OLDaily

Tony Bates offers a response to my paper on connectivism. Unsurprisingly, he does not agree with the position I outline. He argues that "any theory of learning... needs to go beyond the mechanics of physiology" and to explain "consciousness and thinking, rather than wondering what neurological reactions are happening." He also argues that "there are serious questions about his explanation of how networks learn." For example, he asks, "what do the nodes actually contribute to the network other than sending signals?" He is also dissatisfied with my treatment of pedagogy and argues "to ignore the social and economic factors that influence teaching, learning and the construction and use of knowledge." (p.s. and for the record: I was not "home tutored". I spent 13 years in the Ontario public school system, another year in community college, and then an additional nine years as a university student at the Universities of Calgary and Alberta. I have since then built on that base, but I am no more of an "auto-didact" than Bates is). Anyhow, I offer a reply to the first part of the review, with (probably) more to come.

I thought that this was a relatively good article in that it explores a number of features of web3 that go unmentioned by others (and critics especially). But I think we need to be clear that we can talk about web3 without talking about crypto-currencies (like bitcoin or ethereum), just as we could talk about web2 without talking about PayPal or advertising. 'Play to earn' models and NFTs are the web3 equivalents of affiliate marketing and commercial e-textbooks. The web2 or web3 model makes them possible, but it doesn't make them necessary. But it is true that while "web2 is about communication and sharing (like Instagram and Facebook), web3 is about creation and ownership where users can retain control."

There have been various attempts over the years to make asynchronous interactive audio a thing, but this effort has the feel of something more substantial than previous attempts. By pushing a button in the iHeartRadio app, listeners record a 30-second voice message. This message then becomes available on the app's CMS "so the voice recordings are available to use, live on air, within about 10 seconds after sending." I'm not sure I'd broadcast them live without listening to them first so it seems to me that iHeart's next step will probably be to offer an audio transcription 'preview' so broadcasters can read what the message says before deciding to use it.

Michael Feldstein writes a long post saying he doesn't know very much about web3 but that it probably won't help education very much. Now having offered a MOOC on web3 as it was back in 2019, I think I can say that I do have a pretty good understanding of it, both at a technical level and at a social level. Now Feldstein's article is valuable in that it raises many of the questions and confusions people have (for example, he should know self-sovereign identity tokens not only allow people to manage their own logins, they also enable zero-knowledge proofs, which means people can validate credentials without giving out personal information). But I will say that this probably isn't the time for a Economy 2.0 statup - not in a world of content-addressible decentralized resources.

What I like about NASA (aside from the obvious thing about space exploration) is how well it does outreach and learning resources online. I consider it a model of what other scientific agencies (including the one where I work) could do to advance interest and engagement in the field. This page points to a number of free NASA resources that illustrate what I mean here.

This item was posted last September but it showed up in my feed reader only today, and it's certainly worth passing along. It describes cmi5 (EduTech wiki says "cmi5 probably stands for computer managed instruction, fifth attempt"), which has been in the works for a number of years now. If I had to summarize it in a sentence, I would say that cmi5 basically takes SCORM and combines it with xAPI. In other words, it describes the conditions for launching and running a learning resource, and includes information about how to record the activities undertaken using that learning resource. For more, see the cmi5 catapult, "a freely available, open-source cmi5 content player and cmi5 conformance test suite for use by DoD stakeholders or and other e-learning technology vendors."

This is the listing on ProductHunt for Coda 3.0, a document editing tool that features a large number of integrations with other applications and pre-packaged templates for them (called 'packs'). I devoted an episode of Stephen Follows Instructions to it. It's an hour and 20 minutes long, but it covers everything. If you want to see how to approach a new application and learn it from scratch on your own, watch this video and observe my process as I work through the options, make assumptions, and try things out. Would I actually use this product? Well, I like it a lot, but all of your data is stored on their service, and there's no way to export it at all. To me, that just creates too large a risk, and so I would never use it as an authoring platform.

This is yet another 'how we responded to Covid' article. But I want to include it here to demonstrate how much more useful it is posted as a blog post than published in a $169 textbook few will read. We also find it reflects some understanding of online learning prior to 2020. "Rather than simply recording his existing lectures, David opted for a more compressed format inspired by podcasts. He broke the material into 15-25 minute tightly scripted chunks that students could consume in whatever way suited them." Also worth noting: "the focus group found that requiring a raised hand to speak, whether real or virtual, gave the meeting more structure than an in-person, organic discussion, and made space for everybody to express their view." I've always preferred that approach for any group more than a few people

This article is a reminder that, in a way, some things never change. The 'stickers' in question are things like the gold stars handed out four students to display somewhere (the fridge, the wall, the lunch box). They're the pre-internet version of badges. The gist of this article is that teachers should avoid the use of stickers, for two major reasons: first, because they are based on "assuming that the child has full control of his or her behaviour and could demonstrate the desired behaviour if they wanted to, but sometimes choose not to do so"; and second, because "when we incentivize behaviour, we run the risk of extinguishing pre-existing intrinsic motivation." Via Doug Peterson.

Tony Bates reviews Fayed & Cummings (eds) (2021) Teaching in the Post Covid 19 Era. I'm not about to shell out $US 169 for this book, even if that works out to around $2 per chapter. As he says, "no-one is going to read this 764 page book from end to end. Readers are going to dip in and out." He also suggests that though it offers "many examples here of innovative teaching, particularly using synchronous technology," the lack of a comprehensive index, unifying themes, or sense of organization mean it's really basically a list of related articles. Also, readers "would think that online learning was invented in March, 2020. Very, very few of the articles made any reference to all the 30 years of research." Finally, as Bates says, "the real problem here is not online learning, but what are increasingly outdated and ineffective lecture methods, which become even more ineffective when transferred online without modification. That surely is one of the most important lessons from the pandemic." The other - to me - was the value of open publishing. Too bad all this work will remain hidden from view, with nobody but the most dedicated (like Bates) reading it.

Honestly, I think it's cheating a bit to make predictions for the year in the last few days of February, but that's what this article does. And reading them, they feel more like predictions of what publications will write about, as opposed to what will actually happen in the field. For example, the idea of the metaverse isn't new and won't actually make great gains in 2022, simply because it's expensive and time-consuming and takes a lot of work to get even minimal results. The discussion of critical race theory, the second concept raised, will continue, but while equity and inclusion are important issues generally, the efforts to suppress discussion of race will again have minimal impact on actual practice, except maybe to raise awareness a little. The great resignation, concept number three, could have a significant impact where teachers are undervalued and underpaid. The same could be said about the idea of pandemic heroes. Will any of these "shape education"? No.

This is a headline guaranteed to catch your eye, and possible to cause howls of protest. Here's the claim: a study (15 page PDF) has found that "in a remote environment, an artificial intelligence (AI) tutoring system can outperform expert human instructors." Specifically, "In this randomized clinical trial, VOA feedback demonstrated superior performance outcome and skill transfer, with equivalent OSATS ratings and cognitive and emotional responses compared with remote expert instruction." Why? "It helped students achieve higher expertise by bringing awareness to their metric goals during resections and setting measurable performance objectives, 2 effective strategies of learning theory." Moreover, "This AI intervention saved approximately 53 hours of expert supervision and formative assessment over 13 weeks compared with the instructor group while resulting in comparable OSATS scores." Something to think about.

This article covers a report (102 page PDF) from the American Council on Education (ACE) on the Education Blockchain Initiative, funded by the U.S. Department of Education. This initiative includes Blockchain Innovation Challenge and the report covers the development of that project. Four projects involving teachers and educational resources are featured. Overall, the use cases focus on learning and employment records (LER), verifiable credentials (VC), and decentralized identifiers (DID). The value of the report is the detailed description it offers of each of the projects, from use cases to the development of a minimum viable product (MVP) and subsequent development. Related: ACE Education Blockchain Initiative​.

This feels more like a press release than a news article, but the existence of TooShare should be noted. According to the article, "TooShare is a new concept that is 100% African and that combines an e-learning platform and a social network. Its objective is to share knowledge by connecting learners, trainers and training institutes through a technological environment that uses the codes and features of social networks, coupled with a Learning Management System." I'm not sure that this is new, exactly, but if it earns some success, all the better.

Alex Usher responds to an op-ed in the Globe and Mail where Sheldon Levy, interim president of University Canada West (UCW), discusses a UCW white paper (28 page PDF) proposing (in Usher's words) "there should be a university entirely devoted to the techsector, which involves a lot of work-integrated learning and the participation of firm representatives on committees that design and modify curriculum." Usher's four questions boil down to two good ones: why focus on the tech sector, rather than tech skills? And why not a polytechnic, instead of a university? But more, to me, this illustrates the issue that arises when industry designs education: it serves its own interests, rather than the students'. And that's why the education sector needs to be arms-length from industry, in my view.

I think this is a pretty challenging concept, not in the sense that it's hard to understand, but in the sense that it may be hard to accept. This article looks at the UBC political science rich transcript, which does a few things over and above providing a list of the course grades. First, it presents an overview of the course contents in a word cloud. Second, and more significantly, it "captures a student’s skills training across 23 areas by tallying the number of political science courses a student completed in which a specific skill was one of the course learning outcomes." It also describes the type of assignments completed and the student's participation in group work. It's problematic, at least to me, because the measures are outside the student's control. For example, the score of '6' in 'work effectively with others' isn't a grade, but merely the number of courses where this is identified as an outcome.

The University of Calgary Gauntlet - where I was an editor in the 1980s - printed its last weekly newspaper on paper in 2017. But it updates daily online. That pretty clearly is the future of news media generally, though it still comes as a surprise to many. That's why I find it so odd that student newspapers in the U.S. are still pondering their future. Where should they put their efforts? Well, not TikTok or Twitter, though that's part of the picture. It's POSSE - Publish on Own Site, Syndicate Elsewhere. Anything else puts you at the mercy of the platform. Though newspapers, which are experimenting with Facebook and Google's AMP, haven't really learned that lesson either.

To be clear, this paper studies teachers as learners. Specifically, "Three elementary teachers were involved in a series of in-depth, one-on-one self-directed online learning (SDOL) sessions where they informally used the Internet for their professional learning." Specifically, the study employs " Thinking aloud (as) a method for generating direct data about the ongoing cognitive processes that occur during learning." I think this is a useful mode of study and have provided examples of my own in my 'Stephen Follows Instructions' video series. I think we learn more in this paper about how to study this kind of data than from the data itself. And I think we see a bit of a Heisenberg effect here, where the fact that their experiences are being voiced aloud actually shapes their experience, and leads to different forms of learning (for example, "teachers in our study also critically evaluated the content and architecture of the online environments"). Note that there's an image missing (from Beach, 2020). Also, I'd have preferred to see a table listing what they said rather than what buttons they clicked. Related: Beach & Willows, Understanding Teachers’ Cognitive Processes during Online Professional Learning: A Methodological Comparison.

I will credit the authors for their honesty, because as I read this study, pretty much everything that could go wrong did go wrong. The study was a decent size but of course not representative of students (consisting of the oft-studied students at a midwestern U.S. university), and involved only one proctoring system (Proctorio). They wanted to ask whether online proctoring made students anxious, and whether this effect was more pronounced for minority or low-income students. They couldn't get Pell grant status information because it was deemed too sensitive, so they relied on a proxy measure of whether students were first generation. On one question, only 22 (6.6%) said they felt anxious about the exam. The authors argue that "being remotely monitored by webcam appears to be a source of anxiety for some students" but we don't have any basis for comparison with (say) being monitored in person. In the conclusion, the authors talk more about the effect on the instructors than the students. Image: Erik Johnson.

No more MailChimp! Last Friday MailChimp lost the ability to read my RSS file I was using to create email newsletters. This was always a dodgy approach at best, but MailChimp has been truly awkward to work with (the fact that it's an advertising platform geared toward 'audiences' and 'campaigns' makes it even more so). So today I learned the MailGun API and wrote a script to send email newsletters directly from gRSShopper. It's a lot easier to work with. I still need to rewrite the subscribe function (unsubscribe is at the bottom of your email and should work, though you can leave the testing to me :) ). If you have any problems with the new version of the email newsletter, please let me know.