Stephen Downes

The real question being considered here is I think buried in the second half of this article: "How much does it matter whether or not we can accurately observe, characterize, and validate the processes at work inside large reasoning models?" And the answer is: it depends. "If LRMs can supercharge mathematics research the way AlphaFold did for computational biology, this line of thinking goes, why not embrace them, idiosyncrasies and all, and just verify the results?" Which makes sense to me. Indeed, the idea of 'reasoning' as "arriving at a sound conclusion by linking together intermediate steps that logically follow from each other" might itself be wrong, an artifact of centuries of rationalization rather than rationale. "When an LRM produces a correct answer — along with pages of 'thoughts' showing how it got the result — intuition tells us that the two must be linked. It's hard to imagine that process and outcome may have little to do with each other." But intuition may well be wrong.

Read the first two sections of this post but ignore the rest. The first section points to limitations of AI use in "the old economic model" and suggests a new approach based on the question "What do you want to bring into the world?" The second section defines this new approach as 'the possibility economy', "a social and economic system in which value arises from expanding the range of actions that people and communities can imagine, evaluate, and realize." The old language of, say, innovation, priorizes outcome. The new language "involves using technology, knowledge, and coöperation to develop new purposes, services, institutions, and forms of action." In my old-school language I think of this as an economy (and pedagogy) of affordances. As I said, you can stop here; the remainder of the article just runs the concept through a number of topics in an uninspired and automatic way. 

Nobody cares how I learn and listen, but bear with me. One of the things I learned while learning French was how little of what I 'hear' I actually hear. When I am listening to someone speak, although all the sound goes into my ears, what I'm actually doing is 'hearing' only a few words (or word fragments) in each sentence and filling in the rest cognitively. It's natural in English, but of course in French I didn't have the vocabulary or skill to fill in the words I'm not actually hearing. So my focus has to be that much more intense, and the environment a lot more quiet, for me to comprehend what is being said. It makes me wonder how other people listen, and whether they 'hear' by hearing or some other process. "Hearing," as Ann says, "is not the same as listening." Anyhow, that's why when I give presentations I'm careful to add enough text to the slides to help people get past limitations in hearing. "Keeping the barriers to hearing, and the different ways that listening can be supported is one way to get to the possibility of learning in accessible pedagogy."

I was raised on phonics but early in life discovered there was a second way to read that did not involve sounding out words. This way, I could read a lot faster without losing comprehension. I don't begrudge the phonics; it helps my writing a lot to be able to 'speak' it internally as I compose.  But I would never have wanted to depend on phonics alone. Even as it is, it left me with a deficiency in formal and abstract thinking, which again I had to painstakingly recover later in life. That's why this article - and the research it references - is important to me. As Catherine Compton-Lilly and colleagues argue, "based on decades of reading research, continuing to focus only on phonics is educational malpractice." There's no one best way to teach reading; it depends a lot on context. "Complexities related to the teaching of reading lead to different children learning to read in different ways."

A classroom is like a little society (as no doubt many others have observed). It is a closed information environment, especially if external media are banned and if the students don't rotate teacher to teacher. It is (as the title of this blog suggests) something that can come to resemble a cult. Therein lies its strength - and its danger. Take this post about creating, supporting and recognizing a set of values over the course of a year. This can be a powerful tool for conditioning students - but we need to understand that it's a form of conditioning, and not a learning or cognitive exercise. Are the underlying values freely chosen by the student? No. Is there room for an alternative set of values? No. This is why education has become so politicized in some societies, and why it's so important to push back against this.

I can only imagine how much work it took to set up this brilliant parody of educational technology (as I type the support chatbot is having a meltdown). It has everything - ridiculous reserach papers, a haunted book, genius-level learning that starts at birth, 24-hour degree programs. Enjoy. Via Doug Levin. For more, see the Kickstarter.