Vic Work: notes on learning, technology and play

Sometimes a great notion comes to my head, around coffee #2 on a Saturday. If I decide to do something with the notion, then it turns into a small experiment - with technology, or learning, or working differently. This is a write-up of one such experiment. To sum if up: I revived two of my old devices and am now using them as daily drivers. The gear - what am I using? The machine I use for 9 - to - 5 work doesn't change. I swapped my smartphone and my personal computer. My smartphone changed f...

1 You refuse to chuck it into an "AI" enshittificator. 2 You want real eyes / ears / noses / hands / tongues all over it. 3 You exercise the right to a completely shit & private first draft. 4 You treat your work as thinking out loud, and you're picky about who you think with. 5 You try to remember pre-web rhythms, depths, textures, resistances, patterns, ways of packaging an experience. 6 Or to come up with some brand new ones. ...

William Morris was born on this day, 190 years ago. Here's what the internet might have looked like if he had helped build it. Never any DARPA / military involvement Definitely woke Socialist, and aimed at "making more Socialists" Artists retain control No heaps of profit? No problem Beautiful, or useful, or GTFO Possibly more accessible to neurodivergent users (incl Morris himself) No cops No wasteful inaccessible luxury No centralising / monopolising tendencies ...

Good studies published on enthusiasm and usage of "AI" in #education professions. Administrators & curriculum designers seem keen. Teachers - wary. Question asked by journalists and pundits is "what would it take to change that?" Automatic assumption here is that it needs changing - "must get teachers on board!" "Taboo" assumption would go the other way - "must get admin staff off GPT!" Real value of study, to map where "AI" will/won't thrive, ignored. Such is "AI" discourse. McGehee, N. (20...

Studies suggest that paper>screens for reading & handwriting>typing for notes.* Here's a workflow to learn & share. 1 Get book from library 2 Get notepad & pen 3 Sit down, read book, take notes 4 Notes -> plain text (imagetotext dot info is surprisingly good) 5 Plain text -> #PKM tool 6 Repeat Skip steps 4 and 5 for 100% offline & private learning. You may learn better this way, & what you'll learn will be nobody's business. DOIs: 10.1101/2023.08.30.553693 &...

This story is well-documented: any parent working for a social media company is likely to restrict the way their own kids use social media. They want none of that on their kids' screens. Early days for "AI" still, but an interesting thought experiment nonetheless: If you took a sample of executives & employees from the top 100 "AI" companies, how many would you see using & benefiting from their product in their day jobs? How many would let their kids hang out in the worlds they created? ...

My big 2024 plan: Work Offline. It can mean many things. Here are ten interpretations. 1 With my browser closed 2 In a plain-text file 3 In airplane mode 4 On old devices 5 On paper But also: 6 Asking questions face to face 7 On the road, on my bike 8 Alone, just sitting, breathing 9 Somewhere I've never been 10 Outside the comfort zone "Online" is going to change so fast, & not for the better. Time to figure out what your "offline" really does for you. (You do have an "offline", right?...

Daft Social exists, is free to use and has the energy of an Ello/Posterous elegy combined with a punk zine's middle finger and a Zombo.com-style grin. "Kind Words" exists, costs $5 on Steam and is the latest game to break my heart with real human connection. Every time a Big Platform breaks something for us on purpose, you grieve and fume a little. But then, sometimes, you code back. "Oh, the periphery They throw good parties there" (Fiona Apple) ...

These articles get shared often by my English teacher colleagues: "These words make it obvious that your text was written by AI". I looked through the list & call bullshit. I - & many of my colleagues - taught most of these phrases to IELTS learners. There are no shortcuts to learning writing. No "skeleton keys" to detecting machine writing, either. If you can't see that by stigmatising "AI phrases" you stigmatise learner interlanguage - then you don't really "get" learning, human or mac...

Kids read differently on screens & on paper. Their brains seem to want to go deeper when reading paper material; they stay shallow when reading screens. The temptation here is to engineer your way to a solution. Build a more paper-like screen. Or a more engaging one. The big question, with no easy answers, is: Since neither screens nor paper sheets come as "nature" to them, how did we end up nurturing the "screen = skippable, paper = important" mindsets? //doi.org/10.1101/2023.08.30.553693 ...

I've had an on-and-off thing with GTD for over 15 years. I've tried, and paid for, more apps and plugins than anyone needs in their lifetimes. Every January meant a new paper planner with a new promise, and every February meant a new hangover after the planner got neglected. Funny thing, among all this - I was doing so well. I had jobs, and jobs upon jobs. Bringing home the bacon, you know? Living the dream. People's projects kept coming, and I kept saying "yes", because I always felt like work...

There are two types of language learning stories that are relatively easy to write, whether you're writing books or blogs, articles or policy papers. First, there's the case of the relentlessly positive author, whose mission seems to be to list all the ways in which being a language learner changes your life and the world around you for the better. And then, there's the bitter and suspicious voice, out to prove that all things multilingual, multicultural and multifaceted are somehow too dangegou...

My VPN app upgraded itself to Version 4. fanfare V4 was an 850 megabyte download, constantly used 200-300MBs of my RAM, and no longer came with the command-line interface (CLI). sad trombone I decided to uninstall v4 and re-download the cli. It's a 30 megabyte download and has no RAM footprint to speak of. Some things are worth every bell & whistle you throw at it. Others are just plumbing: ugly AF and oh-so-crucial to get right. Know which ones you're building / maintaining. ...

Yup, you could try to hack at Google's #algorithm. LinkedIn's. X's. Do what you're told, when you're told, with the right bells & whistles. Feels good to catch that wave. Then you're off it, and that's a bummer. That's when we buy ads to get back on - to bypass the lineup. "Algorithm" means "A finite sequence of rigorous instructions, typically used to solve a class of specific problems". You have those. For work. For study. For play. For art. For decades. Hack these instead. Be your own puz...

A survey of community college students in the US, adapting to remote study, has intriguing data for #LearningDesign. Only a minority of these reported 0 tech issues. For most, #OnlineLearning was a slow, clunky experience, draining their confidence. Our #EdTech remains tied to privilege - our starting place is where RAM is cheap, broadband is reliable, and coding small isn't necessary. The learners let down by this live just outside Harvard or MIT. Build for them. DOI 10.1007/s11528-021-00587-8 ...

Each Cruse car needed to be helped out, on average, every 2.5 miles. Cruse employed more remote employees to fix this than it would have drivers. All so we could live out our Jetsons fantasy. Each shiny and polished OpenAI product had to be scraped clean of profanity, bias, and racism at prototype stage by remote workers in Kenya, on a pittance. All so we could fantasise about the coming benevolent singularity. Each new neural network is trained by gig workers in developing countries, without ...

A few months ago, to manage burnout, I switched off every LinkedIn notification. Things were quiet for a while. Then a LinkedIn email in my inbox. A new feature, you see - hadn't been part of my purge. Gotcha! Then another one. Another new thing. Back to 1-2 emails, & 3-4 notifications per week. Rob Nixon (2011) defines "slow violence" as "[A]n attritional violence that is typically not viewed as violence at all". If your platform keeps reverting to slow violence - check for your nearest exi...

Backspace key on keyboard: stops working Capitalist self: "I was thinking of getting a new keyboard anyw - WHY DOES THIS ONE COST £259" Techbro self: "There must be a way to map an unused key to become Backsp - WHY IS DEBIAN SO COMPLICATED" Luddite self: takes screwdriver, yanks the Backspace key out, cleans out The Unspeakable Gunk Backspace key: works Self: disgusted but re-integrated Let this be a lesson about our order of priorities when fixing tech problems. Have a good weekend. :) ...

Going through the list of recent updates to WCAG, the web #accessibility guidelines, I smiled. The way #WCAG is shaping up is good to see - not just for web users with disabilities. It makes sure e.g. you don't have to enter the same info more than once. Or that CAPTCHAs aren't too devious. If you're serious about anything you build online, I suggest you start as close to WCAG as possible & move out from there. You'll be building helpful, humane tech for tired & overwhelmed users. Good s...

The reason why AI seems so good is the same reason why our sneakers / burgers / phones seem so cheap. It's someone else's labour. Unacknowledged, underpaid. Sure, soon enough the workers will get too expensive for the models to keep using them. But by then, the AI models will have their shiny feet in our door. If you can't see anyone doing the work on your content by then, chances are it's you. (Rowe, Niamh. “Millions of Workers Are Training AI Models for Pennies.” Wired, October 16, 2023.) ...

The first Noble Truth tells it simply: there's suffering. Duhkha. Unease. Standing unstable. A bumpy ride. In #DigitalLearning, it seems, it's paraphrased: there is social anxiety. Lots of it, everywhere you look. (DOI 10.1186/s41239-023-00419-0) There are always dozens of requirements for everything a #LearningDesigner builds, coming from all sides. Might as well add this thousand-year-old question: "will this here make learners more or less anxious?" Most rides are bumpy enough without us. ...

There is a paper out today suggesting that #SystemsThinking could learn from Darwin (DOI: 10.1073/pnas.2310223120) It concerns weather systems & astronomy, but what if the following were true about anything you build / design? "[T]he system will evolve... if many different configurations of the system undergo selection for one or more functions." One-size-fits-all stopped being reasonable in businesses, #EdTech, or anywhere. Build something else. Smaller. Hungrier. Or angrier. Watch it adapt...

Visual artists whose works have been scraped to train AI models are now building tools to protect their art. Glaze & Kudurru are two recent examples. With language, you'd be forgiven for thinking there's no such firewall. But meaningful language (as Ursula le Guin would argue) is always an event between people. Curiosity, contexts and personalities are all the defence you need. ...

Computer says NO. Computer says NOT ELIGIBLE. Computer says it will access your location even if you thought you told it not to. It's always been OK to talk back at it, you know. Except now it's even chattier, & better at bullshitting. Our ways of thinking about how we talk through / with computers are stuck in ivory towers of academic & technical English (DOI: 10.1002/9781118584194.ch6) We need plain ways of knowing how to talk back at computers before all but the nerdiest of us get sil...

It was surprising to me that algorithm and rhythm have unrelated etymologies (look them up; I shan't spoil the fun). The ideas are closely connected, if you let them be. It's easy to think of algorithms these days as these mysterious black boxes that somehow rule the machines we use. It's easy to forget about the codes and technologies we ourselves are coded by ( DOI: 10.1007/978-1-349-26624-1_7) The reason cyborgs exist is that organisms & machines are equally ruled by rhythms and algorithm...

The Balance app for meditation is an example of an app I've really enjoyed using recently. (I'm not affiliated) It's simple, versatile, and generous. The guided meditations adapt to the time you can spare, the skills you want to build, and the experience you have already. The app also mixes the visual + auditory learning well, giving you a micro-lesson on screen before delivering the rest of the meditation through your headphones. It's free to try for one year. Go search it out. ...

Researchers now begin to understand what meditation does to us (DOI: 10.1016/j.bbr.2018.08.023). There's another important effect, though. These 13-15 minutes spent with yourself are not spent doing other things. Research already suggests that social media use messes with your cortisol recovery (DOI: 10.3389/fpsyg.2017.01609). I've been replacing "doomscrolling" with "dudescrolling" before bedtime these days; sitting and watching the thoughts in my head unfold. Take care of your inner wiring. ...

According to data from Carta, the number of startups closing due to dissolution / bankruptcy in 2023 points to "the most difficult year for startups in at least a decade." Put that schadenfreude / indifference on ice for a minute & consider this: Many of these startups' employees used to devote time to keeping key repositories stable & user-friendly. They used to make sure the piping & plumbing worked for them. Now they're busy hustling for work. The web breaks slowly at first, then ...

We now know that pregnancy re-models a mother's brain (DOI:10.1126/science.adi0576). Hormones prepare the brain structure to begin responding to a world with new priorities. On a smaller, less intense scale, consider this: these kinds of changes happen to us every day. Joy, fear, pain, reward, pride, peer pressure - and all the hormones we're always awash with as a result... It would be easier to list things that don't have such effects. (Many of them are to be found at schools, ironically) ...

We were on our way home after viewing a house. "Sharp right," said the satnav. "Oh no," I thought. "That's the narrow lane again." 50 meters later, we came up against a line of cars, no way of passing them. I had to reverse these 50 meters. I feared and hated that moment. The house was lovely, but the area around it was built for big cars and confident drivers. Not me. Whatever you build - one of the better design questions you can ask is, "could I still use it if I was having a really bad day?"...