of elliptical values" question.
An electron density representation is easier to visualize. And far less epicycle-ly than orbitals. And easier to avoid error and artifact ("layer after layer", "electron pairs", etc). Chemistry education content is notable for leaving both students and teachers steeped in misconceptions.
Though how/whether it could be used to teach orgo is an open question. I've only seen a couple of isolated attempts. Needing new content, and non-trivial software, with isolated professors, has made progress unlikely, even if possible.
I've an barely-started lockdown project, using ASE/GPAW[1] to precompute trajectories and densities, for an AR browser interactive - playfully tugging on atoms blobbing as small molecules, realistically.
Imagine we taught elephants the way we teach atoms. Here's an early medieval drawing that looks... yikes (they're actually not that bad, so the analogy isn't great). Here's a table model. And a hose model. And an orbiting cell model that has everyone imagining breakdancing jellyfish. But nothing with the slightest resemblance to an elephant photo or toy. Let alone a realistic interactive.
If anyone knows of efforts to do better, I'd appreciate hearing of them - thanks!
I don’t know about you, but reading that first paragraph of the organic chemistry textbook was not a drag. It may not have been Carl Sagan’s Cosmos but I don’t understand how any generally curious person wouldn’t find it interesting. For the record I never took anything beyond Chemistry I in college.
https://landgreen.github.io/physics/index.html
I've heard the argument that making it easier to quickly understand something reduces long term understanding. That's probably true if it reduces the time spent thinking about the subject, but I hope that by making the visualization process easier there is more time for reinforcement through example problems.