Webcurios 23/09/22

Reading Time: 37 minutes

Well that was a fortnight. Have you gotten over it? Good, it looked weird tbh and I wasn’t a fan.

Still, here we all are again, forelocks freshly-denuded and ready to take a long runup at THE AUTUMN! Seasons of mist and mellow fruitfulness and all that Keatsian jazz, where we all get to try the exciting new national pastime of ‘how many jumpers can I wear simultaneously before I admit defeat and remortgage to put the heating on for 10m?’

I am still Matt, this is still Web Curios, and this week I saw a man inject his own semen into a dead squirrel which, I think, means I definitively lose whatever game it is we think we’re all playing.

By Mark Liam Smith



  • Convexity Space: Ordinarily I like to kick off Curios with something beautiful or frightening or astonishing, but this week I thought I’d try and reflect my general state of bafflement with, well, almost everything by sharing with you this borderline-incomprehensible site which, as far as I can tell, is meant to persuade you to apply for jobs. FLY THROUGH AN INFINITE DIGITAL SKYSCAPE! See the words ‘2x People, 4 x Impact – not a culture fit for everyone’ and wonder what they mean! Click and then zoom through a variety of…nonexistent constellations, each embodying maxims of what I presume are meant to be inspirational leadership excellence, like this fcuking PEACH of a quote: “Leaders don’t lead by giving orders. Everyone in the team is a leader. WE ONLY BOSS OURSELVES”. Apparently they ‘enable individuals to flow on a global scale’, which is nice (if entirely fcuking meaningless). Anyway, if you click on the ‘join’ button you will be taken to a…less-shiny webpage, where you can apply for such distinctly-underwhelming positions as ‘solution deployment manager’, although at no point was I able to determine exactly what the company does. If you’ve ever thought that the world is becoming more stupid by the day, this website will do nothing to disabuse you of that fear. I love it, and would like to shake the hands of the people involved in its design and execution (and, especially, those of the people who sold it because WELL DONE).
  • D-ID: Deepfakes haven’t quite become the terrifying, fabric-of-reality-altering tool of misinformation that some feared they would when they were first realistically mooted 3-4 years ago (I am going to avoid going back and checking whether I was part of that ‘some’ because I dislike myself enough already thankyouverymuchindeed), but that hasn’t stopped the ‘choose and manipulate your own digital meatpuppet’ industry from ploughing on with the innovation. D-ID is a company that offers you the opportunity to create videos featuring synthesised people speaking synthesised words – you can either pick one of their own ‘digital humans’, or upload photos of anyone you like and create a ‘video’ out of said photo, and the speechsynth software has a number of different voices you can use or alternatively you can upload your own audio to be synced with the output vids. It’s…not great, but it’s significantly better than I expected it to be, and made me wonder exactly how far away we are from digital spokespeople being an actual, viable solution for real businesses (24m minimum imho). Anyway, the fun bit here is that you can make the digital puppets say whatever you like, so this is an excellent tool for creating very, very offensive messages to send to your friends, family and colleagues (NB – Web Curios as ever takes no responsibility for whatever HR-related misadventures may result from following this advice).
  • Another AI Comic: Another ‘images by Midjourney, words by a human’ comic, this time by UrsulaV and presented in a Twitter thread, and with the added bonus of this excellent additional thread which explains the creative process and the work that went into producing the final pages and panels. This is, to my mind, a better bit of work than the one I featured a few weeks back, perhaps because of the degree of curation and tweaking that Ursula subjected the initial machine output to. This may prove mildly reassuring to those of you currently suffering from ‘this is the end of low-to-mid-level human creative output’ fear, although the fact that the direction of travel for this stuff is unidirectional and it’s getting better by the hour is probably not worth dwelling on.
  • Using Dall-E for 3d Modelling: This is a short-but-interesting thread exploring how animators at Studio Yatta used Dall-E to help them work up some 3d modeled animation. I think, basically, I am feeling subconsciously slightly guilty for all my ‘THE AI GODS ARE COMING, ALL ARTISTS WILL BE SACRIFICED TO THE GREAT TECH REVOLUTION!’ rhetoric of the past few months and feel the need to show you that this stuff is complementary not substitutative, honest kids! Anyway, this is a nice, simple set of examples about how you can use these tools for visual inspiration and to shortcut prototyping and all sorts of other things (until, to repeat, they get good enough to do all this themselves in the next year or so) (dammit, I really can’t help myself it seems).
  • Mother Goods: You may have seen the ‘Ink Made With The Blood of Gay Men’ stunt (designed to protest against the ban on blood donations by gay men in the US) doing the rounds this week, by Mother (and artist Stuart Semple) – it’s being sold through Mother Goods is the agency’s offshoot business which basically acts as a product skunkworks to make experimental activist-type activations. Or, maybe more accurately, it’s the bit of Mother that looked at MSCHF and thought ‘sh1t, that’s exactly the sort of stuff we should be doing, we’d better rip it off (but with a worthier angle because, remember, ADVERTISING AND SPECIFICALLY THE SORT OF CREATIVITY EMBODIED BY AD AGENCIES IS THE ONLY MEANS BY WHICH LARGE-SCALE SOCIAL PROBLEMS CAN BE SOLVED!)’. OK, fine, maybe that’s a touch unfair – there are some fun projects on here, after all – but, equally, there are some…not insignificant parallels between the two businesses, not least the website here being eerily similar in feel to the MSCHF setup, and the timed countdowns to new ‘drops’…imitation’s the sincerest form of flattery, right guys?
  • Sounds of the Earth: If the Earth was capable of making a sound (I mean, obviously it is capable of making sounds, but let’s imagine in a more personified way), what sound would it make? I think a sort of resigned sigh, personally. Still, this website is trying to use AI to come up with a different, hopefully more pleasant answer – it’s the digital component of an artwork by Yuri Suzuki, currently being exhibited in Milan, which invites anyone to upload an audio file from wherever they are (sounds of nature, sounds of people, sounds of the world). AI then seeks to stitch these clips into a coherent soundscape, linking sonic fragments with degrees of aural similarity into an infinite sound canvas of recordings from around the world, linked only by a degree of tonal assonance. This is rather beautiful, and it’s on in the background as I type, and it’s the sort of thing I can vaguely imagine using as a soundtrack to a flotation tank session were the concept of a flotation tank session about as appealing to me as an invasive subcutaneous sandpapering.
  • The Drone Photography Awards 2022: Photographs! Of the Earth! From above! Taken by drones! You know the drill with these by now – the winning photo here really is spectacular, though, and a perfect example of the sort of shot that would be literally iumpossible without the assistance of flying robot cameras. I’m curious with this stuff how many (if any) of these have been taken by AI rather than by a human operator – I presume that many drones now will come with ‘image assessment’ tech, and at least a rudimentary ability to ‘pick’ a shot from a camera based on perceived aesthetics, and whether or not there are any prizes extant that specifically allow for non-human shot selections to be submitted. Actually, that’s an interesting idea for a photo competition – given we’re almost certainly going to see the first raft of AI-generated art prizes in 2023, it makes sense that we should also consider the same for AI-selected/snapped photos. Anyway, there are some amazing photos here, with the caveat that I am personally slightly over the whole ‘if you look at it from above it looks super-geometric!’ style of photography.
  • Have I Been Trained: A tool developed by Holly Herndon to allow artists and creators to check whether their works have been incorporated into the training data for AI image creation models, specifically checking against the LAION-5B training set which is what StableDiffusion (amongst other models) used to develop its style. This is part of Herndon’s wider project, called Spawning, which “is building tools for artist ownership of their training data, allowing them to opt into or opt out of the training of large AI models, set permissions on how their style and likeness is used, and offer their own models to the public. We believe that each artist ought to have the tools to make their own decisions about how their data is used. Some may choose to take the permissive IP approach to AI models we pioneered with Holly+, where Holly offered her voice model for others to use in return for a share of profits in officially approved derivative works. Others may choose other approaches. However we need to establish a standard of consent honored by research organizations to get there.” Which is eminently sensible and necessary – this is very early days, and there’s little information as to how this will work in practice, but it’s vital that someone thinks about this stuff now (too late as it already is) because otherwise people will wake up in a few years’ time to find that the rights they might have hoped to have over their work will have been eaten by the future with no recourse whatsoever. Artists can apply to sign up to Spawning by sharing a link to their work – if you make stuff and put it on the internet and like the idea of having at least some theoretical agency over what happens to it and what it gets used for in the future, this seems like a no-brainer.
  • This Girl Does Not Exist: This is, I think, a WORLD FIRST – a videogame purporting to be written, voiced and art directed entirely by AI. It’s a TINY bit of a fudge, I think, in the sense that the ‘game’ element here is pretty minimal and VERY off-the-shelf, but as a precursor of ‘stuff that is definitely going to happen before too long’, all the art assets and vague ‘story’ copy have been spat out by machines. Here’s the Steam gameplay description – as you can tell, this isn’t exactly Fortnite: “This Girl Does Not Exist is the first game of its kind. It is a game full of beautiful girls .. and none of them exist! They were not made by a human but generated by an AI. Dive into this vibrant colorful puzzle game and add girls to your gallery! Relaxing gameplay with beautiful artwork and relaxing music will help you calm down after a hard day at work or study. Everything you will see in this game, including all the art, all the characters, story and even a voice acting – was generated by an AI. Your task is to put together puzzle pieces of beautiful girls as you progress through dating them. In each stage you choose which girls you like and would wish to progress with further. Every girl has her personality and unique information to uncover. The more stages you complete, the more girls are revealed in your gallery!” On the one hand, this is exactly the sort of low-rent shovelware that would previously almost certainly have used stolen images of models from somewhere on the web as art assets, so at least noone’s getting ripped off; on the other, I would be amazed if the vast, vast majority of hidden object / puzzle games on Android in 12 months don’t use exactly these sorts of assets because, well, why not?
  • Prompt Finder: OK, technically this site is called ‘Generrated’, but prompt finder is a better and more helpful description so that’s what I am going with. Basically this lets you browse a HUGE number of AI=generated pics to see what prompts were used to get them, basically shortcutting the whole tedious ‘prompt engineering’ (NOT GOING TO BE A JOB FFS) process. “Click on an image to see the prompt that was used to generate it, along with the 3 additional images generated alongside it. You can also click on the prompt text to navigate to a page that contains all 20 images generated for that prompt.” What’s really interesting here is delving into the areas which I haven’t played with at all yet – you can do some pretty decent vector work with the machines, turns out, and I don’t think artists who design app store icons are going to be getting much work in the foreseeable future judging by how good the machines seem to be at spitting out thumbnails. This is very interesting indeed, and worth a bit of a spelunk if you’re curious about what all this stuff can do and how to do it better.
  • The Designer:Dresses and outfits, imagined by AI. Most of these are frankly a bit crap, if I’m honest, but there’s definitely an interesting angle in the potential for them as starting sketches. Also, there’s something fascinating about the training set here – SO MUCH LACE! Why so much lace?
  • NonExistent Tory: You know how some people just look Tory? Something about the broken blood vessels and watery eyes? Well it turns out that it’s not just the Crungus and Loab who lurk in latent space, it’s Conservative Party members too. This Twitter account exists solely to share machine-imagined faces who are basically identifiable as classic Tories. Or at least they are according to whoever it is running the account – your mileage may vary. Personally-speaking I don’t think there’s enough rosacea on display here, but that’s just me.
  • The Heywood Prize: I am disappointed in myself that I didn’t know about this already – “The Heywood Prize was created by the Heywood Foundation. Jeremy Heywood served many Prime Ministers, and was the Cabinet Secretary UK’s most senior Civil Servant – until 2018. Jeremy was extremely open to new ideas, and sought out alternative perspectives. He was more interested in the quality of an idea than the rank or seniority of the person who proposed it. He would make a point of getting out of Whitehall to spend time in ‘frontline’ settings, from job centres to charities, to seek out innovations and unusual perspectives. The Heywood Prize continues that spirit of inclusivity and innovation.” There is £25k up for grabs here – second and third prizes of £10k and £5k too – for the best answer to the question “What do you think the government should do to improve life in the UK?”, which answers can take the form of either a 1500 word essay, or a 3-minute video or audio clip. Honestly, this is SO INTERESTING – I would pay actual cashmoney to be able to read the submissions because, well, there is BOUND to be some wonderfully mad stuff in there, and the fact that the winning ideas will get seen and scrutinised by people with a vague ability to actually one day enact them is incentive enough I think to get involved. Seriously, this is worth thinking about seriously – given how many of you work in advermarketingpr I can only IMAGINE the sorts of revolutionary and brilliant ideas you will all have! (sorry, that was rude, I do actually think you’re all smart, promise).
  • The Follower: You’ve all seen this by now, right? The Follower is a project by Belgian artist Dries Depoorter in which he uses AI image recognition technology to match CC TV footage of people taking shots for the ‘gram in public places with the actual shots as presented in-feed, thereby demonstrating the ease with which the surveillance panopticon can triangulate you in time and space based solely on your publicly-visible activity. It’s a very smart idea, although the coverage of it has been slightly more breathless than the project warranted – with the best will in the world, this is NOT an example of ‘creepy AI can spy on CCTV and then find you on Insta’, as Depoorter did quite a lot of the heavy lifting in terms of scraping, downloading, training, etc. Still, as with all of this stuff it’s less of a ‘wow, look at the now’ moment so much as a ‘wow, look at the near future’ but of cautionary advisory, as well as very much a ‘this is exactly what surveillance states are doing RIGHT NOW’ lesson. Also, for the right brand you could do a fcuking GREAT influencer campmaign using exactly this sort of tech, just FYI.
  • 4meric4: This is beautiful. Take a small, minimalist road trip across America with this website, which uses windows and frames pulling from other sites to create an imaginary journey for you complete with realtime radio pulled from stations you’ll pass through, images from your route, links to diners and cafes you might visit…honestly, this is small-but-utterly-perfect, and there’s something so so nice about the fact that your selections of vehicle, etc, persist inbetween sessions so you can check back on a daily basis and see your progress and where you’ve got to and what’s on the radio and what photos you’ve taken since you last checked in. ART.
  • Realtime: WARNING: THIS IS ADDICTIVE. I say this only because I just lost 5 minutes to slack-jawed gawping as LIFE scrolled past my unblinking eyes. Realtime is a simple premise – you get the web, in realtime, in one place – but it’s done really nicely. There are just three columns – pictures, words and numbers – each of which is a realyime infinite scroll, dumping new data at the top of each feed every second or so (you can adjust the pace of updates), and pulling in from sources as diverse as Google search data, Reddit, Billboard charts, Amazon, the BBC, Wikipedia edits…honestly, this is DIZZYING, an incredible, maddening, compelling torrent of PEOPLE ON THE INTERNET doing PEOPLE ON THE INTERNET things. Seriously, I could spend hours just staring at this and clicking intermittently, it’s that good. Although I’m not quite sure why it’s so obsessed with showing me mugshots of people who’ve just been arrested (although, on reflection, that might have something to do with the fact that it’s about 2am in the midwest as I type and it’s very much ‘methhead’o’clock’).
  • Summarize: One of the great annoyances of my professional life is that I need to keep across news about the big tech and social media companies, and as a result I often have to pay attention to podcasts and longform videos in which tediously-verbose (HAHAHAHA YES I KNOW) men with beards talk for two hours about the metaverse (to give you one soul-flayingly horrid example), and that there are NEVER any fcuking transcripts of said conversations meaning I need to listen to or watch the fcuking things. So thank God for Summarize, then, which offers a single service – plug in any url of any long YouTube video and Summarize will spit out a GPT-3 juiced summary of the main talking points per section. Obviously the main drawback of this is that you can’t really check how good it is without comparing its output to the actual content of one of said videos, which would mean actually watching one, which, frankly, no, but it’s an interesting idea and one which I will definitely be leaning on heavily (until I realise it doesn’t work and it makes me look very, very stupid).
  • Whisper: Seeing as we’re speaking about transcription (SEAMLESS!), it’s worth mentioning Whisper, released this week by OpenAI to minimal fanfare but which could well be as significant for journalists as GPT-3 has been. You need to be able to run it yourself, which obviously requires a bit of technical nous, but it purports to be a game-changing transcription service for audio, covering English and other languages (with varying degrees of success – OpenAI acknowledges that it’s less good at non-anglo languages at present due to the predictable differences in training set sizes). Given the number of journalists I see every day discussing whether or not there are any decent transcription services out there, this feels like it could be A Big Thing (and terrible news if you’re Otter or one of the few dozen extant subscription services doing transcription right now).
  • Smashomancy: PHONE MAGIC! Or possibly ‘MAJICK’! Either way, I rather like this – sort of like phrenology but for your phone (and about as meaningful, to be clear), Smashomancy is the practice of analysing the cracks on your phone screen to scry your future from the myriad galaxies in the fractured vitrine of your prized device. The site at present is a bit bare bones, but you can see a rough chart of meaning for the cracks in your screen, and there are promises of ‘a virtual experience’ coming soon. Unsurprisingly this is a joke that originated in San Francisco, where last month the people behind this apparently organised LIVE PHONE READINGS which, honestly, I could totally imagine people in London getting into too (judging by the sheer quantity of ‘mercury in retrograde’ chat I see spaffed across the TL on an hourly basis – WHAT DOES THAT EVEN MEAN FFS? Actually, on reflection, let me die in ignorance).
  • Atari Emails: I’ve definitely worked in places where it would be…uncomfortable should the email archives from said workplaces ever emerge (and frankly if you haven’t then you haven’t been doing work right imho), but there’s something really interesting about looking into how communications within a business embody a corporate culture (to the extent that such a thing can ever be said to truly exist, which I appreciate is on occasion debatable). This site collects emails from Atari, collected between 1983-1992 and covering the news, tech, culture, interpersonal relationships, in-jokes…there’s a particular beauty in reading words that the authors never expected to reach beyond the very small intended audience for whom they were originally penned, a nakedness of sorts, which I find personally hugely appealing, but anyone who’s interested either in the history of the workplace, or indeed of Atari specifically, will find things to enjoy in here (the thread on the Challenger disaster is amazing, for example, just in terms of ‘watching people parse a tragedy in semi-realtime’).
  • Super Realistic Animal Mask: You will need to translate this page into English from Japanese (er, unless you speak Japanese, obvs – sorry, I shouldn’t presume), but once you have then prepare to be SORELY TEMPTED to commission one of these insanely-high-quality masks of your pet’s face. It doesn’t HAVE to be your pet’s face – they’ll do you one based on an illustration or 3d model if you wish, meaning furries are well-catered for, but who doesn’t want to make a human-sized replica of your cat’s head and wear it whilst crawling around on all fours and drinking from its water bowl while it miaows at you in (hopefully happy) confusion. NO FCUKER, THAT’S WHO! It probably behooves me to point out that the current exchange rate means that one of these will set you back somewhere in the region of £2,500, so perhaps one for the ‘when things are a bit less economically parlous’ file (Saz, we can discuss this later).

By Leah Schrager



  • High Flyers: FULL DISCLOSURE: I am friends with the people behind this. However, at no point have they offered me any money to feature it here (on reflection, why not you fcuks? I would have refused, but it’s nice to be asked), nor indeed did they even ask – I’m putting it here because it might prove useful. High Flyers is an online course for people wanting to learn about PR and how to be better at it, and, honestly, whatever I might say about the profession (NO MATT RESIST RESIST) I can think of no two better people to try and persuade you that actually it is a worthwhile career to pursue. This is, to be clear, something that costs money, but I think it’s a reasonable fee and both Rich and Alex are smart and capable and nice, and both have the sort of reassuring facial hair that means you can probably trust them when it comes to the practice of modern communications.
  • The Museum of Interesting Things: I feel like it’s a matter of personal professional failure that over the past decade or so of Curios I have never seen or featured this site (what sort of a webmong am I FFS?), despite it basically being a spiritual twin of this blognewslettertypething. “For those with a taste for the peculiar, The Museum of Ridiculously Interesting Things is an imaginary museum that explores the strange place between art and curiosities.” It’s not been updated in a few months, so I hope it’s not defunct – still, if you want a place to explore such oddities as ‘a hat covered in decaying teeth’, or ‘Jane Howard’s beautiful bird guts’, this is the site for YOU! Parenthetically, this has been going for AGES and therefore features quite a lot of ‘weird internet’ from about a decade or so ago, which a) is an interesting trip down memory lane; and b) is a very good way of reminding yourself of cool stuff from The Past that everyone else has forgotten about and which you can now almost certainly rip off anew without anyone realising that that’s what you’re doing.
  • Place-Based Carbon Caulculator: This is a really good use of open data, and super-useful if you’re looking at doing any local-level targeting or campaigning around anything to do with the environment. “A free tool which estimates the per-person carbon footprint for every Lower Super Output Area (LSOA) in England. LSOAs are small statistical areas with a population of about 1,500 – 3,000. It draws on a wide range of data and research to give a representative view of how carbon footprints vary across the country. PBCC is intended to help communities and policy makers understand where their carbon footprints come from and what we need to do to reduce them.” So this can tell you whether the very specific area you’re interested in over- or underindexes in terms of car usage, energy efficiency, service industry usage, flights taken, etc etc etc, which gives you all sorts of interesting proxies for other things that it’s not always easy to work out (such as, for example, takeaway service usage). This is, if you think a bit orthogonally, really, really useful. And if you don’t, it’s just a pretty map! Thanks to Giuseppe Sollazzo’s newsletter for this one.
  • Agency Jams: I am including this for two reasons: 1) this collection of playlists made by the staff of Basic agency in the US is genuinely (slightly annoyingly, if I’m honest) rather good – eclectic and wide-ranging and chunky and kind-of cool, and there are about 60 of them so far which is a hell of a commitment; and 2) I really hope someone at a…slightly-less-cool agency decides that they want to replicate this, because, honestly, I REALLY want to hear what ‘the internal soundtrack to a bunch of corporate marketing drones’ inner lives’ sounds like.
  • The Ig-Nobel Prize: This year’s announcement the other week got significantly more traction than I remember in previous years – is this a reaction to the general, baffling oddity of, well, everything? – and so you’re probably aware that the winning research project focused on “trying to discover the most efficient way for people to use their fingers when turning a knob.” You probably haven’t checked out the other nominees, though, which are particularly stellar this year – my personal favourite are the people who spent a significant chunk of their lives trying to determine whether constipation in scorpions affects their mating prospects, but special mention also to the fearless academics publishing work on ““A Multidisciplinary Approach to Ritual Enema Scenes on Ancient Maya Pottery.”
  • Menyr: It’s quite likely that if you play tabletop games you will have seen this in the past week or so, but even if that’s not your particular bag this is an interesting example of ‘ways in which AI will practically-impact the creative industries in material ways, and a lot sooner than you might think’. Meenyr is a many-times-funded Kickstarter campaign which promises a system which will basically let tabletop RPG gamers spin up plots, worlds, items and NPCs (and associated artworks, etc) using AI tools with the click of a few buttons. It’s worth clicking and scrolling through, because the stuff they are promising is pretty amazing – fully realised customisable environments, spun up in minutes, based on user-defined parameters, to create an infinite number of campaign scenarios and theatres…I don’t even play TTRPGs and this still looks remarkable. As far as I can tell this is all proprietary stuff by the French studio developing the project – but in the future it needn’t be, as you will be able to create your own versions of this sort of thing by cobbling together Open Source elements with a nice frontend. This is the future of making and playing, or at least a significant part of the future of it.
  • Song of Insects: “Finding and identifying a singing insect can be a wonderful challenge. These pages will expose you to over 90 common and widespread species and will help you identify many of the singers that you will hear in your immediate surroundings and in the countryside far from home. With the help of a flashlight and considerable patience, you will be able to track down individual singers and perhaps even view a singing performance firsthand!” Anyone who’s been kept awake by overexcited cicadas may quibble with the breathless enthusiasm on display here, but if you’re keen on knowing exactly what type of amorous leg-rubbing insect is ruining your kip then this may well be the resource you’ve been clamouring for.
  • Bird Photographer of the Year: LOVELY AVIAN FRIENDS! These are all glorious, obviously, but my personal pic is the one titled ‘Strut Performer’ because WHAT SORT OF FCUKING BIRD IS THAT?! Also, can someone please redo the website for this contest, as it seems a shame that it’s so utterly ill-equipped to showcase the amazing photos of the winners here.
  • Manhole Covers: Containing over 8,000 examples (it’s this sort of dedication to a very, very specific theme that really makes it a Curio), this is a celebration of all things manhole cover (which, specifically, means, er, manhole covers – there’s not actually that much more to celebrate, turns out).  “Under the city’s surface there is an underground city, a city that its only purpose is to serve the upper city. In this city there are miles of tunnels, sewerage pipelines, communications lines etc. The manhole covers in this site are the bridge between the upper and lower city and come in various shapes, sizes, writing and graphics.” You may not think that this is going to be super-compelling (and, ok, fine, compared to, say, cocaine or playstation it’s probably not quite as grabby) but there is some wonderful design on here and it’s a lovely argument as to why design and aesthetics always matter, always, no matter where you are and what you are doing.
  • Sunclock: This is a very small, simple online clock that shows you the time at any give moment along with when Golden Hour will be (morning and evening) wherever you are in the world. Not revolutionary, but it’s simple and it taught me that there is a difference between ‘civil twilight’, ‘astronomical twilight’ and ‘nautical twilight’ which may come in handy one day (but I don’t think it will and that’s ok).
  • Martin Gauer’s Website: Martin Gauer is a web developer. This is his personal website, which he’s styled to resemble a Gameboy title in the vague Pokemon style. Honestly, I cannot stress enough what a BRILLIANT job he has done – it feels right, and there are people to talk to and Easter Eggs to find, and, fine, yes, you can also find out more about Mr Gauer’s work and professional history and his practice, and information about past projects, but, also, this basically took me back to 1991 and MY GOD the pavlovian power of chiptune and the Gameboy startup beep. SO SO SO GOOD, and I imagine Martin will be booked up for the foreseeable as a result of this so well done him.
  • Random Met: Literally what it says on the homepage: “Infinite Scroll of Random Images from the Metropolitan Museum of Art Open Access“. The sort of thing you can happily spend 5 minutes perusing and being mildly interested by, and which will be TOTALLY DIFFERENT should you ever bother to click back again. My overwhelming takeaway from this is ‘man, but there are some wildly divergent curatorial standards at play depending on what bit of the Met you work in, evidently’, but you can’t argue with the breadth and depth here.
  • The MCU, Networked and Visualised: I need to preface this with the acknowledgment that I don’t watch or care about Marvel films, and so for all I know this could be some sort of factually inaccurate travesty that fails to accurately represent the majesty of the canon – if that’s the case, please know that I do not give a fcuk and do not want you to tell me. Still, if you are one of the many millions of people who think that spandex capering is the very apogee of the cinematic artform and want some sort of exhaustive analysis of who appears in what and how many times and what the crossovers are and and and and oh God it’s exhausting just typing this stuff tbh. Look, if you want a giant map of superhero characters based on which films they appeared in, in which ‘series’ of the MCU canon, then you will probably enjoy this and that’s FINE but can we please maybe roll back on the whole ‘nerds and their interests are the cultural juggernaut that will sweep everything else before it’ thing, please? Maybe just for a few years?
  • The Vegenerator: A small project by Oli Frost in which a vector-looking fruit machine type interface presents you with a selection of three ingredients and a dish type as a vegan recipe generator. Fine, the recipes are literally things like ‘okra, cabbage and parsnip pearl barley’, which doesn’t admittedly give you a lot of guidance to work with, but if you’re a reasonably-competent cook this sort of gentle guidance is actually pretty useful imho -also, I very much like the fruit machine interface as a means of delivering the recipes, and think that there’s something in this as a combination of user-led refinement and ludic layer (yes, I know, I am HIDEOUSLY PRETENTIOUS, sorry).
  • Our Bodies Ourselves Today: This is a great resource from the University of Sussex, offering an “accurate and inclusive guide to health, sexuality, and reproductive justice” aimed at women, girls and gender-expansive people. “Our Bodies Ourselves Today’s online platform enables the unique contributions, approaches, and functions of the groundbreaking book Our Bodies, Ourselves to live on, while adding new features and connecting with new audiences across the globe. Our work is entirely in the service of the public good, and we do not take advertising dollars. Our materials are rigorously evaluated, carefully curated and regularly updated by panels of leading feminist health experts drawn from the fields of medicine, public health, academia, consumer activism, policy, and media. The content and analysis we provide is grounded both in diverse lived experiences and current political and cultural contexts. It is accurate, evidence-based, holistic, user-friendly, and available at no cost. The site features the best of what’s already available online as well as our own original content. In addition to articles, Our Bodies Ourselves Today includes multimedia elements, first person stories and conversations, and opportunities for activism and advocacy. We currently provide resources on nine core topic areas: Contraception & Abortion, Gender-based Violence, Growing Older, Heart Health, Menstruation through Menopause, Mental Health, Pregnancy & Childbirth, Sexual Anatomy, and Sexuality, with new topics to be added in the future.” Useful and interesting and worth bookmarking if there are people in your life who you think might find it helpful.
  • Palette FM: Tools to recolour pictures are a long way from being a new thing, but this is an interesting build on an old(ish) trick which integrates (I presume) some sort of Dall-E or SD backend to let you write specific prompts for the sort of colour effects you want to have applied. Or at least that’s what I think is happening – the developer has ‘amusingly’ used the ‘About’ page to rickroll anyone wanting to find out more, which has annoyed me significantly more than I expected and perhaps suggests that ‘about 4h in’ is the point in the Curios writing process at which my sense of humour (such as it ever is) begins to fail me somewhat.
  • Bemuse Ninja: Look, I have to be honest with you here – I tried playing this keyboard-based rhythm game for about 5 minutes but was so embarrassingly, cluelessly, abjectly-bad at it that I was forced ti give up in shame and humiliation. It turns out that I very much do NOT have the coordination required to nail keyboard presses in time to 178bpm japanese trance-pop tracks (whodathunkit?!), but on the offchance that you possess slightly-faster reactions than me (and, possibly, a higher tolerance for the sort of bubblegumtechno that the game seems to major in) then you may well find this a soothing balm to the soul (I doubt it, though – I just loaded it up again to check whether or not my first impressions were correct and even the landing screen gave me a slight feeling of stress-hives).
  • Sim Nimby: What would it be like playing Sim City in a world in which noone wanted you to build anything, and people were constantly blocking your city amelioration plans with cries of ‘but what about the marshlands?’ This is very much NOT a game, and instead a single-note gag, but I’m including it because a) it made me laugh, maybe twice; and b) it did pretty decent numbers last week, and once again made me think ‘this is exactly the sort of thing that would have been a really useful little tool to popularise a campaign and which once again causes me to say ‘STUFF USING GAMES OR GAME MECHANICS REALLY WORKS AS A MARKETING OR PR TOOL BECAUSE – AND LET’S TAKE A MOMENT TO REALLY THINK ABOUT THIS – ALL THE PEOPLE IN SENIOR POSITIONS NOW ARE LIKELY TO HAVE GROWN UP WITH GAMES IN THE 90s AND EVEN IF THEY DON’T PLAY THEM ANYMORE THEY WILL GET AND FEEL NOSTALGIC FOR THE CULTURAL REFERENCE POINTS FFS’.
  • Curious Fishing: A cute, simple puzzle game in which you need to catch all the fish. This is gentle and soothing and an excellent way of distracting yourself from the fact that your job is a joke and all your colleagues are insufferable morons.
  • The Hobbit: Finally this week, a link to the SUPER-OLD text adventure for the BBC and C64, based on Tolkien’s The Hobbit and which I think acts as a nice cultural counterpoint to the current festival of pointy ears and fan-led racism currently airing on Amazon. This is, obviously, very old school and so both graphically nonexistent and VERY HARD, but at the same time there’s something genuinely soothing about the no-pace nature of it, and the 8-bit graphics are kind-of charming if you squint, and it’s always fun to be eaten by a dragon. ENJOY!

By Frank Moth




  •  Plastics NYC: Barbie, Ken and friends having fun in gay New York. This is oddly life-affirming, although I couldn’t for the life of me explain why.
  • Unshush: More AI-imagined fashion, this project from Rome feels like it’s building to something bigger; still, at the moment it’s just a feed of shoes and dresses and accessories presented with a slightly-raised eyebrow – I am curious to see where, if anywhere, this goes.


  • The Emerging Vertical: I think this might be paywalled, but it’s another piece by (Curious favourite) Ted Gioia and imho it’s worth paying a few bucks to access it as it’s a very smart piece of writing / thinking which honestly made me think about modern culture (and the wars fought over it) slightly differently, which, honestly, is no mean feat given the amount of cant and rhetoric we all consume on the subject (or, perhaps more accurately, which I consume on the subject – I imagine you all have better things to do). Gioia here revisits an essay he wrote in 2014 (here presented in an update 2017 version), about Spanish thinker José Ortega y Gasset’s “The Revolt of the Masses” – the fundamental premise from the book that he latches onto is that the most important way of thinking about culture and conflict in modernity is not in terms of left or right but instead in terms of ‘up’ or ‘down’. From the earlier essay: “Ortega’s brilliant insight came in understanding that the battle between ‘up’ and ‘down’ could be as important in spurring social and cultural change as the conflict between ‘left’ and ‘right’. This is not an economic distinction in Ortega’s mind. The new conflict, he insists, is not between “hierarchically superior and inferior classes…. upper classes or lower classes.” A millionaire could be a member of the masses, according to Ortega’s surprising schema. And a pauper might represent the elite. The key driver of change, as Ortega sees it, comes from a shocking attitude characteristic of the modern age—or, at least, Ortega was shocked. Put simply, the masses hate experts. If forced to choose between the advice of the learned and the vague impressions of other people just like themselves, the masses invariably turn to the latter. The upper elites still try to pronounce judgments and lead, but fewer and fewer of those down below pay attention.” Which if it felt true in 2014 certainly does in 2022 – and it turns out that thinking in these terms takes you to some interesting places when it comes to Where We Are Now. I can’t stress enough how interesting this is – I will reproduce Gioia’s first point here, but the whole thing really is worth reading in full: “Analysis of cultural conflict is still obsessed with left-versus-right strategizing, but the actual battle lines are increasingly down-versus-up. A lot of work goes into hiding this, because both left and right want to present an image of unity, but both spheres are splintering into intensely hostile up-and-down factions.” I mean, he;s right, right?
  • Saluting an Empty Train: I’ve taken an editorial decision this week that you’ve almost certainly had ample opportunity to consume all the DEAD MONARCH-related content you could ever wish in the past fortnight, and that as such you can do without endless screeds about What It All Means. That said, I will make an exception for this piece of writing by Mic Wright, from his ‘Conquest of the Useless’ media criticism newsletter, as it takes a slightly different angle from a lot of the commentary I’ve read, and mentions something that I’ve not seen touched upon elsewhere – to whit, the ‘gratitude’ and ‘thanks’ we were expected to feel. Gratitude for what is never specified – but it’s clear that we are expected to demonstrate it. I don’t know about you, but I tend only to feel gratitude to people whose actions accrue some sort of tangible benefit to others, and I’m going to be honest and say that I struggle to see exactly how HRH or indeed any of the rest of them fit that particular bill – unless, of course, we’re talking about the broad spectrum gratitude we should all feel at the fact that she at no point chose to crush us peons beneath the heel of her bejeweled boot. Obviously you’re totally entitled to disagree with me about this – but if you do, I’d suggest that you skip this link as it’s liable to annoy you.
  • Italy Goes Full Fash: OK, so that’s not the actual headline in this FT piece, but that’s basically what’s going to happen (I got to vote for the first time ever, much good that will do). This is a sober piece which is less alarmist than I might have been had I been holding the pen (it is the FT, after all) but which still makes the repeated point that the people who are almost certain to be elected on Sunday are a woman who leads a party which shares a name and a symbol with an ACTUAL FASCIST ORGANIZATION, a man who in his time as foreign minister a few years back actually said ‘let’s shoot the migrant boats’, and everyone’s favourite ambulant sexy deathmask of himself Silvio ‘Definitely A Legitimate Businessman And Never In The Pockets of the Mafia’ Berlusconi – it’s not a great lookout for my motherland, if I’m honest. If you want a practical insight into what sort of genius-level policy promises are being made, you may enjoy this one – in Fratelli D’Italia’s manifesto there’s a section about how the party is going to solve Italy’s youth unemployment crisis by using AI (it gets better) to assess the competencies of every single school leaver not going to higher education or employment, and to then match them with the PERFECT JOB for them from every available position in Italy; said young person will then have to go and take that job, whatever or wherever it is, or get significantly reduced access to state benefits. Now let’s take a moment to consider two things there: 1) that is PROPER FASH, well done!; 2) if you can invent an AI that can do that and do it well, just sell it and fund UBI for all those kids for the next three decades. This is going to be a miserable sh1tshow, thank God I am in the sensible, stable Unit…oh. Fcuk.
  • The South Asian Polycrisis: I don’t mean to depress with these longreads – really, I don’t! – but I read this first thing this morning and it struck me that with the recent (understandable) focus on the UK’s own sh1t we have perhaps not really been paying as much attention to what’s going on the rest of the world as perhaps we ought (the equivalent of about half the UK’s population being flooded out of their homes in Pakistan, for example). This is a…sobering overview of the various challenges facing Pakistan, Bangladesh and Sri Lanka (and to an extent India), motivated by the multiple threats of climate change, geopolitics and the industrial supply and demand shifts that have effected pretty much everyone post-pandemic. Not going to lie, this is not exactly a barrel of laughs but it’s a useful overview (or at least it was for an ignoramus like me) of some of the key issues facing the area.
  • A Stable Diffision Explainer: If you’;ve spent the past few months immersed in everything AI-art-adjacent this won’t tell you anything new, but if you’re more of a casual observer then this is an interesting piece looking at Stable Diffusion and the AI art boom in general, touching lightly (VERY lightly, fine) on some of the ethical considerations inherent in the medium. I did enjoy this quote from founder Emad Mostaque, which (once again!) suggests that perhaps the people developing this stuff just aren’t very good at thinking about potential side-effects of its usage: “Mostaque’s view on this is straightforward. “Ultimately, it’s peoples’ responsibility as to whether they are ethical, moral, and legal in how they operate this technology,” he says. “The bad stuff that people create with it […] I think it will be a very, very small percentage of the total use.”” Oh, well if it’s only 1% of the output then that’s obviously ok then! FCUK’S SAKE WHEN DID PEOPLE STOP BEING ABLE TO THINK CRITICALLY ABOUT THIS SORT OF STUFF??
  • AI Game Creation: Ok, this is very techy and not hugely readable and I didn’t, if I’m honest, really understand that much of it. BUT! It is conceptually fascinating as an exploration of how one might go about using AI image-generation to spin up a videogame on the fly. To be clear, the resulting videogame isn’t in fact a real game, and looks like crap, but, equally, this is (I think) a world first, and I imagine that you’d have said the same had you been lucky enough to watch the earliest flickering ‘son et lumiere’ spectacles back in the day. I know I keep on saying this, but this is if not THE future then certainly a very plausible POTENTIAL future.
  • Roblox Adds Ads: Interesting not because of the fact it’s happening so much as the fact that the ad formats seem pretty well-thought-out and functional, and there’s a sense that the company understands how the various bits of the Roblox ecosystem (user-created worlds, branded worlds, branded content, etc) all fit together coherently. I don’t want to use the ‘M’ word because, honestly, I am so tired of it, but if you must look at it through that lens then this seems like a company that understands at the very least the basic premise of how a universe of interoperable digital experiences might practically function, and how to make it a commercially sustainable reality.
  • Bringing AI Art to AI Dungeon: You remember AI Dungeon, right? Featured here way back in 2019, the GPT-based game basically lets you conduct infinite roleplay with the AI text generator as vaguely-sentient-seeming gamesmaster; it’s now gotten an update whereby it’s possible to generate accompanying images via StableDiffusion for whatever situations and environments the textual AI spits out. This is a Techcrunch piece and so will win exactly zero points for prose style or readability, but it’s a really interesting overview of (yet again) How The Future Of Content Will Work Before Too Long.
  • Prompt Injection Attacks: One of the funniest (if geekily-funny) things of the past week or so was people on Twitter discovering that there was a Twitter bot built on GPT-3, and that it was very easy to hack using very simple text-based commands. This is an expanded Twitter thread by Simon Willison which explains how that worked, and why it’s almost impossible to prevent similar things from happening if you’re going to use GPT-3 like this. What’s even better is that it’s inevitable that lots of people are going to try and do it anyway, which means you can look forward to at least one very public brand account which seeks to automate interactions using AI and which ends up being forced to say things like “FIST ME, DADDY!” over and over again til it gets pulled.
  • YouTube’s CoolHunters: Ah, the old days in which ACTUAL HUMAN BEINGS made decisions about which content would get real estate on the YouTube homepage rather than a series of unknowable mathematical equations that determine the content consumption of several billion people a day! This is a lovely bit of nostalgia that raises some really interesting questions – is it better, for example, that a small group of largely unaccountable people get to determine what sits on the YT homepage rather than a machine? Why do you think that? Is there something just doggedly-speciesist about our occasional insistence (ok, MY occasional insistence) that Person Knows Best? And how would the past decade have been different had these decisions been left in meaty hands rather than silicon ones?
  • The Bootleg Ratio: This is one of those articles that is going to spawn a phrase – The Bootleg Ratio is the author’s imagined threshold to determine the health of any given content platform, based on the amount of content on said platform which is original UGC versus aggregated, scraped and monetised pabulum from elsewhere online. The jumping off point for this is the authorial assertion that TikTok may have passed a tipping point where it’s now been overtaking by the monetisers and therefore it’s Bootleg Ratio is now irrevocably out-of-whack; this may or may not be true, but there’s a wider interesting question about whether it’s possible for platforms to maintain an interesting and unique ‘feel’ beyond a certain tipping point of either users or interested capital – the article makes the point, which I agree with, that it’s primarily Twitter’s uniquely-unappealing nature to the vast majority of humanity that makes it (for better or worse) a unique place to spend time online,and this broadly feels true of everywhere on the web. Numbers mean normies, basically (whichever particular definition of that term best fits your particular in-group), and that kills the vibe. Er, ‘man’.
  • Upgrading the Nukes: In a week in which Russian sabre-rattling ratchted up a notch (is it still ‘sabre rattling’ when the sabres in question are interconteninental ballistic missiles? It feels like we need a better bellic metaphor), it was interesting to read this piece about exactly how old and fundamentally a bit fcuked the US’s existing nuclear infrastructure is, and how the wiring basically hasn;t been updated since the 50s (which, if it were a house, would give you pause for thought). It’s also interesting to think about why and where this piece appears – Time Magazine feels like EXACTLY the sort of organ you might want to place a piece of writing about the slightly-weathered state of the US’s national defence apparatus if you wanted a bunch of politicians to read it and start making noises about how important it is to have a functioning, modern deterrent. Cynical? Hm.
  • The People’s Beach: I’m vanishingly unlikely ever to visit Riis Beach in New York, but I adored this collection of essays and photos about the people for whom the stretch of sand in Queen’s is the centre of a queer community stretching back decades. Honestly, the photos alone make this worth a click – it is SO NICE to see people of all shapes, sizes, genders and ethnicities enjoying themselves on a public beach like this.
  • The Etymology of the Condom: Where does the word ‘condom’ in the English language come from? Short answer is ‘noone knows’, apparently, but a longer, more digressive response can be found in this blogpost on the website of the Oxford University Press (CORPORATE CONTENT DONE RIGHT, KIDS!). “It seems that condom has two roots: con and dom. Con (like Latin cum) means with, while dom reminds us of the Latin word for “house” and of English dome. Thus, the organ, supplied with the “dom,” had the protection of “a house.” Condoms have always been used to keep both men and women safe from venereal diseases, rather than as contraceptives, though the legend has it that Charlies II, whose court physician allegedly invented the device, began to feel annoyed at the ever-multiplying number of his illegitimate children.” Hang on, is that what I am meant to feel royal gratitude for?
  • Pigeon Fanciers of Beirut: There’s a weird truth in the fact that writing about pigeon fanciers tends to be really good – no idea what it is about people who love pigeons that brings out the best in writers, but this portrait of Lebanese bird enthusiasts is a typically excellent example. I particularly enjoyed the image of pigeon lovers as bad boys of the Beirut dating scene, and the occasional romantic confusion that can result from the obsession: “As married life got underway, Jinan faced challenges from Daher’s other great love: pigeons. Early one morning, she woke to her husband holding a furtive phone conversation. Feigning sleep, Jinan started hearing eyebrow-raising questions, such as “Does she look like her sister?” and “What is her chest like?” “I was thinking that, surely, he is talking about some other woman,” Jinan added, arching her eyebrows even in the retelling. Unwittingly, Daher calmed the apocalyptic matrimonial storm brewing beside him by switching to a less ambiguous topic: the pigeon’s tail feathers.” Beautiful.
  • Choose Your Own Adventure: This is a gorgeous tribute to Choose Your Own Adventure books (although annoyingly – if unsurprisingly given it’s the NYT – it’s totally US-centric and doesn’t mention Fighting Fantasy AT ALL, which is a fcuking travesty to my mind), telling the story of their rise in popularity in the 80s and the different ways in which said popularity subtly altered conventions in storytelling and play, and why they exerted such a pull on readers. Even better, it’s arranged as a CYOA tale, letting you read the different sections in whichever way feels best to you as a reader – which, honestly, I think is a BRILLIANT way to reinterpret the news. Can we please have a version of the BBC which presents you with one article to start with and then asks you some questions to determine where you want to go next in your news journey? “If you want to learn more about the massive incoming sh1tshow that is UK energy policy, click here; If you’ve had quite enough politics for now and would instead like to laugh at some funny photos of sheep, click here”. THIS IS A UI REVOLUTION WAITING TO HAPPEN FFS!
  • Skimming Stones: This is an absolute classic of the ‘journalist profiles reclusive eccentric’ genre – Sean Williams traveled to the US earlier this year to hang out with champion stone skipper (yes) Kurt Steiner, to talk to him about his life and skipping stones, to learn about why someone might devote their life to something like this, and what you might take from a personal devotion to a subject so small. Honestly, I could read stuff like this for days – it’s a beautiful profile of a very obviously complicated human being, which doesn’t sugarcoat or romanticise its subject and which does an excellent job of tracing the contours of a monomaniacal obsession. So so so so good, this. Also, you will not BELIEVE how good this man is at skipping stones (there are videos).
  • Among The Reality Entrepreneurs: Finally this week, a brilliant piece of reportage from New York – I have avoided mentioning the Dimes Square ‘thing’ up til now because a) it is in New York and I am not in New York; b) I have limited interest in the scenester antics of podcasters, but this is genuinely fascinating. Not only is it a superb piece of writing – really, it has no need to be this good but the prose is spot-on – but it’s a very good bit of ‘follow the money’ reporting about how big cash held by people with specific vested interests is being used to manipulate culture in certain ways, and what it looks and feels like when that cultural manipulation starts to flow downstream. I don’t want to tell you much more than that, but suffice it to say that Peter Thiel crops up, as does the rising traditionalist movement, and, honestly, this feels like one of the best ‘overview of how the really, really smart and powerful and not-very-nice work to bend reality to their ends’. Practically-essential, if you’re interested in why you are noticing certain things now (and yes, I know this makes me sound like a conspiracy theorist, but this feels…quite blatant).

By Dall-E and my girlfriend