Webcurios 02/12/22

Reading Time: 40 minutes

This week I thought I would ask GPT-3’s new chatbot to pen the opening – this is what it came up with, based on my request that it ‘write me a humorous, cynical opening paragraph to a newsletter about interesting things on the internet’:

“Welcome to our weekly newsletter, where we share the most interesting things we’ve found on the internet (or at least the things we found interesting enough to click on). No guarantee of accuracy, truthfulness, or even goodness, but hopefully at least a little entertaining. This week: a bizarre conspiracy theory, a hilarious video, and an unexpected use for a chicken. Enjoy!”

You know what? FCUK YOU AI I AM BETTER THAN YOU LIKE FCUK CAN YOU MANAGE MY UNIQUE AND IRREPRESSIBLE STYLISTIC FLOURISHES AND BOLD USE OF CAPITALISATION I AM A SPECIAL AND UNIQUE HUMAN SNOWFLAKE AND DON’T YOU FORGET IT.

(next week I will train it on the entirety of the Curios corpus and then kill myself as I am rendered otiose by machine).

I am still Matt, this is still Web Curios, and I am increasingly feeling like I would rather like to be replaced with AI if I’m honest with you.

By Miki Kim

WE START THIS WEEK’S MIXES OFF WITH A RATHER WONDERFUL THREE-HOUR SET OF PLEASINGLY-LISTENABLE HOUSE FROM LEE BURRIDGE!

THE SECTION WHICH IS CURRENTLY SPECULATING AS TO WHICH OF THE CURRENT CROP OF HORRIBLE PLUTOCRATS IT WOULD LEAST LIKE TO ALLOW TO STICK A BUNCH OF ELECTRODES INTO ITS GREY MATTER, PT.1:

  • GPT3 Chat: It’s that time of year in which agencies around the world desperately flail to get their ‘TREND PREDICTIONS FOR NEXT YEAR’ presentations out of the way, which means that many of you will either be reading or writing a bunch of pointless crap that says things like ‘We predict that AI will be big in 2023!’ I doubt, though, that any agencies will be quite honest enough to admit that, based on this link, what will actually happen in 2023 is that anyone with an ounce of sense will start outsourcing 90% of the shovelware content creation work that forms so much of agencies’ bread and butter these days to the machines – because, honestly, based on this, no fcuker will be able to tell. OK, fine, perhaps I’m being slightly hyperbolic, but if you’ve yet to check out OpenAI’s new GPT3-based chatbot then please click the link and have a play and come back here and continue reading once your mind has been properly blown. This is, quite frankly, fcuking ASTONISHING – it’s based on the tweaked new GPT-3 model called DaVinci03 which OpenAI released earlier this week and which you can read more about here, but the main takeaway is that there’s now a convenient, in browser toy with a conversational interface which you can command to write whatever you want for you, and it is SO GOOD. Honestly, just try it on something like ‘can you write me a 600 word article on the history of Facebook?’ or ‘what are the basic tenets of confucianism?’ or even moderately-esoteric stuff, like ‘can you write me a romantic sonnet expressing my love to a person called Calliope, which specifically references her peculiar love of millinery and which contains a marriage proposal?’ – it NAILS it. Honestly, I was a bit sniffy earlier this week when messing around with the base-level new model, but the addition of the interface is game-changingly good. The potential implications for this are insane, but, at the very least, this feels like it’s ushering in the end of ‘kids, go home and write me an essay’-type homework – honestly, a busy teacher would be hard-pressed to tell the output of this machine from a standard C+/B- student’s work – as well as being the first technological innovation that made me think that the era of most written content online being machine-penned is almost upon us. I can’t stress enough how astonishingly near-magical this feels, and also how deeply unsettling, like we’re standing on the edge of something that may well be a terrifying precipice full of as-yet-unknown monsters. This unrolled Twitter thread gives a decent overview of some of the things which you can do with this, and the things which it still struggles with – you may find the software’s inability to do anything which even halfway looks like ‘reasoning’ somewhat comforting.
  • Chat With Your Inner Child: Of course, human nature being what it is, this new tech almost immediately saw someone somewhere online doing something…ever so slightly odd with it. Step forward Michelle Huang, who went moderately-viral earlier in the week as a result of her having used this new tech to, er, have a ‘conversation’ with a specially-trained version of the new GPT-3 software which she had ‘fed’ a bunch of text from her childhood diaries – said training meant she was then able to ‘converse’ with a software model which responded to her in a style reminiscent of her as a child. Which, I think, is such an insanely-future idea that we should take a moment to really chew over how mad it sounds – someone trained a computer on their childhood diaries so that they could subsequently have a conversation with said computer as though they were talking to a past version of themselves. Honestly, this is a MIND-FLAYINGLY ODD concept and I am simultaneously deeply-fascinated and slightly troubled by it – there’s another thread here covering some of the reasons why this might not be a wholly psychologically healthy thing to do, but, well, who am I to tell you not to create a digital ghost of your childhood self? NO FCUKER, etc etc.
  • Fabled: Do you think David Walliams’ recent spate of negative publicity means that maybe, perhaps, some other people (maybe non-famous ones?) might be allowed a shot at selling some kids books? Who knows, but let’s hope so as time may well be running out for authors of reading material for small folk – whilst I don’t expect AI-generated novels to be troubling actual novelists anytime soon (although on reflection, given the quality of some of the stuff self-published on the Kindle store, this may in fact happen sooner than I expect), it does rather feel like the slightly more formulaic childrens’ book market could well see AI make serious inroads. Welcome, then, to Fabled, which uses a combination of AI tools to spin out stories for children based on simple prompts (or at least it will – it’s currently in beta, but you can apply for waitlist access) – “Instantly create custom books about any topic or theme! Make personalized kids stories, baby books, adult novels or fan-fiction. Simply give us a sentence and our AI writer will do the rest.” It will churn out accompanying illustrations too, letting you specify an art style to direct the aesthetic however you choose – honestly, I look at this and I think ‘hm, those companies that churn out those generic ‘your kid’s name in a storybook!’ books are going to have a field day with this stuff’, and also ‘if you’re a small-scale kids author this is going to feel like someone hammering the coffin down over your face’. Fabled isn’t alone – even Amazon is getting in on the act, with its new Alexa-based system that uses AI to create animated stories on-demand for its screen-enabled domestic surveillance devices – and whilst we’re evidently going to see some alarming headlines about ‘AI creates snuff story and terrifies children’ in the not-too-distant future, it also seems reasonably clear that the number of people able to make anything approaching a living from writing books for kids is going to take an absolutely massive fcuking hit in the next 24 months.
  • Instoried:. CONTENT! MORE CONTENT! The ceaseless demands of the content-hungry maw of the web are becoming ever more insistent, while the rewards for actually bothering to make anything become ever slimmer – which is why any sane advermarketingprdrone will be looking for ways in which they can eliminate themselves from as much of the pointless slog of CONTENT CREATION as possible. InStoried is one of a growing number of subscription services which promise to help you churn out the pointless words that make up 99% of the modern web – it will write straplines and blogposts and social media copy and, basically, all the crap, largely-pointless words that exist only to fill up the infinite spaces of the internet and to be read by the Google Spiders, and it’s keenly-priced at 250 quid a year for 20 licenses, and, honestly, how much did you spend on copywriters last year and was it REALLY worth it? I say this as someone who occasionally gets paid to write – I know, it’s RISIBLE, isn’t it? – and, honestly, most of the time I simply wouldn’t bother paying for a human these days because most corporate content is read by approximately 17 people and, frankly, who the fcuk cares if it’s machine-written or not? Which I appreciate is exactly the sort of attitude that will see us sinking in a swamp of machine-dreck within a few short years but, look, it’s been a long year and I am tired and part of me just wants to sink into it and just sort of suffocate.
  • NotContent: On a similar note, NotContent is an interesting idea – an agency which is quite open about the fact that it uses a combination of the AI stack to do its work faster and smarter and cheaper than the competition. I have no idea who these people are or whether there work is in fact any good, but am including the link because, seriously, there are going to be a LOT of these shops spinning up in the next 12 months and attempting to undercut larger agency structures and, honestly, having worked within large agency structures here and there over the risible mess that I like to call my ‘career’, it’s about time too. If you’re not using this stuff as a streamlining element of your creative process then you are probably soon going to be losing work to agencies who are, basically, simply because they will be able to be cheaper than you (and, almost certainly, because your work’s not special or exciting enough to warrant the ‘artisanal human premium’ that you’re charging).
  • GPTweet3: Not, to be clear, the actual name of this service, but allow me the terrible ‘gag’ – this is a little webtoy that will spin out tweets for you on any topic you tell it to, with any tone you specify, even mimicking the general feel of a specific other Twitter account should you wish. If your job involves having to fill out endless content calendars and having to draft an infinite number of pointless Twitter bromides that will one day be punted out to a largely-uncaring audience then a) I am so, so sorry that this is what your life has become; and b) this may well stave off the suicidal feelings for at least a week or two.
  • Natural Language Playlists: Ok, this doesn’t totally work but I like the idea and there’s something fun in the ‘fuzziness’ of the interface and what it delivers. The premise is simple – you write in a description of the rough sort of playlist you would like to create, click a button, and in a few short seconds this site will have created something for you from Spotify based on your specifications. I just gave it ‘happy jazz for a summer evening’ (literally antonymical to my current mood and situation, as I sit in a dark and freezing cold kitchen with a hot water bottle stuffed under my jumper) and a quick glance at the tracklisting it’s given me suggests that it vaguely-understood the brief – I can’t guarantee that it will be similarly effective with more esoteric or complex requests (“A soundtrack that perfectly evokes the feeling of anxiety that precedes a major international football fixture which you expect to win but fear you will lose”, for example), but it’s definitely worth a play – I would be fascinated to know how this works under the hood.
  • Jodorowsky’s Tron: People have been playing around with the new version of Midjourney this week, and the consensus seems that it’s currently the leader of the AI art pack in terms of the quality and style of its outputs – you may have seen this selection of images, effectively imagining ‘what if Tron, but directed by infamously-batsh1t Mexican auteur Alejandro Jodorowsky?’ (the first batch of which are at the main link, the second which you can find here), but they’re worth examining up close because…well, because the machines are getting so much better so much faster than I would have expected. Contrast this with the stuff people were getting excited about six months ago and it’s astonishing to see the degree, and speed, of progress – there’s a nice selection of images in this post by Renee Walter which showcases some of the other ‘what if film x, directed by person y?’ mashups people have been playing with, and whilst, yes, fine, this stuff does tend to produce better results when doing what is basically style transfer work than when ‘imagining’ something entirely new, it’s hard not to see this as our first, tentative steps across a massive fcuking creative Rubicon.
  • AI Synths; This is just a Twitter thread, fine, but it struck me as a lovely little example of what it’s possible to do with all this stuff – AI art meddler Fabian Stelzer has created a series of odd, imaginary synths, dreamed up by machines to his specifications, and has animated them and then gotten another AI to imagine what music each of them might make, and I honestly get giddy at the thought of what people are going to be able to make with this stuff by this time next year.
  • Old London Photos: And then, of course, there’s the very real fact that if we think we’ve had a dodgy few years in terms of our inability to determine truth from falsehood online then we should probably brace ourselves because, honestly, we have seen NOTHING. These images are a case in point – created using Midjourney, they are a TERRIFYINGLY convincing selection of images which mimic with eerie accuracy the very particular visual style you see in old photographs that have been retouched and sharpened by digital imaging techniques.Except, of course, they are not – these are not real photos, the people in them never existed, and if you look closely then you will see that there are occasional anomalies in the photos such as the fact that at least one horse appears to be floating above the ground unaided by limbs. But, honestly, a cursory glance would see you convinced that these are proper old images which have just been digitally retouched a bit – as evidenced by the fact that they were happily accepted as real by the subReddit in which they were originally posted. We’re on the cusp of another popular panic about THE PERILS OF DEEPFAKES, except this time there will actually be something to worry about.
  • Woolitize: This is a version of StableDiffusion that has been trained on wool, and which therefore will create BEAUTIFUL images of whatever you like, as long as you don’t mind the fact that all the outputs will look…well, distinctly wooly. If you’ve ever wanted to see what your favourite pop star might look like if they were knitted by your nan then this is the model for YOU!
  • Cleo: WARNING: THIS LINK WILL MAKE YOUR LAPTOP SOUND LIKE IT HAS EMPHYSEMA. Or at least it will if it’s as sh1t as mine is – you may well have some sort of insanely-powerful beast which will render this with nary a complaint – but it’s sort-of worth it, if only to see what a digital recreation of a piece of interpretative dance all about the life of Cleopatra is like. I don’t really understand this – it’s contemporary dance, which to me is about as comprehensible as Urdu – but it’s a nice (if heavy) piece of digital work and I am glad it exists.
  • Wyth: I am including this website solely because it contains some of the most astonishingly-meaningless examples of copywank I have read in YEARS. If any of you can explain to me what the fcuk it is that this company does I will be hugely grateful – seriously, what in the name of Christ does this mean? “Wyth is designed to facilitate a virtuous cycle of experiences converging on a single platform. We call it the Circular Experience. The Circular Experience is Wyth’s human-centered approach to technology aiming to connect social, interactive and collaborative frameworks to enable deeper relationships in every aspect of our lives.” I mean, obviously it means nothing, but what is it MEANT to mean? Can we please start asking questions like this when we read copy this bad? Can we stop letting these cnuts get away with these crimes against language?
  • Powder World: “Powder World”, reads the onsite blurb, “is a lightweight simulation environment for understanding AI generalization.” Honestly, though, that MASSIVELY undersells it – what Powder World in fact is is a little browser-based sandbox that lets you experiment with all sorts of interlinked physics models in a tiny little pixel world. Drop vegetation everywhere! Set fire to it! Douse the flames with water! Cackle maniacally at the feeling of godline power that ensues! I mean, yes, ok, there is doubtless loads of really impressive maths happening here which I am totally glossing over in favour of cooing over the physics toy aspects of the whole thing but, well, I reckon you’re probably not reading this for my trenchant analyses of emergent phenomena simulation engines (or at least I hope you’re not, and that if you are that the disappointment isn’t too crushing).
  • Not Pink: Click the link. Remind you of anything? Yes, that’s right, someone has FINALLY ripped off MSCHF wholesale – this is a quite astonishing aesthetic lift (seriously, click this and check out their homepage and do a mental comparison and marvel at the chutzpah), and as far as I can tell the basic premise (mysterious brand punts out odd little creative game/experiment things on a semi-regular schedule) is the same. No Pink only have one project under their belt at the moment – a game in which they invited people to attempt to come up with the closest match possible to an AI created image, which is a fun-if-lightweight idea (lol like I have ever had a ‘heavyweight’ idea in my life) – and their next drops in 10 days and I think is going to be something involving AI and a popular Twitch streamer; worth watching, if only to see whether it’s possible to rip off the MSCHF concept wholesale or whether it only works if you’re a bunch of trust fund kids in Brooklyn.
  • Mocopi: This is super-impressive, at least in theory – this is new kit from Sony which comes in the form of a series of small sensors which you affix to your wrists, ankles and head and which, at least according to the launch blurb, will let you do full-body mocap using only said sensors and a smartphone. Which, obviously, is huge – it takes VTubing out of the bedroom and into the wild, for a start, but it also opens up the general mocap industry to anyone which, when allied with the boom in AI image generation and modelling, means that any kid with an imagination and a phone can theoretically create fluid mocapped animation with minimal spend and resources – I very much get the impression that we’re all going to be looking back at Avatar 2 in a decade’s time and laughing at the fact that Jim Cameron spent billions on creating space jungle smurfs when we were only about a year or so away from every single 12 year old being able to make the original Avatar using some of this kit and some Snapchat skins.
  • Mindmelt: This is GREAT – pick a music genre, streamed from some popular online radio station or another, and enjoy this frankly brilliant visualiser to accompany the sounds. I promise you, this is superb – hallucinatory and glitchy and straddling the gulf between abstract weirdness and odd metacultural commentary, and I promise it’s all I can do not to abandon Curios here at 825am and just stare at this, entranced, for the rest of the morning.

By Laura June Kirsch

WE CONTINUE THE HOUSE-Y THEME NOW WITH THIS EXCELLENT MIX BY JOHN TALABOT! 

THE SECTION WHICH IS CURRENTLY SPECULATING AS TO WHICH OF THE CURRENT CROP OF HORRIBLE PLUTOCRATS IT WOULD LEAST LIKE TO ALLOW TO STICK A BUNCH OF ELECTRODES INTO ITS GREY MATTER, PT.2:

  • Taking Shape: Whilst obviously there are some downsides to the current and coming boom in AI image generation and the associated tools – huge swathes of the artistic community potentially having their earning potential decimated by machines, the as-yet-unknowable impact it will have on visual styles and aesthetics, the as-yet-underexplored questions of bias and prejudice inherent in the models we’re building, etc etc etc ad infinitum – there are also some positives, for example the hope that this might see everyone’s least-favourite software behemoth Adobe lose at least some of their stranglehold on the digital art and design marketplace. It does feel rather like Adobe doesn’t quite get how potentially fcuked it is – this website, designed to showcase a bunch of modern design styles and how YOU TOO can recreate them by – SUPRISE! – investing heavily in the Adobe Creative Suite – doesn’t really acknowledge the extent to which there are, and certainly will be, a bunch of other, free, non-Adobe means of getting these sorts of looks and feels and outputs that don’t require you to sell your soul in perpetuity for an After Effects license. This isn’t a bad site per se, it’s just…pointless, and also doesn’t do what it says it will – the homepage suggests that it’s to ‘explore questions of modern art and design’, something which it singularly fails to engage with on any meaningful level. Still, at least it scrolls nicely.
  • In My Nature: Subtitled ‘A Multimedia Hike With Cyborgs’, this is a GORGEOUS project where three actual, real-life cyborgs (in the real sense, not just in the weak ‘I have a smartphone, you have a smartphone, we are ALL CYBORGS NOW’ sense) go on a hike and you can see how their augmented senses alter their experience of the world around them. “What is it like to choose your own senses? Join three cyborgs from Barcelona on a multimedia hike inspired by our interviews with them. They will show you how their electronic enhancements change what it means to be human.” You’ll probably recognise at least one of the three participants – Neil Harbisson is instantly-recognisable from his numerous TV appearances over the years, and the fact that he has an actual antenna poking out of his skull – but each has augmented their body in interesting ways, and each tells an interesting story through illustrations and prose and photography about why they have chosen to build on their existing senses in the way they have, and how their experience of ‘cyborg-ness’ alters their relationship to the natural world around them. After reading this, it is entirely possible that you too will want to install some weather fins in your head, and I say GO FOR IT.
  • Noggin Boss: There are many, many reasons I will never be rich, but one of the principal ones is that I literally have the opposite of business instincts – I am largely incapable of predicting what will be a popular success and what will not, and am as a result never going to end up being an early investor in surprise breakout consumer hits. Which is a longwinded way of saying that I think these hats (for that is what Noggin Boss is selling) are, by a long way, some of the most stupid-looking articles of clothing I have ever seen in my life, and I cannot for the life of me understand how the fcuk it is that they appear to be selling so many of them. How best to describe a Noggin Boss hat? Hm. Ok, imagine a baseball cap. Now imagine that said cap is enlarged to somewhere in the region of five times its original size. Now imagine said enlarged cap perched on the head of a normal-sized person. It looks ridiculous, doesn’t it? Yes, yes it does, and yet here we are. I think we’re probably too late for this to ship to the UK in time for Christmas, thank God, but if YOU want to be the first person in your postcode to wear a comically-outsize cap then this may be the link for you.
  • Count Things: I am THRILLED that this exists – this is an app that exists solely to help you count large numbers of uniform objects, at scale. Imagine, if you will, that you are a…I don’t know, a scaffolding magnate. You have a massive pile of scaffolding poles in a yard somewhere, which you want to accurately enumerate (this is a HIGH OCTANE thought experiment, isn’t it? Never let it be said that Web Curios doesn’t truly engage the mind!) but you don’t actually want to have to stand in front of said pile counting each pole one by one. Enter Count Things! This app contains all sorts of pre-built models to count common supplies and materials (bricks, tiles, all sorts of joists), and the people behind it will even work with you to create your own model if there’s something you want to count at scale that they don’t currently support (badgers, skulls, teeth, etc). The only thing about this that could possibly be improved would be a brand tie-in with Sesame Street – I want the number of girders delivered to me by a cuddly vampiric number obsessive.
  • Tidbyt: Look, I appreciate that times are hard and we are all feeling poor and parlous, and therefore the idea of pointing you at frivolous ways to spend what little disposable income you have left after you’ve dealt with the basic ‘not dying’ expenses is perhaps not exactly ‘on trend’ for the fag-end of 2022, but, well, I have no idea who any of you are and for all I know you might all be plutocratically rich thanks to judicious crypto investments (lol) and can happily drop the £180 (oh, ok, dollars, but still) to purchase your very own customisable artisanal pixel display screen thingy, which you can use to display everything from email alerts to the time to the weather to data about where your family members are on their commutes…basically if you’re a particular type of geek, the sort who likes messing around with a Raspberry Pi and who once owned one of Berg’s TinyPrinters, then you might well like this too.
  • Instafest: Yes, yes, I know that you have all done this thing already – that thing where you connect this website to your Spotify account and it spits out a festival lineup based on your listening habits, and you share it online to show what a fascinating and unique person you are based on your music tastes – but completeness compels me to include it here as well. I’ve been interested to see this doing the rounds far more than Spotify’s own ‘Wrapped’ content so far this year (although in fairness that’s possibly because Wrapped only came out late this week), which perhaps speaks to the fact that this shows you MORE STUFF and also perhaps because the Wrapped stuff this year is (imho at least) pretty horribly copywritten (you don’t need to fcuk with this stuff! We’re all narcissists who think that the world should care about our listening habits and would share this data even if you just gave us the artist names and nothing else, you don’t need to call us fcuking ‘astronauts’ for having listened to 17 hours of Diamanda Galas in 2022 ffs!). Anyway, I don’t fcuking care what music you have listened to this year so don’t fcuking tell me.
  • Pattern Collider:  “Pattern Collider is a tool for generating and exploring quasiperiodic tiling patterns. Every pattern that you create has a custom URL that you can bookmark & share.” This is slightly-hypnotic and geometrically-dazzling, and it made me want to sit and basically create my own intricate bathroom tile design (before I spent too long staring at the shapes and patterns and my visions started to go a bit funny).
  • TLDR: FULL DISCLOSURE: I have only given this a cursory play, and so I can’t either vouch for its brilliance or that using it won’t cause you some sort of significant professional or personal embarrassment. Still, the brief play I had with it made me think that there’s possibly utility here (and this is definitely only the tip of the iceberg for stuff like this – the obvious next extension is AI crawlers that do the same thing but without you even having to ask) – TLDR is a Chrome extension that effectively lets you batch-create AI generated summaries of any urls you point it at, whether they house text or audio or video, meaning that (in theory, at least) you could find a bunch of sources for research and then point the machine at them and get digestible notes on their contents rather than having to wade through X,000 words and hours of audio and YouTube videos. I wouldn’t, obviously rely on this for anything serious, but I can envisage scenarios in which it’s genuinely useful to get a top-level overview summary from a wide range of sources, if only to get a vague steer on an idea or concept. I probably wouldn’t trust this right at this moment, but it feels very much like a coming future.
  • Lungy: Look, full disclosure, I am including this link solely because I find the name of this app to be almost-impossibly awful. “Lungy”? LUNGY????? It sounds like the World Health Organisation’s ill-fated attempt to create a mascot to raise awareness of bronchial illnesses! It sounds like what you call the kid at school who once suffered such an extreme coughing fit that they dislodged a small piece of meat from inside themselves which landed on Melanie King and meant she was never quite the same again! WHO NAMED THIS? But, er, more seriously, it’s actually an app to help with breathing exercises, which, honestly, is probably quite useful and I shouldn’t mock. But, really, ‘Lungy’?!
  • Trendwatch 2016: A PERFECT Curio, this – Trendwatch 2016 is a YouTube channel which features 50-odd videos which are…just fcuking weird, honestly. I don’t want to tell you too much more, just click the link and dive in and immerse yourself in the strange. It feels…hm, how to describe it? It’s very much in the same aesthetic/vibe(sorry) space as the ‘infinite corridors of liminal space’ stuff, but with less of the self-conscious wackiness of a lot of that stuff, and I have no idea whatsoever who makes it or why it exists, or indeed why they posted two videos six years ago and then nothing else until February of this year, or where it is going. If you enjoy Scarfolk then you will enjoy this – they are not similar, but they also are (more high-quality critical cultural analysis coming up shortly!).
  • The Covid-19 Dream Journal: I can’t be bothered to dig it out, but in the early days of the pandemic I featured a link in here which invited people to share the dreams they were experiencing during the pandemic to see whether any clues could be drawn as to the impact of lockdown and isolation and the general sense of directionless fear that everyone was experiencing on the collective unconscious. I am not sure if this page, collecting people’s submitted Covid dreams, is the result of that particular project or something else, but if you’ve ever wanted to have a browse through all the things that people’s psyches threw up during The Great Plague of 2020-? then this will please you immoderately. Whether or not you believe that these are faithful recollections of people’s real dreams or a chance for people to indulge in a bit of creative writing is up to you – I haven’t gone through enough to see whether there are common themes that emerge, but I have read exactly enough to know that I am personally not that interested in hearing about other people’s dreams.
  • Oimo: A collection of beautifully-coded little browser-based physics toys, all by a person known as Saharan who I think is from Japan. Click on the ‘works’ tab and enjoy a couple of dozen beautiful little physics-y webtoys – I spent an embarrassing amount of time playing with the fabric model, but you will all find something to amuse you here if you look closely enough.
  • The Sheffield Tape Archive: Via the excellent ‘Things’ magazine comes this superb archive of music from Sheffield – “A series of archive recordings from around 1980 onwards: sheffield bands, demos, concerts and rarities..” Want to listen to a bootleg of a Butthole Surfers gig from 1987? YES YOU DO! Want to explore the work of the (to me at least) largely-unknown 80s beat combo called, magically, The Fcuk City Sh1tters? FILL YOUR BOOTS! There’s a wonderful mixture of the classic and the very obscure here, along with some proper historical curiosities like a campaign recording of Monster Raving Loony Party founder ‘Screaming’ Lord Sutch – a properly good archive collection, this.
  • Appropedia: If one of the things on your tentative list of ‘plans for 2023’ is ‘attempt to make my life more sustainable and perhaps less reliant on things like the crumbling British national infrastructure which it increasingly looks like failing entirely in the near future’, you may find Appropedia useful; it’s a Wiki which features all sorts of useful information on things like installing appropriate solar power, or living roofs, or proper composting, or bike-powered generators – in the main, I am including this link for my friend Ben who lives on a boat and I imagine will find all this stuff hugely useful (on reflection I could possibly have just texted him this link, but then what the fcuk would I put in the newsletter?).
  • The International Society of Antique Scale Collectors: Do YOU have a deep and abiding passion for antique weighing devices? Do YOU want to learn more about them, perhaps with a view to amassing a collection of these magical, wonderful devices? I mean, the answer to this is almost certainly ‘no, of course not, I have a full and rich life and I do not need to sully it with mechanical devices from The Old Times’, but just in case one of you is after a really obscure and ostensibly-incredibly-tedious hobby to pursue in the new year then, well, WELCOME TO YOUR SCALE-Y FUTURE!
  • Door: This is a website which hosts a load of music, with the following loose curatorial guide: “At the end of the 90s, we ripped albums that we found in physical stores and took them to the internet. It was during this era that we built a content channel with a noble purpose, that of listening. Soulseek’s directories were cities and “emigrate to a new land” was a common feeling. Back then, connecting to the Internet required a desktop computer, a good local provider, modem, and time. Life was concretely and cybernetically constituted,a division that no longer exists and -without automatic playlists or advertising- finding material was the product of research so the user was, at the very least, selective. With free internet on the streets and the advent of the smartphone, the latest generations are now easy recipients of unrequested information. All this, before touching a wire or having a thoughtful moment. Curated by romi, door is a music selection for listening and dancing in closed spaces.” I have listened to half a dozen of these over the course of the week, and each has been excellent in its own way – I highly recommend bookmarking this page and dipping in when you fancy some guided listening featuring some genuinely obscure (to me, at least) and eclectic stuff.
  • Internet Archaeology: Twitter’s wobbles over the past month or so have elicited some interesting conversations about the preservation of online ephemera, making this link a timely one. Internet Archaology is a project seeking to preserve not so much individual websites but the broader cultural artefacts that are created by our shared online experience – “The Internet is a network for contemporary cultural expressions. We act within the web and reproduce and produce cultures in the process — making it a virtual cultural space. The network also provides access to the artifacts and cultural assets of the past. Many museums, archives, or libraries have digitized large parts of their collections and made them available online. This very confluence of past and present represents our new digital heritage: We must protect and transmit it for future generations.” So, for example, exhibits so far include Kony2012, Harambe, a bottle of p1ss from an Amazon delivery driver…this isn’t perfect, but I think that there’s something genuinely interesting in how we seek to preserve movements and memes and general vibes of particular online moments and eras, and more experimentation and discussion of the best means of so doing is A Good Thing.
  • Planet Pizza: If you were to attempt to explain the concept of ‘pizza’ to aliens, how would you go about it? If you’re the creator of this website, you would do so via the medium of a wordless website that seeks to communicate the nature of humanity, maths and tomato sauce to a theoretical offworld presence. Why? WHY NOT??? “Pizza is simple, versatile, delicious, and most importantly, worldwide.  Every country in the world enjoys pizza to some degree; and, pizza is just “pizza” in most languages of the world, with only mild variations in the pronunciation.  As a species that strives for that sweet release of dopamine, pizza is a worldwide peacemaker and a perfect representation for one of the little joys which makes a human, human.  Earth very well may be the only planet in the Milky Way Galaxy, or maybe even the only planet in the Virgo Cluster of galaxies that enjoys pizza.  If extraterrestrials visit as friends, explorers, or researchers, it would be very beneficial to use this site to simultaneously describe broad sections of our math and science system as well as to teach the makeup of one of our planet’s most agreed upon pleasures.” This is pleasingly pointless and perfectly overthought.
  • Hindsight 2030: This looks like a rather fun idea, whether as a game to play for fun or as some sort of icebreaking/teambuilding exercise (god, that’s a depressing juxtaposition, isn’t it? Let’s just stick with the ‘fun’ usecase, shall we?) – “Hindsight 2030 is a quick and lighthearted game for exploring possible futures. Each player or team will pick one target headline for the year 2030 and then create a timeline, also made up of headlines, that shows how the world gets from today to that headline in 2030. Then, they will create one final headline, set after 2030, that serves as an epilogue.” The only potential downside here is the potential for it all to get very apocalyptic and a little depressing rather quickly, so perhaps start each game with a clear set of directional guidelines and some baseline rules (“Matt, why must your future headlines always revolve around crying robots on a dying planet?”).
  • Music Historian: Ooh, this is really fun. Pick any musical artist you can think of and this will tell you what their most-played song is (fine, based on Last.fm so limited data, but still), along with artists that the platform believes are similar to them – as a result, I can now happily inform you that Cannibal Corpse’s most-played song is their cheery little ditty ‘Hammer Smashed Face’, and that part of the universe of related music that Last.fm thinks exists around the band includes the incredibly-named bands Cattle Decapitation, Waking the Cadaver and, inexplicably, a country singer called Alan Jackson. This is a lot of fun, and feels like the basis of quite a fun online quiz game if you could be bothered to build one.
  • The Jerry Lawson Doodle: Time was I wouldn’t bother featuring a Google Doodle on here, but I figure that a relatively small number of us bother actually visiting the Google homepage these days and therefore won’t have seen this – this is a gorgeous bit of work, which “celebrates the 82nd birthday of Gerald “Jerry” Lawson, one of the fathers of modern gaming who led the team that developed the first home video gaming system with interchangeable game cartridges. The Doodle features games designed by three American guest artists and game designers: Davionne Gooden, Lauren Brown, and Momo Pixel.” Not only this, but it’s also a really robust little game creation engine which lets anyone who fancies it make their own simple browser games and share them with the world – honestly, this is a GREAT one to bookmark and point your pre-adolescent children at when you want 10 minutes of peace over the coming holidays.
  • Gourdlets: The last miscellaneous link of the week is a BEAUTY – Gourdlets is basically a tiny, pixel-y, browser-based Sim City clone, except there are no budgets and no goals and all you have to do is build whatever you fancy building. It’s a demo for a full game, but, honestly, there’s enough functionality in here to keep you happily citybuilding for hours, and there’s something super-charming about the art style and the way that the cute little inhabitants of your semi-urban paradise interact with your city and each other. This is GORGEOUS, and you should all play it.

By Jason Shulman

FINALLY IN THE MIXES, AND ESPECIALLY FOR ALEXANDER WHO ASKED FOR MORE JAZZ, HERE IS A LOVELY HOUR OF JAZZ AND LOUNGE MIXED BY THE APPROPRIATELY-NAMED JAZZ N PALMS! 

THE CIRCUS OF TUMBLRS!

  • Blombluss: I don’t really understand why this has been built as a Tumblr but, well, it has, so here it is! This is actually a quiz all about bad ports of classic videogames, so if you’ve been desperately craving a ten question quiz about old games and the obscure systems they may or may not have been ported to then this will please you no end (it will also please you if you were ever a devotee of the teletext videogames magazine ‘Digitiser’, to whom this owes a very obvious and welcome debt of stylistic gratitude).
  • MurdochHere: Last updated nine years ago, this Tumblr collects images of Rupert Murdoch spotted in the wild. I REALLY want to know why this was started and also why it stopped when it did – it seems to have been maintained by someone who worked closely with him, and I wonder whether he found out it existed and had poor Nathalie Ravitz killed (NB – Web Curios is in no way suggesting that Rupert Murdoch is the sort of man who would have someone murdered, honest).
  • Enchanted Book Art: Rare book illustrations, because who doesn’t love a rare book illustration? NO FCUKER, etc.

THE TROUGH OF (INSTA) FEEDS!

  • Howard Lee: If you’ve ever found yourself in an ASMR video rabbithole you may have come across videos of people painting things with an otherworldly degree of photorealistic precision – Howard Lee does that exact thing, taking photos of real life stuff, cutting a square out of the photo, and then painting in the blank til you can’t tell that there’s a point where the photo stops and the painting begins. Which, yes, I know, is a horrible description, but you’ll get it when you see it – this is mind-boggling, or at least it is to me.
  • Kevin Is In New York: Home Alone 2, presented on Insta in clips. I personally don’t really ‘get’ Home Alone (yes, I am a joyless husk, why do you ask?), nor indeed why anyone celebrates the markedly-inferior sequel which features Donald Trump, but should you be differently-minded then perhaps you will enjoy this.

LONG THINGS WHICH ARE LONG!

  • How Do We Create Digital Life?: Now, fine, your initial reaction to that question might well be “Jesus, Matt, why the fcuk would we want to do that? It’s bad enough being made of meat without inflicting the horror of existence on a blameless digital entity!”, but work with me here for a minute. This article is…interesting. I don’t necessarily believe that this stuff is as possible as the author seems to think, nor indeed that what he describes is necessarily tied in with crypto and web3, but the general thinking about what digital ‘life’ might look like, and how it might end up working, is fascinating (and, honestly, if you follow this thinking even a little beyond where this article takes you, not a little unsettling). Digital ‘life’, by the way, is defined by the author here as a digital entity characterised by the following: “It needs to be able to survive past its creator’s death; It needs to be able to reproduce; Its descendants need to be able to evolve; It needs to locally reduce entropy”, and the discussion of how such entities might be created and what they might be used for is genuinely interesting (but, again, also possibly a *touch* more unsettling than the rather pollyannaish tone adopted by the writer, and certainly not the sort of thing that we ought to pursue ‘just because we can’, which appears to be the general conclusion of the piece).
  • Where Is The Innovation?: Or, more accurately, where is the useful innovation? This was sent to me by Katie (thanks Katie!) and is a really interesting read about the degree to which so much new tech stuff from the past decade or so has left so little in terms of a meaningful mark on culture and society – technologies have made money, and financial bubbles, but there’s an increasing degree to which questions are being asked about what meaningful impact we’ll be left with once the froth dissipates. The piece looks at this question from the perspective of research and what can be done to improve its utility, but the basic premise is neatly summarised here: “Our question is whether newly hyped technologies, like the Metaverse, Web3, and blockchain, have any chance of changing this basic picture. There are many reasons to be skeptical that they can. In many ways, the Metaverse and Web3 are merely a pivot by Silicon Valley, an attempt to gain control of the technological narrative that is now spiral­ing downward, due to the huge start-up losses and the financial failure of the sharing economy and many new technologies. Huge start-up losses along with the small markets for new technologies have brought forth novel criticisms of Silicon Valley. If we are correct that the newest wave of hot technologies will do almost nothing to improve human welfare and productivity growth, then elected officials, policymakers, leaders in business and higher education, and ordinary citizens must begin to search for more fundamental solutions to our current economic and social ills.”
  • Minecraft Bots: The one area in which it seems fair to say that we can point at real technological process with tangible effects on culture and society is the emergent field of machine learning – it’s clear to anyone who’s paid any attention to this stuff over the past 24 months that we are building tools now that are going to completely reshape the way in which business and culture work in the not-too-distant future (although, amusingly, none of us have the faintest idea as to how!), and this essay, about OpenAI training machines on videos of people playing Minecraft to teach said machines how to do it, is an excellent example of how quickly this stuff is moving and what we might do with it. This is less about Minecraft than it is about the technique being used here as a step-change in how we go about this sort of training – YouTube provides us with an INCREDIBLE training set for any number of things, and the ability to effectively (yes, I know that I am massively oversimplifying this) point a bot at a YouTube channel and tell it to, basically, ‘learn’ is fcuking INCREDIBLE.
  • Galloway on AI: My standard Galloway link apology (“you’ve probably already seen this, but in case not…”) aside, this is a really good overview of why this stuff is so exciting and where we are at now and why it’s not just a fad (probably). If you know a lot about the field then you can skip this, but if you’re after a decent ‘state of the market’ overview for beginners then this is a clear and cogent bit of thinking/writing.
  • How Dall-E Works: This is SO GOOD. I have a…moderately-fuzzy understanding of how LLMs and image-generation AIs sort-of work, but this article (the first in a short series), explaining how OpenAI’s Dall-E manages to magick pictures out of thin air in simple terms and without resorting to HARD MATHS, was genuinely eye-opening. I promise you, if someone like me (who is so bad at maths that he understands literally nothing taught beyond GCSE level, and who once was reduced to near-tears by the mere concept of ‘logorithms’) can understand this then ANYONE can.
  • The GPT-3 Cookbook: This is SUCH a good resource, and is a hugely-useful overview of all the different things that you can use GPT-3 (or another LLM of your choice) for – it offers a short explanation of what an LLM is and how it works, but then is straight into explaining and describing a bunch of potential usecases, from getting it to write code to getting it to do translations, cleaning up copy to writing things from scratch. Read this and then work out all the ways in which you are going to be rendered obsolete in the next five years or so!
  • The Generative AI Revolution in Games: It upsets me how many times this year I have linked to the A16Z blog – but, in their defence, they do occasionally post stuff that is far more interesting and useful than the standard ‘thought leadership’ pabulum spat out by VCs, and does occasionally seem to contain actual examples of actual thinking. This, for example, is a good overview of some of the ways in which generative AI might usefully be employed by the games industry – whether or not you’re interested in videogames, it’s a useful read as an illustration of all the different ways in which the current AI tech stack might disrupt (sorry) an industry, which might prompt you to have a think about what you do for a living and how it might be utterly upended by AIs (and, more positively, how you might exploit that for your personal gain).
  • Automation Drives Income Inequality: Which, fine, I expect we already knew, but it’s interesting to see this articulated so starkly in a new study from MIT, which shows “most of the growth in the wage gap since 1980 comes from automation displacing less-educated workers.” This shouldn’t come as a surprise, but it’s useful to bear in mind after half a dozen pieces focusing on the coming wonder of the AI revolution: “A central conceptual point, Acemoglu says, is that automation should be regarded differently from other forms of innovation, with its own distinct effects in workplaces, and not just lumped in as part of a broader trend toward the implementation of technology in everyday life generally. Consider again those self-checkout kiosks. Acemoglu calls these types of tools “so-so technology,” or “so-so automation,” because of the tradeoffs they contain: Such innovations are good for the corporate bottom line, bad for service-industry employees, and not hugely important in terms of overall productivity gains, the real marker of an innovation that may improve our overall quality of life. “Technological change that creates or increases industry productivity, or productivity of one type of labor, creates [those] large productivity gains but does not have huge distributional effects,” Acemoglu says. “In contrast, automation creates very large distributional effects and may not have big productivity effects.””
  • The Weird Right: This piece is about US politics and the US right specifically, but, as ever with this stuff, I think we ignore the extent to which we’re downstream of America to our peril. What’s interesting is the extent to which this weirdness is a post-Trump thing – “This is a different—though parallel—phenomenon from the often raucous, conspiratorial personality cult that surrounds Donald Trump and his devoted base. This new turn has predominantly manifested among the upper-class and college-educated right wing. Indeed, as Democratic strategist David Shor noted, as those with college degrees become more left leaning, the remaining conservatives have gotten “really very weird.” In this well-off cohort, there exists a mirror of the excesses often attributed to the college-educated left, fairly or unfairly: an aversion to mainstream values and an extreme militancy.The ascendant weird right will likely struggle to sell its deeply anti-patriotic vision to many voters. In these segments of the mostly young, online-influenced American right, the optimistic vision espoused by Ronald Reagan’s “morning in America” has been discarded. The elite educated right has moved even beyond the overt pessimism of Donald Trump’s “American carnage”—now disgust with equitable citizenship, personal liberty, and democratic self-governance is commonplace. Fed by an endless outrage cycle and a motivated and well-resourced donor class willing to pour money into increasingly reactionary think tanks like the avowedly anti-democratic Claremont Institute, right-wing thinkers and activists have begun to identify the foundational pillars of the United States itself with immorality and adopted a new fascination with medieval Catholicism and imported European extremisms. Today, the right has shed its American and conservative roots and seeks a radical shift—a national “refounding.” Indeed, leading right-wing intellectuals like John Daniel Davidson have said that “the conservative project has failed” and that people like them constitute the educated vanguard of a “revolutionary moment.”” Obviously the UK is different, and we have our own peculiarities and nutters, but keep an eye on the fringes over the next 6 months or so because I have a feeling we’re going to see some of this stuff replicated over here as well.
  • Why Fascism’s Coming Back: Feels appropriate to post this 12 hours after one of the world’s most famous people happily and openly stated that he thought Hitler had some good points – I think this piece by Umair Haq is a bit ‘fearmongering 101’, and I don’t for a second buy his ‘America as a last redoubt’ lines, but in general it’s hard not to look at the baseline analysis (to whit: fascism thrives in circumstances where significant numbers of people suffer a vertiginous fall in living standards simultaneously over a relatively short period of time) and think ‘hm, seems legit’.
  • Yolo Personal Finance: OR, “Young People Spend The Pain Away”, OR “Klarna Is Going To Be Looked Back On Just Like Wonga Currently Is” – here’s one for your 2023 trend decks (THEY ARE FCUKING SLIDES FOR THE LOVE OF GOD PLEASE CAN WE PUT THIS ‘DECK’ THING TO BED?)! In the face of worsening economic conditions and increasingly-precarious employment, and a world in which everything does rather seem to be falling apart at the edges and seams, are young people being frugal and tightening their purse strings? Are they fcuk! Instead, the author of this piece in Dazed asserts, it’s all about the frivolous yolo spending, whether motivated by a lack of planning ability or the very real belief that civilisation will collapse long before the time comes to pay back the debts. You can literally base

an entire strategy for any number of consumer brands on this article – it won’t be very good, fine, but it probably doesn’t matter!

  • The Amazon Adpocalypse: Or ‘no, it’s not your imagination, shopping on Amazon fcuking sucks now unless you know exactly what you want to buy’ – this is a look at how the transformation of Amazon from ‘business that sells everything’ to ‘a business that is based on cloud software and advertising which also happens to have an awful lot of warehouse space’ has meant a worse customer experience for pretty much everyone, sellers and buyers alike, and how it doesn’t matter because of the company’s size and near-indelible place in the popular consciousness as ‘the everything store’.
  • This Is England: The World Cup continues to underwhelm, but I enjoyed this piece in The Face about the weird feeling around the England national team and the strange evolution of the country’s relationship with the current incarnation of the Three Lions. These paragraphs in particular resonated with me: “It’s taking place when living in England itself feels like a laughing stock: three prime ministers in seven weeks, a public being relentlessly ripped off, a conveyor belt of cartoon villains who fall in and out of power, occasionally appearing on our feeds to say something like ​“heating is woke”, launch an assault on tofu, or to whip up excitement about sending people to Rwanda. Everyone’s knackered, everything’s divided. The Queen’s death revealed England’s identity issue. An island of myth and mantra, a teddy bear from Peru hoisted up as a grief totem. A monarchy that’s fading into insignificance, a nation mourning the future as much as it is the past. Screwfix posting its condolences. A king barking at his subjects over leaking fountain pens.”
  • TikTok and Bad Food: Or “How algorithms are unexpectedly messing with society in ways we are yet to fully understand, part x of an almost-infinite series” – this piece is about the rise of inexplicable/horrifying-looking food content on TikTok (did you see the onion water girl? You have to see the onion water girl), and how, yet again, human behaviour is being manipulated by algorithms we don’t comprehend in ways we can’t predict. This piece is, honestly, a fairly standard ‘hey, look, weird internet trend everyone’ bit of copy, but I think what’s more interesting is to think about where this leads us and what happens when an entire generation of humans grows up being exposed to food preparation only in the context of this sort of horrific ‘made for virality’ content – basically what I am asking is ‘what if there is a direct throughline from Onion Water Girl to everyone in 2062 being incapable of cooking anything other than the 30 canonical ChefClub recipes that have resulted in ⅓ of the world’s population now having type-2 diabetes?’
  • Greece Is Capitalism’s Petri Dish: This is SO interesting, and feels like a bigger story – this piece in VICE looks at the partnership between Volkswagen and the small Greek island of Astypalea, which has become a testing ground for VW’s electric cars, buses and scooters, and a whole app-based ecosystem to manage said vehicles, and how this works and whether or not it feels like a good thing to give a massive international conglomerate this sort of access to and control over public infrastructure. The article mentions other areas in Greece where similar things are being tried – and why not, right? The country’s still reeling from being fcuked 17 ways by the 2008 crash and why wouldn’t they sell off access to themselves as a market and a testbed for new tech to some deep-pocketed corporate partners? Except, of course, there are QUESTIONS over transparency and influence – but, perhaps most interestingly, the bigger questions should be around how common this stuff is likely to get. Think, for a second, about the UK right now, a country in which nothing seems to work and where the money to fix it no longer seems to exist. Do you seriously think that a UK government wouldn’t seriously consider an offer from a massive international tech business to help fix some of its infrastructural problems with experimental innovation, in exchange for, I don’t know, lots of lovely data and maybe some preferential treatment on future national contract bids? I have a sneaking suspicion that we will look back on PPI and PFI as being just the tip of the iceberg for ‘interesting’ corporate involvement in the national architecture.
  • The Problem with EVs: Look, obviously electric vehicles are better than petrol or diesel vehicles – but, er, they are also problematic in their own way, turns out. This is a typically-excellent piece in Rest of World which looks at the specific issues faced by Indonesia, one of the countries whose natural resources are being mined to meet demand for batteries and associated technologies used in electric vehicles, where the human and environmental costs associated with the rocketing demand for nickel and other elements and minerals in the EV trade are slowly coming to light. This is, I warn you, one of those articles that might make you feel less-than-positive about the direction of the world.
  • The New Jennicam:  You all remember Jennicam, right? The first EVER lifestreamer (or at least the first one that anyone had ever heard of), Jennifer Ringley broadcast herself in realtime, all the time, from a series of webcams around her apartment, showing her sleeping, showering, occasionally-fcuking…it was, at the time, an incredible phenomenon, a true species first in a way, and the precursor to the society of the streamed spectacle we now ‘enjoy’, and now others are following in her footsteps on Twitch. One such streamer is Emily, who for over a year now has been on Twitch 24/7, playing games and sleeping and eating and chatting – this piece profiles her and her stream, and inadvertently pinpoints the way in which our relationship with people online has changed in the 25-odd years since Jennicam started this whole thing (the professionalism, the money, the fact that there doesn’t appear to be any joy in the experience for Emily whatsoever) and the ways in which it hasn’t (creepiness and entitlement and the fact that this doesn’t seem in any way a healthy way to live, for anyone involved).
  • Brand Restaurants: I had NO IDEA that Gucci has a restaurant, let alone that said restaurant (in Tokyo, in case you were minded to visit) has this year been awarded a Michelin star, but I rather enjoyed this piece about the surprising number of high-end brands that have experimented with opening eateries as part of the infinite, multidirectional expansion of their brand identity. As the piece notes, these restaurants will all lose violent quantities of money, so the benefits are far more intangible – but I wouldn’t be surprised to see this sort of thing becoming more commonplace as brands (and the awful people like me who occasionally advise them) continue to seek offline means of connecting with audiences in an increasingly-online-only world.
  • CoreCore: In some respects this piece (via Shardcore – thanks, Shardcore!) is total bullsh1t (honestly, read it and try and make it make sense beyond ‘here is an aesthetic that seems almost deliberately non-aesthetic but does in fact have some rough unifying principles that we’re not really capable of articulating’), but on the other it feels like there’s a certain truth it’s pointing at but not quite articulating. The overall gist is that we’re seeing a move towards more obviously, deliberately ugly, scruffy, tonally…wrong, or baffling, content on TikTok, and that this is part of a movement of sorts – where I think this is interesting and half-right is that I think this is a factor of platform maturity; you saw this on Insta too, as it reached saturation, with the growth of collage-y cut-up aesthetics and rough, zine-ish graphics and a general sense of wanting to kick against a prevailing aesthetic orthodoxy, and it feels like a lot of the stuff described (and embedded) in the piece is the same sort of thing but for TikTok. Which, if we follow this to its logical conclusion, suggests TikTok will be a cultural deadzone in about two years time, just like Insta has become.
  • Japanese Web Design: This has been everywhere in the past week or so, but with good reason – this is SO INTERESTING, and such a clever use of data analysis, to explain how Japanese website design is peculiar to that nation, and why that might be. Again, even if the subject matter doesn’t immediately grab you, this is just a really good example of storytelling and argument-building using data.
  • Cruising The Village: OK, this is a bit of a peculiar one, but bear with me here. This is the Phd thesis of Michael Atkins from 2013, which I stumbled across online this week and which is all about the practice of cruising amongst gay men in the city. Which may or may not interest you as a topic, but which is rendered SUPERB by the fact that all of Atkins’ interviews with various men on the subject are represented here in comic form, to help protect the identities of the people he spoke to. Honestly, I wasn’t expected to find this so interesting, and so moving, and I ended up reading far more of this than I expected to. Skim it to find the comics, but it’s worth dipping into the writing too as it’s a fascinating portrait of a culture, and of a less-online time.
  • Hello, World!: Sheila Heti is one of my favourite novelists, and I would read her shopping lists given the chance; this is a really interesting formal experiment, in which she conducts a series of conversations with various AI chatbots and transcribes them. This links to part one of five – this is LONG, but if you’ve any interest whatsoever in either AI and the idea of ‘sentience’ in machines, or in the manner in which these sorts of tools can form part of the authorial process, or indeed in the extent to which they can be useful interlocutors from the point of view of self-discovery, then it’s pretty much required reading. I don’t think it totally ‘works’ as an exercise – though of course it depends on what you perceive the aim of the exercise to have been – but I found it fascinating to read.
  • Insensible Loss: Surgery and sickness – a doctor writes about the relationship between their work and their sense of their won body, and the extent to which one impacts and affects the other, and, honestly, this is so so so so so good. Look: “I spend my days and nights invading strangers’ bodies. I put my finger in their mouths and gag them, feeling for firmness suggesting tumors; I snake a thin camera through their nostrils and into their throats to see the organ that gives these strangers their voices. I assess bodies impassively, looking for infection, inflammation, hidden cancer. I cut bodies open, dissecting them layer by layer; I suture them back together again using needles curved like hawks’ talons. Strangers’ bodies are where I work and where I learn. I know things about my patients’ bodies that they don’t know and can’t see. I know things about them that they have no reason to know. Some people want to see inside themselves, and they ask if their operations can be filmed, or if they can take their tumors home. “Where do my tonsils go after the operation?” one patient asked me during my training. “Do you keep them in a jar, and once the jar is full, you get to graduate?””
  • 52 Things: We close this week’s longreads with the 2022 edition of Tom Whitwell’s ‘52 Things I Learned This Year’, which once again is fascinating, baffling, funny, sad and sort-of terrifying, all at once, You will learn more from reading these, and thinking about them a bit, than you will from reading three million agency trends presentations, I promise you. My personal favourite this year is number 44: “In March 1967, the CIA tested Acoustic Kitty, a live cat with a microphone, battery and antenna surgically implanted. Sadly, on its first public trial, the unfortunate animal was run over by a taxi.”

By Mark Macevoy

AND NOW, MOVING PICTURES AND SOUNDS!: