OLDaily

JsonLogic applies rules to data. "JsonLogic isn’t a full programming language. It’s a small, safe way to delegate one decision. You could store a rule in a database to decide later. You could send that rule from back-end to front-end so the decision is made immediately from user input. Because the rule is data, you can even build it dynamically from user actions or GUI input." In a case study today I saw it used as a way to help users easily construct queries to a database. Nifty.

From the cool tools department: "D3.js is a JavaScript library for manipulating documents based on data. D3 helps you bring data to life using HTML, SVG, and CSS. D3’s emphasis on web standards gives you the full capabilities of modern browsers without tying yourself to a proprietary framework, combining powerful visualization components and a data-driven approach to DOM manipulation." I haven't tried it but I've seen it in operation and it looks pretty useful.

This is a pretty good paper based on two case studies at the University of Saskatchewan where OER and class-specific, closed-content videos were designed. The first was for a course in veterinary microbiology and the second in music research methods. In addition to the cases, the authors draw out some recommendations and practical considerations for OER video production. The most interesting part was the table of 'highest priority' open video content: at the top were labs and demos, while the least important (for OER) were course-specific information and FAQs. Which makes sense.

I've had a mostly inactive account at iris.ai for a couple of years now (it has recently been warning me it will be deleted if I don't do anything) and used it to access a decentralized data architecture service so I have a sense of the capability the AI company is bringing to this project. The focus of this particular project is 'the AI Chemist', which is a nice hook, but the real story is Iris's access to the CORE library: a metadata dataset (title, author, abstract, publishing year, etc.) for some 210 million articles, and a full text dataset of 29.5 million articles. The AI Chemist, meanwhile, was actually introduced back in June, 2020 (something the Jisc article doesn't mention) when Iris had about 18 million articles. So the big change here is the introduction of all that metadata from the millions of closed-access papers. That doesn;t really seem to me to be a great leap forward.

Here it is: the ed tech conference proposal generator. The adjectives don't really work but the titles are otherwise realistic. It would be a fun conference game to challenge people to give a two-minute impromptu talk based on generaled proposal titles.

There's value in this post, but also especially in the lengthy comment thread that follows. The premise here is that the associate editor of an academic journal writes, "it is my strong suspicion that the peer review system is finally broken beyond reasonable repair." So what can be done? Suggestions range from overhauling the system to something called slow philosophy. These days, though I am frequently asked, I rarely complete reviews. At first, I didn't want to support subscription-based journals. Also, journals stopped caring whether I was qualified. Then they treated it as an obligation after I've published a paper. But you know, we don't need a review system any more, at least, not in the sense of two blind reviewers deciding whether a paper deserves to see the light of day.

OK, here's the slightly simplified explanation of what a Non-Fungible Token (NFT) is. Take any digital content, and run it through an encryption function known as a hashing algorithm. This produces a unique identifier for that content. Next, create a unique identifier for a person (typically the address of their digital wallet). Write a statement saying the first belongs to the second. Embed this statement into a blockchain so that it cannot be altered. You have created an NFT. To the extent that this is useful in education, you can now define the future of NFT in Edu (your results may vary, but this article takes you through some of the possibilities).

Mark Johnson looks at the various tools he has over the years thought might lead to a personal learning environment (PLE) - "an idea that really came to nothing." Some of the tools for using AI are the latest candidate. "AI has the property of being able to return more variety than it is given." And this was basically what was true of the other tools. The idea was always to be able to begin with something relatively straightforward and easy to master, and to use it to open up a world of possibilities. I think that's a good insight. "The motivation to learn technical skill comes about through our hunger for greater variety. We need to design our pedagogy from this principle."

This post in the BERA Blog is intended to introduce us to the inaugural edition of #JoyFE newspaper, "a group of further education (FE) professionals who want to ‘rethink education and take action’", and more to the point, to frame it as an example of "an example of ‘unleadership’ an embodiment, experienced and practised, rather than a theoretical construct." I'm not sure the language of 'leadership' really fits here, one way or another, and whatever I may think of the content of JoyFE or why this publication merits mention in BERA, I applaud the idea of practitioners taking to virtual pen and virtual paper to get their ideas out there and start sharing them.

One common experience with commercial online services is that they will eventually turn the screw, reducing services and increasing costs. Thus is the case with cloud storage, including DropBox, which has reduced services recently. So maybe you want to host your own cloud storage service. "Having spent time trying to do this for a little while," writes Ernie Smith, "I’ve started to realize why normal people don’t do this." He looks at a variety of services, including NextCloud, which I've also tried, but concludes, "it’s not for everyone, especially if you don’t want to spend your weekends hunting down bugs." Other tools included Syncthing for file sync across a variety of machines and Backblaze for long-term cloud file storage. But for now, he writes, "the average person is probably better off with something like Dropbox, OneDrive, or Google Drive, at least at this juncture." Via Aaron Davis.

Obviously, if we mean 'conscious like you or I', then neural nets are not conscious, because they're not like you or I. But that's not what OpenAI's Ilya Sutskever was talking about when he tweeted that large neural nets may be “slightly conscious.” It goes to the nature of consciousness. If, like me, you think that consciousness just is experience, then even a neural network can be conscious, albeit in a very limited fashion, not even equivalent to the consciousness of a worm. And if experience is just the stimulation of connected entities, then it's not wrong to say that everything possesses some sort of rudimentary consciousness. I'm fine with that. To me, it's a much easier proposition to defend than the one that says that by virtue of some special feature or another, only humans can be conscious.

If you work in education (and especially education policy) think this is a really important message. "Many of us assume that flourishing in the face of adversity requires a certain kind of mindset... yet such efforts have been largely unsuccessful at reducing poverty and unemployment, and have been derided both by the people they were designed to help and by those advocating on their behalf." Why? Because "the four components of a mindset that are claimed to enable flourishing – locus of control, self-regulation, approach orientation, and being trusting and agreeable – are not only less common in low-income contexts, they fit poorly to such contexts." Rather, abandoning them is a rational response to their reduced circumstances; "they’re regulating emotions and conserving their energies so that they don’t face continual disappointment or overlook very real threats."

Skip past the standard ten-cent interview questions to about half way down the page, where the discussion of critical thinking starts. That's where you get this: "what we need is to try and understand one another... to do this, we need to try and see the world from the other person’s perspective... and this requires both critical thinking skills and a dose of good will. So although both critical thinking skills and good will are useful in isolation, when they are present together – that’s when the magic happens." Exactly. Of course, the corollary is, when you lose the good will, you have also lost the ability to understand one another.

This post argues that people in the open learning community should use open badges. The authors write, "Community responses are instrumental in pushing back against creeping commercialisation of education, it is vital that we pay attention to how such shared effort in technology is used. One way to ensure that our voices are heard is to occupy the open badges space." It's worth comparing this perspective with some of the other perspectives on microcredentials from this week, especially those arguing that they should be managed and regulated. You can be sure that the list thing administrators will want to do is 'keep badges weird'.

I want to pass this along because it shows pretty clearly the value proposition FutureLearn and other online course services are offering to potential learners around the world. "The creative economy," they write, "comprises industries that are based on individual skill and creativity, which have the potential to generate income and jobs through the generation of intellectual property." All you need to earn an income in this economy is a computer, an internet connection... and an education. The article lists a variety of occupations, along with corresponding salaries, expressed in Philippine pesos. For example: Video Editor, median salary: PHP 300,000 annually. If you're wondering, that translates into about $7,500. Annually.

The pile of Lego blocks as the main image should be enough to show that there's nothing new about this "new hybrid learning methodology". It's that old standby, created back in the 90s as learning objects: "Imagine the program design as the creation of a set of building blocks (think LEGO) that can be assembled in a variety of configurations, with each configuration personalized to the interests, needs and constraints of that local site." But I wanted to include this post because of its use of the term 'digital learning transfer platforms' in place of learning experience platforms (LXP). I thought it was a nice touch. That is all.

Peter Greene  minces no words (nor should he): "one element of Learning Loss is just plain made up," he writes. "There is no reason to believe that getting Pat to score higher on the Big Standardized Test will earn Pat more money at work and a better life. None. Raising Pat's test score above the score that Pat would have achieved in some other unboosted alternate universe accomplishes nothing except getting Pat a higher score." This, though, won't stop the hucksters from trying to convert learning loss panic into a money tree.

Some companies make grand claims about the benefit of online tutoring, but their claims should be treated with caution, according to this article. I completely agree. I spent seven years tutoring by telephone for Athabasca University, and sometimes it really worked, and sometimes it did noting. “We haven’t proven that online tutoring is guaranteed to work,” said Matthew Kraft, an economist at Brown University who led the Chicago study (5 page PDF). “But we haven’t gotten evidence to say this is really going to tank... Trust but verify... It’s probably true that some kids make huge gains. Whether all kids did is maybe a different question, and not the one that they’re trying to convince you they answered."

This is just a small thing, but something I've also run into in my own development work, and these considerations apply not only to Google Fonts, but also other content delivery network (CDN) resources such as Javascript libraries, such as jQuery or React, and things like Font-Awesome. The issue is this: the resource provider gathers individual data from the CDN. "Google sponges user data via its fonts on my site, and I am subject to possible blame," writes Alan Levine, referring to things like the GDPR, which prohibits such data gathering. Now you could download and install these fonts and services on your own website, but it's not trivial, especially if you want to keep them up to date and secure (there are tools that do this, but installing them isn't trivial, and I still haven't figured how to put this all together in, say, a self-assembling Docker container).

“IDs don’t belong on the open web,” argues Insider Inc.'s Jana Meron. “I like seller-defined audiences and all of these different things, but really the ID belongs where you have to log in... Publisher data is the most valuable thing for an advertiser.” There's merit to this argument. You shouldn't have to hold your ID out for everyone to see as you browse the web. That's the problem with tracking cookies, and it's the problem with some of the post-cookie proposals. But it's not, I think, an argument for some other form of tracking either. We should be able to self-identify with a single click, but it should be a zero-knowledge sort of self-identification, where you don't actually share personal data with the website. And advertisements - and educational applications - should respect that.