news (external)

I thought I linked to Elle Griffin's No one buys books, and commented "books are over", but maybe I just imagined it, because I can't find the link. Still, I would have done that, and in fairness, feel obligated to post this rebuttal. "BookScan counted 767 million print sales in 2023. BookScan claims to cover 85% of print sales." That's a lot of books, though keep in mind that this averages to about three books per person in the U.S. (population 336 million). These figures seem widely accepted even by those proclaiming a crisis in print book sales. It's one of those arguments where everyone can be right.

Ellie Pavlick tells us "we decided that meaning involves concepts in some way... If you use the word 'apple' to mean apple, you need the concept of an apple. That has to be a thing, whether or not you use the word to refer to it." And my opinion is, this is a category error. So it was interesting to see them actually find evidence of such a concept - "we found a small place in the model where it basically boils that connection down into one little vector... It's like this systematic 'retrieve-capital-city' vector." Is that what it means to, say, 'know' what an apple is? If so, though, then the knowledge and the thing knowing are one and the same thing - there's no 'concept' over and above the 'conceiver'.

The first thing I did as I started to read this item is to look up X.509 Certificates on Wikipedia, which made me gulp a little. It's a bit daunting to grasp from scratch. The assertion in this post is that "the abundance of X.509 certificate authorities who already perform identity proofing for businesses provides a rich resource that can be leveraged to boot the verifiable data ecosystem." There are still questions to ask, for example, can we trust these authorities, are they accountable, are the costs reasonable, are they available, and of course, are they technically feasible for the wider population? That said, the idea is new to me, and represents another step in the thinking toward distributed identity (DID).

According to this statement from Harvard, "Harvard Library will offer new sustainable and equitable open access publishing models to advance open access scholarly communication." That's a good thing, and I support it, though I will remind people that that the Public Knowledge Project (PKP) has been doing this for years. Decades, even. My worry, of course, is that years from now people will talk about this as the 'start' of the open journals movement, as is wont to happen when a top tier institution adopts a practice like this (which is why we read, even in this article, things like, "I hope that many research-heavy institutions adopt our approach. The first Harvard Open Access policy launched in 2008 has been adopted nationally and internationally, and it would be great to see similar reach."

Matthias Melcher considers the question I asked yesterday: "what do we want the outcome of an education to be, comparing a student, an intellectual, a billionaire, (or) small-town inhabitants." Of course, none of these is ideal - but each of these represents in some way the ideals our education system seems to aspire to. But more, these four points are so different it's hard to imagine basing a single system on them at all. P.S. I created the image for this using chatGPT 4 - it struck me how similar it was to working with a human designer (eg., I asked for four people, the first version had five people, etc).

The headline is that the United States may soon ban TikTok, a story remarkable in its own right. The underlying thread is that centralized social media services (TikTok, Twitter, Facebook, etc) have become too unstable for individuals and businesses to rely on. Buried deep is the main story: that the only way to reliably access an audience is to give up on platforms trying to control access to it. Hence, "Vox Media's technology news publication The Verge says it also has plans to federate its own site to have more ownership over its content and audience." Via Ben Werdmuller.

There are many ways to read this article. I'm here for the way the author criticizes billionaires, but that's not why I'm sharing it, that's just a bonus. No, what I ask readers to consider is what we want the outcome of an education to be by considering the four points of a pyramid: a young Marc Andreesen who as a student at a public university was able to develop the first commercial web browser; the second, the author of the piece, an apparent 'intellectual', familiar with the works of Agnes Martin, Claes Oldenburg, Julius Evola and Corey Robin; the aforementioned billionaire who believes people like him "should get to make decisions to reorder life as we know it without interference from anyone else"; or the people in small towns who value "the ordinary comforts of kinship, friendship, craft, memory, legend, lore, skills passed down across generations." Don't answer too quickly. Via Dan Gillmor.

Web bloat is something I take seriously because it impacts both cost and performance. If the reader is using a slower or less powerful device, accessing online services can be difficult or at times impossible (think about how unresponsive some shopping or airline sites are, for no good reason). This article looks not only at page size but also at how much of a load the page imposes on your computer once it has downloaded. For comparison with the web sites on the chart, my own home page comes in at about 140 kilobytes (0.14 mB). There's no Javascript on it and it doesn't have cookies. It should load easily just about everywhere (in my office it loads at 640 ms). The same can't be said of everything I build, though. But I don't stop trying. Via Molly White.

I probably wouldn't use the cute philosophical trick as in this post, but I nonetheless agree with the outcome: that respect for human life is not a form of value. The way you can show this (with cute tricks) is to tie yourself in knots arguing about what form of sacrifice of human life is 'worth' more than another in various circumstances. My observation is that in today's society the concepts of 'worth' and 'value' are so tied into our everyday thinking we can scarcely conceive of a world without them. But that, surely, is an error. The ideas of 'worth' and 'value' are not so fundamental as all that. All of the efforts devoted to (say) pedagogies of care, equity, social justice, etc., are attempts to show this. Image: 'Dignity', originally from the Simpsons.

Would Immanuel Kant - a world-famous philosopher who lived in 18th century Prussia - be as popular as an influencer today? The early evidence suggests he would. Created using an AI is "Manu, a 23-year-old influencer who, as @manumanukant, lives a typical young life in 2024. With an Instagram bio that describes himself as 'Spreading good vibes and deep thoughts' he's a handsome young guy who shares images, selfies (himself looking soulful in bookstores for example), stories, and reels and comments." The image is based on actual photos of Kant in his youth, while the content is drawn from his work (rewritten to be more accessible to a 21st century audience). "Since quietly launching on Instagram and Threads in January, the account, which was designed to peak on April 22, Kant's 300th birthday, has achieved over 384.8 thousand impressions, reached 82.4 thousand accounts, and has an engagement rate of 51.2%."

Creative Commons makes the following statement: "access to cultural heritage is a fundamental right. And preservation, access, sharing, use, and reuse of cultural heritage are all some of the essential functions that libraries, archives and museums fulfill to enable everyone to enjoy that fundamental right." That's similar to what the British Museum says as it stores half of Egypt's cultural heritage within its walls. Maybe Creative Commons should rephrase: 'access to our own cultural heritage is a fundamental right'. That would allow me to reproduce, say, Group of Seven paintings - which are part of my cultural heritage - while preventing me from converting Indigenous traditional artwork into cash commodities if that's not something they want to allow me to do. Image: A.Y. Jackson, Red Maple, from Wikipedia (had this as a poster as a child).

"AI-generated avatars that look and sound like deceased relatives are increasingly popular to console those in mourning, or to hide the deaths of loved ones from the elderly and young children," Viola Zhou reports. What's interesting is the variety of reasons people want the avatars - not just for remembrance, but also to hide the death of relatives from young children. Also: "Lin hopes the bot will become his immortal doppelgänger, speaking on his behalf after his death. 'If my descendants ask 'What was Grandpa Lin Zhi like?' they could just talk to the AI version of myself to find out.'" It's not immortality, exactly; it's more like publishing an interactive autobiography. I need to get around to setting my own up.

This post references some good conference coverage by Phil Hill and considers the question of whether institutions are ready for the collapse of online program management (OPM) services. "Leaders in the sector, including 2U, Coursera and Keypath, never made a profit on the activity, and Pearson and Wiley sold off their OPM offshoots in recent months." The revenue-sharing model was never profitable, according to the articles. Moreover, as colleges acquired the relevant skills (especially during the pandemic) the need to outsource abated. The tech is complex enough that, if you're large enough, you want to do it in-house, if only to stay on top of features and costs.

According to this article, "Post News, a Twitter alternative run by the former Waze CEO Noam Bardin, is shutting down. Bardin says the platform 'is not growing fast enough.'" It was backed by venture capitalists Andreesen-Horowitz. It was "a social platform that also offers users ad-free access to paywalled content from publishers such as Fortune, Business Insider, Wired, The Boston Globe, and others." What they missed, I think, is that people not only want to read, they want to share, but they only place users could share was on Post News itself; otherwise, you'd just be sharing a paywall, and nobody wants that.

"The time has come to break free from the shackles of 'best practices' and embrace the power of effectiveness driven by the true experts in education—the schools and educators who implemented these strategies consistently and with a high degree of fidelity." Certainly there has been criticism of the idea of 'best practices' over the years for precisely this sort of reason. But are 'the schools and educators who implemented these strategies' really 'the true experts'? Certainly they would have valuable feedback, as would any practitioner. But they do not assess their practices scientifically, and they are unable to view outside their own context. The problem with 'best practices' isn't that we're talking to the wrong people - it's that we're asking the wrong question. Image: Perceptyx.

This is just a note to remind readers in Canada that graduate students - who teach a significant proportion of university classes - are expected to live on roughly $15-$25 thousand dollars a year. The explanation is that 'work' is 'capped' at 10 hours a week, so it's a good hourly rate, but nobody believes people teaching university classes are actually working only 10 hours a week. You can read about the strike at Western University here and here. As a graduate student association president some 25 years ago I worked actively on this very issue, and it's disappointing to see it persists to this day.

This post is directed toward ed tech companies and makes the point that "in a fragmented impact ecosystem, ed-tech needs collaboration to prioritize education over technology." In other words, "for a technology to count as educational, the market needs to be run as a partnership industry, where developers, educators, researchers, and students actively work together to develop, implement, and scale what works." The article makes four specific recommendations that seem reasonable to me (though I word them a bit differently): first, the company's self-interest needs to be subordinate to "the 5Es of impact, efficacy, effectiveness, ethics, equity, and environmental impact"; second, "focus more on the quality of evidence rather than solely on the type of evidence"; third, "contribute new ideas for impact measurement and understanding", and fourth, do more than just query practitioners for new ideas.

Meta (aka Facebook) has just released its new AI assistant, Meta AI, Built With Llama 3. Time Bray asks it a simple question, which it gets very wrong. "The problem isn't that these answers are really, really wrong (which they are). The problem is that they are terrifyingly plausible, and presented in a tone of serene confidence." I asked it a question about myself and got a short answer that wasn't wrong so much as very misrepresentative of my actual beliefs (see the image). Note that I didn't need to sign in to Facebook to use it (though I'm sure this will change).

Though this is mostly an exercise in taxonomy, and though it is also badly titled, this post on what Carlos Ortegon, Matthias Decuypere, and Ben Williamson call 'edtech brokers' is an interesting glimpse into an infrequently-discussed branch of the field. Edtech brokers position themselves between educational institutions and the (usually commercial) vendors and services that support them. The authors identify three types of edtech brokers: ambassadors, which act as representatives for specific brands; service engines, that function as search portals offering such things as 'what works' indices; and data brokers, that manage data flows between institutions and vendors. They mediate edtech in three ways: by supporting infrastructure building and standards development, by producing evidence of 'impact' and 'efficacy', and by 'professionally shaping' via development and training programs. I think both taxonomies could be extended with a little thought. See the full paper, Mediating educational technologies (17 page PDF).

According to the authors, "studies on teacher-student interaction (TSI) support tools often focus on teacher needs while neglecting student needs and autonomy." This raises the question of how to enable student needs to be expressed during a class session. They describe the development and testing of a tool called NaMemo2 (built on NaMemo, a tool for remembering student names) to address this need. In so doing they propose a TSI framework called STUDIER (i.e., Sparking, Targeting & Understanding, Designing & Implementing, Evaluating & Refining). NaMemo2 is based on an augmented reality (AR) tool "that allows students to convey their willingness to interact to the teacher as well as show their names to the teacher in physical classrooms." 36 page PDF.