Stephen Downes

The Linked Web Storage Working Group has published four First Public Working Drafts for the Linked Web Storage (LWS) 1.0 Authentication Suite, including authentication for OpenID Connect, SAML 2.0, Self-signed Identity using Controlled Identifiers, and Self-signed Identity using did:key." This gives an indication of what sort of authentication technologies to be looking at in the future. Image: Wikipedia.

There has been a lot of fuss about the many software vulnerabilities discovered by Anthropic's new Claude Mythos model. But the flip side of that is the potential for software developers to fix them. "Defenders finally have a chance to win, decisively," exaults Mozilla in this post describing how it responded with fixes for 271 vulnerabilities. "Encouragingly, we also haven't seen any bugs that couldn't have been found by an elite human researcher... Software like Firefox is designed in a modular way for humans to be able to reason about its correctness. It is complex, but not arbitrarily complex. The defects are finite, and we are entering a world where we can finally find them all."

This is a long and muddled article arguing, in essence, that the mechanisms supporting course or credit equivalency between institutions (aka 'The Transfer Industrial Complex') is not worth the cost. How such an article could be written without even a mention of the Bologna process is a mystery. But I digress. My main objection is Hollis Robbins's supposition (supported via a  paywalled Chronicle article) that no two disciplines (and, by extension, I guess, courses) are the same. "No one thinks there's such a thing as 'playing sports' generically. There's just playing football, gymnastics, golf, wrestling, and shotput, each with standards of excellence particular to it. This is why we don't establish generic-sports coaching centers staffed by noncoaching staff..." This is a ridiculous proposition. There is considerable cross-over between sports, especially in the areas of fitness and conditioning. There are courses devoted to teaching people how to be fitness coaches. When I was in high school, the very same guy taught me football, gymnastics, golf, wrestling, and shotput (I was bad at all of them, but it wasn't his fault). Denying that skills transfer - in any discipline - is folly, and so obviously so, that I suspect ulterior motives behind such arguments.

I want to preface this comments by asserting that not everyone is driven by fear. In some societies, this may be the dominant driver, but in many (and I'd like to include my own) people are motivated as much by hope, a desire to help, ambition for the future, and even the sheer joy of things. So there are other responses to AI and technology than just fear. Now having said that, I should mention that Alexandra Mihai's main point here is to argue that "If we want to support students in developing certain skills, we need to do it intentionally, through course and task design, not by conditioning their classroom experience. And here I mean either by forcing them to use AI or by forcing them to switch off their devices." An intentional response, in other words, rather than the all-or-nothing approach so common in educational circles. I endorse this, with the caveat that we should not be limiting our imagination to courses and classrooms. The world is so much larger.

This is a set of questions that might be posed by students but the answers aren't answers in the usual sense - they probe the question itself, asking why the question matters, what the question tells us (the education system) about ourselves, what we can do in response, and some practical suggestions. That said, I think the text in the first blue section ('why this question matters') has been mixed up with something else (so just skip that box; the rest of the article makes sense).

This is interesting though right at the edge of the tech horizon. The tag line says it best: "Turn your favorite coding assistant into an expert at building and deploying agents on Google Cloud." What that means is that if you use a coding platform, such as Gemini CLI Claude Code, Codex, or Antigravity, you can add Google's 'agents-cli' as a tool that allows that platform to create and manage agents to perform specialized functions. This reduces the load you're putting on your coding platform and allows you to access specialized services without having to build them. Today things like this work in coding platforms, but of course the concept can be applied in the future to any sort of platform you can think of.