On monocultures, institutional agency, and resilience
D'Arcy Norman,
2026/05/12
As Scott Leslie says, this is a thoughtful post on the Instructure hack. Like Phil Hill, D'Arcy Norman leads with Instructure's public communication regarding the incident, which was (and continues to be) not good, with most people interpreting their claim to have "reached an agreement with the unauthorized actor (ie, ShinyHunters) involved in this incident" to mean "they paid the ransom." Though, what else were they going to do? Along with many others, Norman points to the technology monoculture as a point of failure. "It looks like Instructure manages 3 different Canvas environments: 'production', 'beta', and 'test'. And all 8,800+ institutions appear to share those three environments." But the major lesson to be drawn, he writes, comes from UBC. "The UBC response is truly remarkable... (they) quickly put together a cohesive set of resources to support instructors in rapidly adopting different online platforms to meet the pedagogical needs in their courses." This is a lesson in maintaining expertise and developing resilience, even while working with external partners.
Web: [Direct Link] [This Post][Share]
The Ozempic Problem: The Generative AI Norms Forming in Silence, and How Youth and Adults Can Shape Them
Wyatt Pashia,
Getting Smart,
2026/05/12
I thought this was a decent article though the examples are contrived (as so often happens in articles about education). There isn't (and never was) a Ms. Chen or a Priya or a Fitz or a magical sense that detects when feedback is created by AI. Sure, some things are 'cringe', but that can happen when a human writes (it would be like me trying to use the word 'rizz' meaningfully) and it reflects less than careful proofreading more than the sure and only sign of AI. The 'Ozempic' example here is to convey some sort of social disapproval of the use of drugs for weight loss, as though willful exercise and diet is somehow more socially acceptable. The argument here is that "admitting AI use carries the social risk of being seen as less capable, less creative, or less genuine. But we can move the needle by engaging young people directly. A well-designed nudge reframes disclosure from a moment of 'getting caught' into an act of ownership." Maybe. Maybe new norms will emerge - but note, they will emerge, not be 'constructed' through some process of collective meaning-making and choice.
Web: [Direct Link] [This Post][Share]
Recognized Entities v1.0
W3C,
2026/05/12
Dropped today, "This specification describes a data model with which one or more recognized entities, such as one or more persons and/or organizations, can be described as known to perform specific actions, such as issuing or verifying a verifiable credential." The idea is to provide "a cryptographically-verifiable and privacy-preserving mechanism" allowing credential-holders to prove their credentials are recognized. "The specification is designed to interoperate with existing 'trust infrastructures'... while enabling new decentralized ecosystems to be built using verifiable credentials."
Web: [Direct Link] [This Post][Share]
There Are a Million Fediverses. Some of Them Are Louder than Others.
jaz-michael king,
jaz-michael king's blog,
2026/05/12
This post begins as a criticism of the idea that "the fediverse holds this very specific opinion" and evolves into a discussion about chronological feeds where "you replace algorithmic amplification with sheer volume. If those same 3% of vocal, toxic, aggrieved accounts are posting twenty times a day while the 97% of us post once or twice, who dominates your timeline?" It's a fair criticism - I've found the same effect in my RSS feeds, which is why when I built my own, I privileged accounts that post infrequently. In this post jaz-michael king makes a case for curation, but what's missing though is a consideration of how we should be able to shape our own algorithms (which is essentially what I did) rather than rely on the platform or the machine.
Web: [Direct Link] [This Post][Share]
Squashing Gemini Nano
Mark Corbett Wilson,
Talking with machines,
2026/05/12
This is an update on the revelation that "Google Chrome is silently downloading Gemini Nano onto our computers without a clear opt-in or notification." As Mark Corbett Wilson describes it, it's not simply that people are finding an unexpected AI model on their hard drive. "Beyond the AI model, the 'ghost' data in Chrome often extends to the GPUCache. Chrome builds a massive cache of 'shaders' to speed up graphics rendering on your GPU. Over time, these can balloon into several gigabytes and they rarely clean themselves up." Cleaning it up is not a simple task, as described here.
Web: [Direct Link] [This Post][Share]
One Step Forward, One Step Back
Phil Hill,
On EdTech Newsletter,
2026/05/12
This is an update on the Instructure hack with a focus on media questions and the company's response. Overall it's... not good. Instructure obviously wants to limit the bad publicity, but are doing so in a way that keeps the people impacted less informed. This, at least, according to Phil Hill. "The scale of the breach has not been publicly acknowledged. The academic continuity framing is being articulated by everyone except the company at the center of it. Higher ed's trade press is being deflected to a static page. Direct questions about the most material aspect of the resolution are going unanswered."
Web: [Direct Link] [This Post][Share]
There are many ways to read OLDaily; pick whatever works best for you:
- Read in your web browser on this web page, updated weekdays
- Receive a daily or weekly email newsletter by subscribing here
- Subscribe to the RSS feed using your favourite feed reader
- Follow @OLDaily@mastodon.social on Mastodon or elsewhere in the fediverse
- Follow @OLDaily on Bluesky
- Read OLDaily as a LinkedIn Newsletter
- Integrate the open API using the JSON feed
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 2026 Stephen Downes Contact: stephen@downes.ca
This work is licensed under a Creative Commons License.