Stephen Downes

I have never understood the logic of responding to downturns with layoffs. It seems to me magical thinking to expect that earnings will increase when you reduce your productive capacity. It's like responding to being in debt by saying "I'm going to work less to cut back on expenses." Tim Bray cites 'the Kansas experiment' showing that tax cuts and government workforce reductions made it more difficult, not less difficult, to address financial issues. The same with companies. You have all this qualified staff and infrastructure just sitting there, and instead of figuring out how to make money with it, they just let it go. So wasteful. And now companies think they can cut their way to grown using AI. Now at this point it's still an experiment, the way Kansas was before they ruined it. But it's not just that that the experiment is likely to be a failure, it's that with AI they could have (say) doubled their capacity, and they chose to just lay off half their staff instead.

The AI skill people are lacking, says Sean Stowers, is 'discernment', "the ability to decide whether AI belongs in a given task, which tool fits the situation, what good output actually looks like in your specific context, and when the situation calls for your own expertise instead." He cites 'Learning & AI strategist' David Chestnut, who writes that people focus on skills rather than behaviour change. "People can understand AI, relate to it differently, and still revert to old ways of working. Not because they don't get it - but because behavior change has always been hard." None of this is wrong per se but it's too narrow (and people, are talking about it). It's more than behaviour change, more than 'get on board with the new strategy', more than just 'hard'. It's like they're suggesting people take a leap of faith, but there's more to faith than a leap.

The concept of 'warm demander' is new to me, so I'll pass it along. "The concept, usually credited to education leaders like Judith Kleinfeld and Lisa Delpit, combines genuine care and cultural responsiveness (warmth) with high academic expectations and rigorous instruction (demand). It is explored at length in Franita Ware's book, Warm Demander Teachers: Healthy, Whole, and Transformational." It feels like a mild version of 'tough love'. My main reaction is that it seems far more teacher-driven than student-driven, though we read, "the Warm Demander is a facilitative leader, not a dictator. Warm Demanders emphasize student agency, classroom leadership, goal setting, and accountability." Yet look at the language the teacher uses: "I expect smooth, silent transitions... every student must contribute using established sentence stems... etc." 

I admit I spent more time looking at the image than reading the article, wondering why it was necessary to create a fake set of laptop stickers over top of the original stickers. Was it because the original included stickers like 'hacker' and 'rock against war'? No, the fake layer "includes open source projects that began on institutions of higher education" and is intended to illustrate Apereo executive director Patrick Masson's argument that "what's needed now is an open source renaissance for higher education - one that restores community-built infrastructure, institutional agency, and academic autonomy to the center of the educational enterprise." I'm not going to dispute the objective, not the origin story for the applications illustrated, except to point out that some were build in spite of the organization where they originated, not because of it, and that open source authors have long had to work against the institution's desire to keep the tech in-house, to spin it off commercially, or at the very least, to community-source it. Meanwhile, I think Masson's case might carry more weight if authored on an open source platform, not LinkedIn.

This is a good article describing the principle of 'back-propagation' in some detail. This is one of the major algorithms used to train neural networks (we've mentioned it here a lot over the years). The simple explanation is that back-propagation is the process of correcting outputs in response to feedback. But the trickier part is now this happens when we're looking at a neural network with multiple layers (so-called 'deep' learning). Darren Broemmer could go into more detail and describe the mathematics of it, but he doesn't, and the article doesn't really suffer for it. He does look at some alternatives to correct back-propagation around the edges, and considers some misconceptions, including the larges question, which is whether the human brain itself uses back-propagation (answer: probably not, though it needs to solve similar challenges).

The team of Alex Usher and Maïca Murphy point to what is an everpresent reality in New Brunswick, the desire to cut spending on education. They link to a list of proposals circulated by the government - here's a good copy, the copy on HESA is unreadable - that range from closing campuses to limiting financial aid to in-province students only. The list of options displays the usual lack of imagination displayed by governments around higher education, viewing reducing enrollment as a demographic crisis. What is missing (from both the list of questions and the critique) is any discussion of getting more value from the system, such as offering broader community-based services to the whole population (not just 18-25s). And missing is the obvious way to make up $35-50M - tax the Irvings, New Brunswick's local billionaires. This kind of money can be found in their seat cushions.