OLDaily

Welcome to Online Learning Daily, your best source for news and commentary about learning technology, new media, and related topics.

AI can carry out qualitative research at unprecedented scale
Impact of Social Sciences, 2024/11/01

---

Social sciences students around the world aren't sure whether to cheer or fear this new technology."We have developed and launched an easy-to-use platform for conducting large-scale qualitative interviews, based on artificial intelligence... A chat interface allows the respondent to interact with a LLM that collects their responses and generates new questions." I'm sure someone will tell us why AI can't ask people questions and analyze the results, while meanwhile the equivalent of a thousand PhD studies a day will be conducted by the system.

Web: [Direct Link] [This Post][Share]

'We Were Wrong': An Oral History of WIRED’s Original Website
Virginia Heffernan, Wired, 2024/11/01

---

Hopefully this will be your 'one free article' of the month from Wired. Normally I wouldn't touch it, but it's supposedly an oral history of the original Wired website called HotWired. Accept the cookies, then minimize the great big subscription ad.  This matters to me not because of the content (which was and is the focus of these writers) but because of the HotWired Threads discussion area, where I was a regular (presciently, I knew they would eventually kill Threads, so my posts there became the beginning of my Articles section on my website). Part way through they take credit for having the first banner ad, but I'm here to say they weren't first - the actual first banner ad appeared on a specialty website called The Spot (yes, I was a Spotfan too). Via Kottke.

Web: [Direct Link] [This Post][Share]

How ChatGPT Search affects the broader AI landscape
Ben Dickson, TechTalks - Technology solving problems... and creating new ones, TechTalks, 2024/11/01

---

As reported here, ChatGPT can now search the web when generating its responses. Probably this will impact Perplexity more than anything else. I gave it a quick try and found it was nothing special (though in this world of AI it's pretty easy to beco0me jaded). As Ben Dickson says, "all leading AI companies are converging on the same set of features and their interfaces are becoming very similar. This is further indication of the commoditization of the AI space."

Web: [Direct Link] [This Post][Share]

AI and workers
Bryan Alexander, 2024/11/01

---

I personally think it's an easy futurism (or 'prophesy') that simply draws scenarios. "The AI and labor market picture is deeply in flux, with the potential for many local and regional variations," writes Bryan Alexander. "I think we have three possible scenarios (replacement, redesign, no effect) and that we'll likely see versions of each play out in the short and medium term future." Yes, but which will predominate and where? People need to know - they need to make decisions based on which picture is most likely. My own take: if you work in a field that involves in creating, summarizing, categorizing, displaying, representing or curating information, for whatever purpose, your position will be mostly replaced by an AI (or euphemistically, 'redefined with an effect on employment').

Web: [Direct Link] [This Post][Share]

Flipboard Expands Federation in Brazil, Canada, Germany and U.K.
Jessica Jordan, About Flipboard, 2024/11/01

---

I haven't tried following any of these and I'm jut holding my breath in case it all turns out to be based on subscription walls. Still, Flipboard is federating, which is a good thing: "People using Mastodon in any of these countries can now follow news curation and lifestyle coverage from some of their most popular national media outlets. This update brings 150 new publishers and over 2,000 new Flipboard Magazines to people in the fediverse."

Web: [Direct Link] [This Post][Share]

What makes memories last - dynamic ensembles or static synapses?
Jason Shepherd, The Transmitter: Neuroscience News and Perspectives, The Transmitter, 2024/11/01

---

This article raises a fascinating question and while the eight experts asked for responses generally say 'both' it really feels like it ought to be one or the other. The gist of the question: are memories formed by static and stable connections between specific eurons (aka 'engrams') or are they stable 'patterns of connectivity' that may 'drift' from one set of neurons to another over time? "Teasing out how different subfields conceptualize central terms might help move this long-standing debate forward," writes Jason Shepherd. "What is information,' and how is it 'represented' in the brain? What is an engram?" From where I sit these concepts may well prove to be very problematic, as they require a third-person perspective that's just not available inside a working synaptic network.

Web: [Direct Link] [This Post][Share]

AI's "human in the loop" isn’t
Cory Doctorow, Pluralistic, 2024/11/01

---

Here's the big problem with 'humans in the loop' to safeguard AI, according to Cory Doctorow: "Humans are wildly imperfect, and one thing they turn out to be very bad at is supervising AIs." It should be noted, I think, that humans were wildly imperfect before AI as well, and it was just something we dealt with. But also: even if we put humans in the loop, and even if they're not making mistakes, will the be willing or able to overrule the AI? "We've all encountered instances in which 'computer says no' and the hapless person operating the computer just shrugs their shoulders apologetically. Nothing I can do, sorry!" The 'human in the loop', concludes Cory Doctorow, "is a false promise." 

Web: [Direct Link] [This Post][Share]

RIP botsin.space
Colin Mitchell, muffinlabs, 2024/11/01

---

This is the other shoe in the fediverse beginning to drop - server costs are making it too expensive to run servers hosting a few thousand users. Now that's not to say they'll never be sustainable - a cooperative model where users pay the costs might still work. But as a hobby project, a fediverse server has a limited scale. As, I guess, was always going to be the case.

Web: [Direct Link] [This Post][Share]

We publish six to eight or so short posts every weekday linking to the best, most interesting and most important pieces of content in the field. Read more about what we cover. We also list papers and articles by Stephen Downes and his presentations from around the world.

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 Twitter
• Integrate the open API using the JSON feed