If you want a holiday reading project, this might be it. I especially recommend chapter 6 on neural networks and deep learning - it covers such things as multilayer perceptrons, convolutional neural networks, and recurrent neural networks. The book was written in a wiki, with chapters available as PDF, and will be available on Amazon. The book can get complex in places but it also describes pretty clearly what the different types of neural networks can be used for.
This is the transcript of a talk devoted to showing how people have lost faith in the idea that data will solve the problems of 21st century education and technology. "We’re now learning some hard lessons from the rapid expansion of data analytics, algorithms, and AI across society," writes Ben Williamson. "For many educators and students alike, education is more than the numbers in an adaptive, personalized learning platform, and includes non-quantifiable relationships, interactions, discussion, and thinking." There are some good points overall in this article, but I do want to sound the cautionary note that the emphasis on ethics not be the solution everyone thinks it will be. We don't have agreement on ethics, on what counts as good, nor what counts as right, nor even what counts as relevant.
Audrey Watters is not writing a ten-part review of education this year offering instead this quick summary that says, essentially, that innovation in the field has slowed to a crawl and that governments and educators are making the same mistakes over and over again. "If I look back at what I’ve written in previous years, I feel like I’ve already covered everything I could say about 2018," she writes. "I’ll just note here that the most important 'innovator' in this area this year (other than Gritty) was surely the e-cigarette maker Juul, which offered a mindfulness curriculum to schools – offered them the curriculum and $20,000, that is – to talk about vaping." Yeah, that's 2018 in a nutshell.
This paper reports on the first of two projects I did with the Canada School of Public Service. In this project I surveyed the landscape of new learning technlogies, including such areas as mobile learning, crowdsourcing and e-books, and studied the work they had undertaken to support initiatives like this, and given their state of development and need, mapped a development strategy for them.
This article summarizes the EMPOWER project, launched the 2015 by the European Association of Distance Teaching Universities (EADTU). Its goal was "to empower students to become life-long, self-directed learners in open, online and blended-learning environments." It draws on the "‘retention formula’ proposed by Seidman, S=AC+Eid+(E+C)PaM, where S = success, AC = appropriate course choice, Eid = early identification of vulnerable students, (E + C) PaM = early and continuous proactive motivation support." This of course isn't a formula at all (the '+' and '=' signs do not signify mathematical operations; they're just decoration). Still, the paper offers a pretty comprehensive overview of student support for online learning.
This article references a case where a case where a publisher (American Scientific Publishers) was identified as "displaying a problematic pattern of citations". Phil Davis surveys some of the measures journals take to prevent this. "Editors need to be insulated from the business of publishing, which often means keeping them away from their colleagues in marketing, sales, and advertising" and "many publishers have explicit rules that prevent editors from handling their own paper or the papers of authors very closely associated with them." The editing process is an important part of the provenance of the paper, but it is also usually hidden from view. Openness would do a lot more to help here than policy, especially as we move from more formal publishing environments to things like blog networks and data exchanges.
As the news item says, "The Profiles Ontology is an RDF vocabulary to describe profiles of (one or more) standards for information resources" and "Content Negotiation by Profile describes how Internet clients may negotiate for content provided by servers according to profiles." There are still open issues in both drafts. The first is relativly simple and straightforward, while the second addresses what is an inherently difficult and complex problem. I personally am not sure whether it's worth the effort, but then again I'm a bit of an ontology sceptic to begin with. My preference is to define fields and thn let people use them as they wish, allowing convention rather than specification define their meaning and ngotiation.
Like many others I'm sure, I've received those emails to please donate to EdX. But to be clear, the amount they need to survive is only a small blip in Harvard's multi-billion dollar endowment. So perspective is everything. "Slowly the company -- like its for-profit cousin Coursera -- has moved more content and features behind paywalls." It made $54 million in 2016, So now: "students will no longer be able to audit an individual course free for an unlimited period of time. That period will vary, but is usually around six weeks for most courses." So you'd better be prepared to learn fast. The article concludes with pundits saying the market has turned toward Online Program Management (OPM).
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 2018 Stephen Downes Contact: firstname.lastname@example.orgThis work is licensed under a Creative Commons License.