I have a question I hope someone here can answer. I am at something of a crossroads in my career and privileged enough to be able to pursue almost anything. When I tell people the big problems I'd like to make an impact on, it's education, specifically science education. I want more people to understand how the scientific method works and how we use it to gain more understanding. I have passively read about education (Montessori, Bruner, Piaget, Pappert), followed Alan Kay and the smalltalk story, and I was a math teacher for a couple of years. I also teach currently online. I can't help but think that there is more to using computers than reproducing old media, and that the code-debug cycle is a good way to refine our understanding. And so maybe there is something there, using programming and interactivity as a medium for exploration of ideas. However, I don't have much more of a vision than that. I was talking to my wife about it yesterday and came up with an interesting phrasing: I want to do what Montessori did with her learning materials but with computers as a medium. I would appreciate any and all links to people, conferences, reading material, etc., that are related to this idea. Just to be clear, I don't mean teaching kids to code. And I don't mean using software to support the current educational paradigm (e.g., Chromebooks in schools, etc.). People seem to automatically roll back to those two pits.

Changing Minds Computers, Learning, and Literacy. by Andrea diSessa is kinda continuation of Papert's ideas & Logo work. diSessa designed Boxer with the goals of being an environment conductive to "computational literacy", not just learning to code, but using code to explore and communicate ideas. Personally, I found the book does a curiously mixed job. • Me I'm obsessed with many language/env design choices in Boxer but the book actually doesn't cover that too deeply. • It does a not-bad job arguing computers could have a profound impact on society, not just as a tool but as a medium. • But when discussing examples of such impact, they're curiously confined to improving math&physics teaching, which is far short of transforming all of society 🙂 ◦ Well, these just happen to be the areas where they did actual research trying it with school kids.

I’d recommend having a look at Andy Matuschak’s work. Also worth a look: May-li Khoe’s work, especially from her time at Khan Academy. And, pretty popular amongst people in this forum and perhaps too obvious, but you didn’t mention him, so you might also want to take a look at the early presentations and essays of Bret Victor. Also somewhat related: Michael Nielsen.

Another not-exactly-but-maybe-tangential: Mark Guzdial has written prolifically on education research & expanding CS teaching to wider groups, with focus on what they hope to get out of it.

definitely worth reading Teaching Machines by Audrey Watters detailing the history of educational technology

Bret Victor's latest work with DynamicLand is along those lines of helping people understand complex systems

A nice example is DrGeo. It has a long history (Wikipedia says 16 years), so I suspect there is a lot to be learned from its evolution over time.

@Eric Normand Andy Matuschak has the most up to date ideas on imagining learning with today's tech. Worth checking out: https://www.youtube.com/live/AZDX1Vv0HYc?feature=shared Although, as I pointed out to Andy as well, the biggest factor on effective learning for me trends to be map-making for a domain. Today's AIs are a bit hyperfocused on verbal learning, so new ideas are still needed.

You might find our work on Wikifunctions interesting (CACM paper, website). I'd be happy to chat more about it. I can't really put my finger on it, on the exact phrasing, but our believe is that "functions are knowledge" and that making functions available to everyone and to allow everyone to share and collaborate on functions, can make a difference. But we don't have a good way to phrase that yet.

Thanks, everyone!

I use the scientific method everyday both in my work (software architect) and outside work when learning something new. It's simple and effective. However it's not visual -- "using programming and interactivity as a medium for exploration of ideas." For that I use category theory, more, ologs: https://dspace.mit.edu/handle/1721.1/70469 This combination helps me to have: A likely correct understanding of a problem / subject via the scientific method; the result is a structured, static document A provably correct model of my understanding using ologs; the result is an interactive web app. Since I'm in software the interactive olog editor / web app is equipped with a functor which generates code, tests and documentation. In other words I have a full circle: an open ended scientific method where I capture an ever expanding knowledge; a math layer which makes that knowledge correct and visual; then another layer which translates to / runs my knowledge on a medium / domain (software in my case) It would be interesting to see how this applies to other domains than software. The https://www.appliedcategorytheory.org/ community uses this approach on a couple of domains already.

I feel you. If we think if education as limited by the quality of educational content, then the question becomes how do you maximize content quality- and i can't think of better ways than interactive digital experiences. Ivan sent this to me so i pass it to you https://explorabl.es/

Do you specifically not want to use computers to teach kids? I imagine theres software, maybe video games, that let kids play scientist. Like a game that has a physics engine and you can do real physics experiments.

@Scott Antipa Algodoo is quite fun for motion, gravity, optics etc.

@Eric Normand - I have been designing edtech for all my life and captured some of my experience here - tell me if it helps!

Thanks!