[Home] [Top] [Archives] [About] [Options]

OLDaily

Newsletters vs. Blogs and Social Media - Thoughts for Tech Coaches
Richard Byrne, Free Technology for Teachers, 2022/06/03


Icon

The answer to this question isn't one or the other, of course. It's all three (and various flavours of each). You use different media for different purposes. For people who want daily updates from me, there's email newsletters - that's really good for people who don't want to miss anything. The same is true of Twitter followers. For the content curators and knowledge seekers, RSS is really good because it provides a searchable archive and a way to integrate what I write with everything else they're reading. People who need more presence in their content will prefer my web-based presentations, which include audio and video. They and other people who find me by search rather than subscription also benefit from my blog posts. For those who want to see the dialogue with the world as my ideas form and take shape, there's Mastodon (and some of my video playlists) - but be warned, my thoughts are borne out of a lot of extrinsic and extraneous content, such as cycling, video games and photography. And then there's my home page, which includes links to all these things.

Web: [Direct Link] [This Post]


25 Books Every Technologist Should Read
Tom Haymes, Idea Spaces, 2022/06/03


Icon

What this list shows me is that the '25 books' isn't the list of books you must read, but rather, for the author, the list of books they have read. Even assuming the same framework (which depicts technology as being about designing systems to produce outcomes) I can think of alternative (and, IMHO, deeper and better) books in each category. Why the military texts for 'understanding systems' when we woulkd be better served with analogies from business (Senge, say) or ecology (Bronfenbrenner). Similarly, for design, I'd be more likely to draw from architecture (Lockton, for example) than science fiction. But if I did want to talk about worldmaking, I'd be reading about knowledge construction (van Fraassen or Goodman). But that's me. There is not one best list. There's your list, and your choices (war and SF versus ecology and construction) not only reveal who you are, in some important ways they also define who you are. Via Aaron Davis.

Web: [Direct Link] [This Post]


Towards artificial general intelligence via a multimodal foundation model
Nanyi Fei, et al., Nature, 2022/06/03


Icon

This article describes a strategy to produce artificial general intelligence (AGI) with what might be called weak semantic correlation with image-text pairs. Traditional models seek strong correlation, and are therefore based on feature detection mechanisms that can be organized into (say) sentences. This requires a lot of overhead for limited capacity; "only millions of image-caption pairs are collected by years of human annotation"). Instead, "since there do not necessarily exist fine-grained region-word matches between image and text modalities, we drop the time-consuming object detectors." Without this overhead, the system is able to find patterns for vague concepts (eg., 'nature' or 'time') and find non-obvious and non-explicit between things, like metaphors and analogies.

Web: [Direct Link] [This Post]


How Else Are People Using FindClone's $5/Month Face Recognition Service?
Josephine Lulamae, Algorithm Watch, 2022/06/03


Icon

There's no good mainstream coverage of this yet, but we've entered the world where people can use facial recognition software for as little as $5 a month. This article describes a number of ways it's being applied. FindClone is a Russian company, but now because of the invasion of Ukraine people are looking for FindClone alternatives. It's enough of a concern that image 'cloaking' technology such as Fawkes is being developed. I think attitudes about facial recognition may shift over time; we've used it for centuries on a human scale (think: wanted posters, police lineups, drivers licenses, passports). The line will be drawn based on already existing conventions, with prohibitions against using facial recognition for discrimination or stalking, for example. The edge cases are new uses such as targeted advertising, setting insurance rates, social credit, or performance evaluation.

Web: [Direct Link] [This Post]


Five ways to reduce PowerPoint overload
Cliff Atkinson, Richard E. Mayer, Indezine, 2022/06/03


Icon

I think this presentation (15 page PDF) offers good advice, but for the wrong reasons. It advocates for less text and less extraneous content on Power Point slides, and a structure where the words and picture on each slide support one main point. I agree. But not because of cognitive load theory, which (as readers know) I have criticized in the past. No, instead I would argue that the advice is good because humans perceive through pattern recognition. This suggests a model where presenting the same relatively simple (but not too simple!) pattern in multiple modalities (test and audio, but also activity and other sensations) develops and/or activates deeper patterns, layering what is being presented on pre-existing knowledge (aka pre-existing patterns of connectivity). Via Mike Taylor.

Web: [Direct Link] [This Post]


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 2022 Stephen Downes Contact: stephen@downes.ca

This work is licensed under a Creative Commons License.