I'm going to give Daniel Willingham props for a no-qualifications admission that he was wrong in the past. He has argued in the past that students should be taught domain knowledge, rather than critical thinking or other '21st century skills'. But with the observation that "Students are too trusting of what they read on the Internet" comes the recognition that there is a need for "a useful, content-free strategy that could a big difference in student assessment of website accuracy." The one he proposes here is pretty basic: "read laterally. Instead of going through a checklist of features of the website in question (the usual advice) encourage students to get OFF the website to see what others say about it... show click restraint... refraining from clicking on the first result from a Google search (and) use Wikipedia wisely." I think there are more skills, of course, but that's a quibble. More.
"What would happen if someone created a digital platform which schools could use to connect with businesses, experts and industry," asks Steve Wheeler. We're about to foind out because of the service being announced in this post, LikeToBe, which connects professionals with schools and classes. "Using our unique platform and content, LiketoBe connects teachers and students with professionals to provide impartial, real world careers advice and e-pals. Join our network of professionals, become an e-pal to teachers and schools and help us create a grass route approach to career guidance around the world" ("grass route appears to be an eggcorn). Here's the product's investment page on Seedrs.
Venture Beat highlights the funding, but I'm more interested in the direction being taken by the technology. Here's the idea: "We collect dense behavioral data from successful professionals in various roles and use machine learning to build models of which traits separate the successful professionals from the general population." This was the idea behind what we called 'Automated Competency Detection and Recognition'. Study the experts using neural networks, then design training programs to help others develop similar traints. Ah, but it proves to be rather more difficult to implement than to describe. So maybe the funding is the story here, while the product itself is still a testing application. But the idea is still out there, waiting to replace testing and assessment with somnething far more practical. The danger? What if 'expertise' is associated with properties like race and attractiveness? That is, after all, how venture capitalists seem to make their decisions. And college professors, sometimes. But we should keep it out of AI, if we can.
While I want to jump on this and say that my long-held theory has been vindicated, I know that I should await confirmation and test6ing. Still, an article that begins "our brains organize experiences by their similarities, new research suggests" just makes me feel fuzzy all over, because this is the theory I was working on 35 years ago and which underlies most of the work I've done since (here's a quick outline). The research article cited is probably paywalled for you (a ridiculous state of affairs). It's based on behaviour and fMRI studies, which leave plent of room for scepticism. Still, we see the role similarity plays in the creation of knowledge: "immediate post-learning changes in connectivity may reflect a consolidation mechanism that plays an active role in shaping memories over time, in a way that prioritizes their commonalities."
The 2017 Jane Hart's list of the top tools used for e-learning (by mostly professionals) is now available. The list is now 200 tools long, but really, after 50 or so you're getting into the cruft. The top of the list, by contrast, is much more mainstream than in the past: YouTube, Google Search, PowerPoint, Google Docs, Twitter, Word, LinkedIn, Facebook and WordPress. The first big surprise on the list is Trello, a project tracking tool up 21 places to take the 22 spot. Zoom, a video meeting tool, has shot up 38 places to rank 28th.
This newsletter is sent only at the request of subscribers. If you would like to unsubscribe, Click here.
Know a friend who might enjoy this newsletter? Feel free to forward OLDaily to your colleagues. If you received this issue from a friend and would like a free subscription of your own, you can join our mailing list. Click here to subscribe.
Copyright 2017 Stephen Downes Contact: firstname.lastname@example.orgThis work is licensed under a Creative Commons License.