Webcurios 09/09/22

Reading Time: 35 minutes

You’re probably a bit distracted by the news, aren’t you?

It’s fair enough to be honest – it’s not every day a living embodiment of national identity shuffles off this mortal coil – and you’d be forgiven for not giving two skinny fcuks about ‘a bunch of stuff some bloke you don’t know thought worth sharing from the past fortnight’s web’, but, on the offchance that there are a few of you who want a distraction from the mourning, or from the first draft of your epic commemorative prose poem about Her Great Legacy (I really, really hope the papers offer up some space for readers’ tributes – nothing marks the passing of a head of state quite like the poorly-constructed iambic stylings of the barely-literate, after all!), or from scrolling endlessly past the shouting and sadness and anger and point-scoring and memes and really-quite-top-tier insanity that is Twitter right now, or from the nagging fear that none of the other worrying news has gone away, then, well, I AM HERE FOR YOU!

Also, I am travelling again next week so Curios will be back on 23rd September (but then it will be weekly til the end of the year, promise (well, promise-ish)).

I am still Matt, this is still Web Curios, and I can categorically assure you that this is what She would have wanted.

By Anna Koak

WE KICK OFF THE MUSIC THIS WEEK WITH THIS SUPERB CARNIVAL MIX FROM LAST WEEKEND AS AN ANTIDOTE TO THE SOMEWHAT SOMBRE CURRENT NATIONAL SOUNDTRACK! 

THE SECTION WHICH NEVER MET THE QUEEN DESPITE HAVING ONCE WORKED AT BUCKINGHAM PALACE AND AS SUCH IS UNABLE TO CONTRIBUTE ANY PERSONAL ANECDOTES ABOUT HOW SHE WAS VERY DOWN TO EARTH AND INCREDIBLY FUNNY, PT.1:  

  • P0rnpen: I know I normally chuck everything bongo-related at the end of Curios so as to make it easy to ignore, but this particular link presents something of a special case; it falls under the heading of ‘general AI image magic’, and as such thematically needs to sit with the rest of that stuff, and, more interestingly, IS IT BONGO IF NO ACTUAL HUMAN BEINGS HAVE ARE DEPICTED? Still, tedious questions of taxonomy about which noone actually cares aside, be aware that you are two clicks away from ACTUAL NAKED PEOPLE – or, at least, collections of machine-imagined pixels that look very much like actual naked people. P0rnpen (sorry for the bowdlerisation but, well, firewalls) is the first full Stable Diffusion-based standalone ‘imagine me a naked person’ website I think I have seen, and…well, it’s interesting in all sorts of ways (I know that this comes across very much as ‘I read Playboy for the fascinating and eclectic journalism!’, but it’s true!). The landing page lets you select from a variety of parameters to determine what sort of imaginary naked person you’ll generate, and if you click on the ‘Feed’ tab you can see a (realtime?) view of everything that the machine is spitting out – unsurprisingly-but-depressingly whoever has built this has only bothered to make it capable of generating female-presenting outputs, so effectively this is just a cascading wall of computer generated Page 3 pinups, in the main. It’s not exactly surprising that the outputs wouldn’t be more heterogenous – the web is still the web, after all, and in any case this sort of software tends to find penises…problematic to render – but it’s still a bit miserable. More amusing is quite how weird some of the outputs are, which leads to many of the images being a bit more ‘HOW MANY NIPPLES?!’ than your average scudpics. Most interesting of all, though (or at least to my mind), are the various ethical questions that this raises – given none of the people in these images are real or have ever existed, should there be any constraints on what they are depicted as doing? What is this sort of thing doing to challenge / reinforce sexual/social stereotypes, and what sort of preventions / protections need to be built in to guard against said stereotypes? It’s notable that there’s no ‘free text’ image generation here, as whoever’s running it has quite rightly surmised that it would quickly become an absolute horrorshow – but I’d be amazed if there aren’t a bunch of things like this with fewer guardrails running on private servers somewhere. WHERE DOES THIS ALL GO? Anyway, sorry for kicking this all off with bongo – IN THIS OF ALL WEEKS FFS MATT WHERE IS YOUR RESPECT AND DECORUM?? – but hopefully you agree that this is at the very least conceptually-fascinating.
  • Lexica: Remember a few weeks back when everyone was thinking ‘oh, it’ll be fine, I can just retrain as a ‘prompt engineer’ and that will be enough to keep me from the cold streets of London for a few months longer’? Yeah, probably not, sorry. Lexica is a search engine for AI image generation prompts – tell it what you want to generate and it will spit out a selection of Stable Diffusion-generated images that roughly match your search criteria, letting you see not only the pictures but the search terms used to generate them, making it trivially easy to zero in on commands that have the effect or style that you’re after. Obviously you’ll still need a degree of ‘skill’ (or, in my experience, just a lot of patience) to create something ‘good’, but as a way of shortcircuiting the ‘creative’ (fcuk, we’re going to need new vocabulary, I am already bored of putting inverted commas around stuff like this all the time) process this is potentially super-useful (and also as a way of exploring some of the things that people have been generating). Amusingly there was something of a furore in the prompt engineering community this week when this other site appeared (it’s currently a bit flaky re overuse, as far as I can tell) which suggests prompts to you based on your inputs, leading all sorts of people to start getting upset at the extent this was ‘stealing the creativity of the engineers’ to which LOLOLOLOL. Anyway, all this is to suggest that a) if you want to use this stuff properly then there are already a bunch of really helpful tools to help you do so; and b) I don’t think ‘prompt engineer’ is really going to be an actual long-term career option, sorry.
  • The Stable Diffusion Training Dataset: I appreciate that this isn’t the most enticing title, but this is a really interesting bit of work by (friend of Curios) Andy Baio and Simon Willison in which they basically pulled a bunch of the training images used to make StableDiffusion and made them searchable, so you can start to get an idea for the sorts of material that has been used to develop the AI. Andy’s written up a nicely-informative post about what you’re looking at and what it tells you, which you can read here if you’re so inclined, and whilst it is only a very, very partial selection of the training data it gives an interesting look into why and how the prevailing aesthetics of this particular model came about. You can search different parameters – including, yes, how NSFW the image is classified to be, which offers some interesting insights into the degree of…prudery? Sexual…’sensibility’? baked into the machine -but perhaps most interesting is the list of artists that have been ‘scraped’ and incorporated into the model – and personally speaking this sent me down a proper rabbithole of wondering about who will decide which material we use to train the next models, and how those decisions will be made, and why.
  • Summer Island: One of the most interesting things about the current wave of imageAI stuff is seeing how people use it – this particular example is perhaps predictable (especially given the styles the machines seem most-comfortable generating), but no less surface-level impressive for that. Summer Island is a comic, with all the artwork generated by (I think, based on the style) Midjourney, and at first glance this is honestly amazing. As with anything like this I am going to assume that there’s been a significant degree of tidying and touching up behind the scenes, but if you step back and squint this looks like professional illustration work and the composition of the panels is…not bad! Per Clueless, the thing is in fact a bit of a Monet and doesn’t really stand up to close scrutiny (as with all AI stuff, crowd scenes get bizarrely horrific if you try and focus on anyone’s actual face, and I’m sure that professional comics artists could explain to me why the composition isn’t quite right) – but as with all this stuff it’s worth remembering that it would have been literally impossible six months ago, and the direction of travel in terms of the quality of outputs is only one-way.
  • Stories by AI: The Holy Grail of this sort of AI creativity, of course (or at least a Holy Grail), is the ability to magically pipeline words to images with nary a human involved – we’re quite a way from that, but it’s nice to see people experimenting with the intersection of AI words and pictures. Stories by AI is a newsletter, each edition of which presents a short story cowritten by man and machine and then illustrated by AI (with a little help from some humans doing the typing and the imageselection) – these are momentarily-interesting, although they do all suffer from the slightly-self-consciously-wacky ‘look at what the crazy machine imagined next! SO RANDOM!’ style of prose and narrative that all these projects tend to veer towards. This currently exists mainly to prove to anyone feeling scared that there’s still a market for human-penned narratives – although this may be less true for those ploughing the ‘self-published twisty-turny domestic thriller’ genre on Amazon.
  • SALT: This is a really interesting little project, currently living on Twitter. SALT self-describes as “the world’s first fully AI-generated multiplot “film” – a web6 internet adventure where your choices create a 1970s lo-fi sci-fi universe”, which, honestly, means fcuk all to me but which I will try and parse for your (and, frankly, my) benefit. As far as I can make out, SALT is attempting to tell a slightly-creepy horrorish old-school style scifi story, with readers able to vote on the next plot developments which they will then see rendered entirely using AI tools on their Twitter feed. So for example you can vote on whether a scientist should keep a sample of the STRANGE AND MYSTERIOUS ALIEN MATERIAL or jettison it into space, and the winning choice will become a part of the narrative and form the basis for the next short AI-made film they post. There’s a lot of work going into this, with loads of different bits of tech being used to compile the videos, and whilst I’m personally not convinced that Twitter works as a vehicle for this sort of thing it’s fascinating to see how it develops. If nothing else, it turns out that the machines are REALLY good at rendering ‘stuff that looks a bit like Blake’s 7 or other stuff from that era when spaceship interiors were very beige’.
  • Yige: One of the side-effects of the fact that the Chinese (and indeed wider-Asian) app ecosystem is so impenetrable and inaccessible to people outside the region (whether for legal or linguistic reasons) is that we have at best a hugely-partial view of What Is Going On Over There in terms of development. Which, perhaps, is why I haven’t seen anyone really writing much about the Chinese equivalents of Dall-E, Midjoiurney and the rest – one of which is Yige, developed by Baidu. As far as I can tell – which, admittedly, really isn’t very far what with my complete inability to read Chinese – this is only accessible if you have a Baidu account you can link it to, but you can scroll down the page here to see example outputs generated by the software. What’s interesting about this is the degree to which there are obvious recognisable stylistic similarities between the stuff you see here and the software coming out of the West, but also how there are very clear visual differences borne out of differently-weighted datasets, etc, and the extent to which this helps one quickly understand the fact that this stuff is hugely political and the training inputs really matter and decisions about what things are trained on are not in any way trivial and visual filter bubbles are probably going to become a sociocultural thing, aren’t they? I am slightly dizzied by the futurespeculation that this stuff invites, honestly.
  • Animating Stable Diffusion: A proof-of-concept webtoy that lets you input two different Stable Diffusion prompts and create an animated transition between the two, just to demonstrate that you can. You can see examples here – whilst this is relatively-rudimentary, the ‘Tom Cruise flashing you a smile’ one is a good example of the sort of potential this stuff has.
  • AI Music Videos: A whole YouTube channel presenting videos created with AI imagemakers – a sort of line-by-line visual interpretation of the lyrics, presented as stills. So not videos at all, but a nice example of how this stuff can easily generated (a certain type of) visual and stylistic atmosphere. If you want more/better, though, this effort by Ben Gillin is STELLAR, not just because it features an all-time classi track but because of the way in which the visual style of the imagery (and light post-production effects I think) works with the music (it’s worth checking out Gillin’s channel as he’s also done a few of these and they are all excellent).
  • Making Minecraft Realistic In Realtime: Oh, ok, not quite photorealistic – unless your photos are somewhat-impressionistic as though viewed through very thick glasses, in the rain,whilst incredibly drunk – but as a very early demo this is astonishing. Watch the video and marvel at how a prototype plugin of Stable Diffusion tries to turn a Minecraft world into something real-looking as someone plays, and take a moment to think about what this stuff will be able to do in a couple of years – and then about the potential avenues this opens up, where you can (for example) create a rough map in Excel and then output it into a 3d modeled render in infinite range of aesthetic styles at a click. SO FUTURE!
  • Photoshop With AI Superpowers: This is another little demo video, this time showing how a prototypical integration between Stable Diffusion and Photoshop could work – this is, honestly, the ur-example of human/machine visual creativity, and the sort of thing that I imagine all the people promising a better, easier, more creative AI future have in mind when they try and sell you on the vision; I look at this and I momentarily imagine that it would enable even me to create something beautiful (it wouldn’t, obviously, there are some things that are beyond the ability of even the machines) – even if you’re less madly-hubristic than me, the potential here in terms of massively-reduced timescales to create large-scale coherent visual work is self-evident.
  • AI Birds: This Twitter account posts images of birds imagined by machines, in a useful precursor to a near-future in which we’ve managed to accidentally extinguish all avian life on Earth and are reduced to getting the computers to imagine bluetits as we chow down on gruel in the Elizabeth II Memorial Bunker.
  • All The Useful AI Image Links: I know, you’re thinking ‘this can’t be all the links, Matt, that’s impossible!’ and yet it certainly feels like all the links – if you’re playing around with this stuff, whether for fun or professionally, you need to bookmark this link, which presents a frankly insane selection of “awesome tools, ideas, prompt engineering tools, colabs, models, and helpers for the prompt designer playing with aiArt and image synthesis. Covers Dalle2, MidJourney, StableDiffusion, and open source tools.” Seriously, if you’re a designer or illustrator and you want to start trying to use this stuff properly, this is the motherlode.
  • Musika: Just in case you thought it was only the artists having their lunch money nicked and sand kicked in their faces by the terrifying AI future, musicians are also set to suffer – Musika is a toy hosted on Huggingface which lets you quickly spin up short clips of entirely-machine-imagined music, in either ‘classical’ or ‘experimental/techno’ style, and whilst you wouldn’t listen attentively to any of the sounds produced, you might reasonably conclude that there’s literally no point paying the subscription fee for the stock music service you’ve been using to provide tooth-grindingly upbeat soundtracks to whatever blandly-positive pitch pabulum you’ve been forced to churn out. Admittedly the ‘techno’ stuff tends a bit hard towards the ‘experimental’ end of the spectrum – a lot of this sounds weirdly like Gearwhore, which…surprised me a bit tbh – but the classical stuff could TOTALLY be used as a bed for your ‘this startup will revolutionise your B2B marketing journey!’ pitch video. So, er, that’s nice! I wonder how soon we’ll start to see musicians who used to make a living writing for stock sites starting to offer knockdown bespoke compositional services instead?
  • The Meatverse: Just in case you’d forgotten about the (nonexistent) ‘metaverse’, this site will hopefully remind you. It’s a single-note gag, but I very much appreciated all the various multiversal Zuckerbergs festooned across the page, along with the pleasingly-90s webdesign.
  • MetaTommy: I remain unconvinced that there is any current mass-market demand for ‘virtual designer clothes that I can dress my avatar in’ – and yes, I know that SOME brands are making money selling this stuff, but that is not the same thing and that is not all that they are selling – but, credit where it’s due, this site by Tommy Hilfiger made me not hate the concept, which is no small feat. As with most of these things, it integrates with Ready Player Me (who seem to be winning the ‘one avatar to rule them all’ early race, though whether this means anything long-term is very much tbc) but also VRChat and other platforms, and there’s a neat link between dressing up your avatar and being gently punted the real-world equivalent for a meatspace purchase. To be clear – I am very much convinced that this is an expensive waste of time and the sort of thing that only brands with bottomless pits of ‘fcuk around and find out’ marketing budget can really permit themselves to play with; that said, this made me less angry and upset than I expected it to, so, er, WELL DONE TOMMY!
  • Nevermet: How many ‘metaversal dating apps’ do you think there are? Based on that guess, how much of a flex do you think describing yourself as ‘the #1 VR dating app for the metaverse’ in fact is, based on that estimation? Regardless, THIS IS IT! Nevermet is a dating app whose gimmick is that rather than presenting images of your physical self you instead present as your Avatar from whichever virtual platform you prefer to spend time in – so basically ‘Tinder, but rather than meeting up for a desultory coffee before realising that you are fundamentally pheremonally-incompatible you instead dress up as a wolf-person and go and harrass strangers in VChat’ (or at least that’s how I’m  interpreting it). “Our mission is to create a new digital relationship culture, where people can express themselves freely and where meaningful relationships flourish in the metaverse. We envision “limitless relationships.”” – I have no idea what ‘limitless relationships’ means (is it nothing? I think it might be, you know!), but if you like the idea of swiping through an endless parade of surprisingly-buff anthropomorphic mammal-people (I’m sure it’s not all Furries but, well, I think there are going to be a lot of Furries) then DOWNLOAD IT NOW! PS – this is an interesting look at the whole ‘metaversal dating scene’, should you be so inclined.

By Helen Beard

NEXT UP IN THE MIXES, THE LATEST ECLECTIC INTERNATIONAL SELECTION BY VINYL-ENTHUSIAST TOM SPOONER! 

THE SECTION WHICH NEVER MET THE QUEEN DESPITE HAVING ONCE WORKED AT BUCKINGHAM PALACE AND AS SUCH IS UNABLE TO CONTRIBUTE ANY PERSONAL ANECDOTES ABOUT HOW SHE WAS VERY DOWN TO EARTH AND INCREDIBLY FUNNY, PT.2:    

  • Wreck The Brief: Do you work in advermarketingpr? Have you ever done deliberately bad work on a pitch because you didn’t want to win the business? Have you ever dragged your heels in the workplace as an act of deliberate sabotage against THE FORCES OF (WHAT YOU PERCEIVE TO BE) EVIL? If not, why not? COME ON FFS, SABOTAGE IS FUN! I obviously have never done this – when I do bad work it’s because I am simply not very good at my job! – but should you be the sort of person who believes in TAKING A STAND then you might want to get involved with Wreck The Brief, a creative project by Glimpse whose mission is as follows:  “Right now, too many of our finest young creatives, artists, storytellers and communicators are being put to work selling unsustainable, high carbon products and lifestyles that are driving us to destruction. This presents both a huge challenge, but maybe an even bigger opportunity.” The Brief Sabotage Handbook is available on application – they will post you up to 10 copies – and is designed to help you with “delaying, distracting and ultimately wrecking creative briefs from fossil fuel clients such as Shell and BP. The handbook offers playful tips to derail every step in the creative process from strategy to production, from ‘politely interrogate the truth of the planner’s consumer insight with an anecdote about your mum’ to ‘blow the production budget on poke bowls’.” Web Curios obviously isn’t suggesting that you deliberately sabotage your employers’ work for companies that might be considered morally ambiguous at best, but, to be clear, Web Curios is also not not suggesting that.
  • David Bowie’s Dressing Room: David Bowie was without a doubt one of the most interesting creative minds of 20th Century popular culture. So who better to celebrate that than, er, famously-fun purveyors of overpriced imagemeddling software Adobe?! Yes, that’s right, everyone’s favourite guardians of their own brand (NO I WILL NOT CAPITALISE THE ‘P’ IN pHOTOSHOP YOU FCUKS WHAT ARE YOU GOING TO DO ABOUT IT?) have for some reason seen fit to create this small digital experience in which YOU can experience the glorious hallowed ground of Bowie’s dressing room, replicated in your browser. The setup here is a bit creepy if I’m honest – the implication from the scene-setting is that you’re basically sneaking into Bowie’s private space while he’s not there to nose around his personal effects, sniff his smalls and read his diaries, which feels…not very cool tbh – but if you’re a fan then you might enjoy the ability to click around and find Easter Eggs about all his ICONIC STUFF. Although on reflection, if you’re a fan you’re probably already well aware of the significance of the makeup and the boots and all that jazz (this is, by the way, very much 1970 Bowie), so I’m fcuked if I can tell who this is aimed at. Still, it’s OFFICIALLY LICENSED by the Bowie estate, and it’s quite a shiny bit of webwork, so perhaps you’ll care for this more than I did.
  • Crisp Sandwich Day: Matt Round has for a while now been running Buttystock, the world’s premier (only?) stock photography resource for crisp sandwiches. Now he’s decided to GO BIG by declaring October 25th International Crisp Sandwich Day and inviting everyone in the world to get involved. I am including this link because a) it’s silly; and b) I figure enough of you work for places who might be persuaded to get involved with this in some small way. Although I appreciate that if you’re reading this on 9 September this might not be the best morning to try and get everyone excited about a frivolous brand activation involving crisp sandwiches – one to revisit after the funeral, maybe.
  • The Squeaky Wheel: What would The Onion be like if it were written about, and by, disabled people and their life experience? That’s basically the elevator pitch for Squeaky Wheel, a new website which “is the first-ever satire publication that focuses on the experience of having a disability. It challenges common misconceptions, highlights absurdity, criticizes imbalances, and does it all with humor.” This is North American (as far as I can tell from the style/spelling) and I can’t promise that the humour will be to everyone’s taste, but I properly laughed-out-loud at the headline “Boyfriend Finally Pops the Question: ‘So, What Happened to You?’” should you want an indication of the general tone on display here.
  • Sex Positive Social Media: This is an interesting movement / manifesto: “Social media is taking on an increasingly central role in shaping and constraining cultural life, popular discourse, and human sociality. Sex is an important part of this. Yet, social media policies are not very sex-positive. Through their community standards documents and content moderation practices, platforms currently make private, arbitrary and unaccountable decisions about the kinds of sex and sexualities that are visible in online space. We want that to change. Social media rules around what can and can’t be posted shape broader attitudes towards sex and nudity, which in turn directly impact on all of our safety and wellbeing. We believe that we’re healthiest and happiest when sex is not a source of shame but accepted as part of human experience…This Manifesto sets out guiding principles that platforms, governments, policy-makers and other stakeholders ought to take into account in their design, moderation and regulation practices. It does not offer a set of technical instructions, as their implementation will differ across diverse platforms and contexts and their operationalisation requires further investment and resourcing. Instead, the Manifesto builds upon the generative work currently underway with the proliferation of alternative, independent collectives and cooperatives, who are designing new spaces, ethical standards and governance mechanisms for sexual content. ” If you’re interested in how sex and sexuality manifests online and in the digital social space, this is a wealth of interesting resources and viewpoints on how best to manage the questions around content, behaviour, safety and appropriateness.
  • Depths of the Internet Archive: A Twitter account sharing DEEP CUTS and oddities from the Internet Archive – recent highlights include a link to where you can download the ACTUAL Stuxnet virus from back in the day, the note Bjork put on her personal website in 1999 about the Y2k bug, and an hour-long ‘introduction to the internet’ video from 1995 which features someone driving down an ACTUAL HIGHWAY (it’s a METAPHOR, do you SEE?). Superb.
  • The Geocities MIDI Archive: WOW. This is a searchable archive of over 150,000 songs extracted from the Geocities archives – basically a motherlode of 90s/00s tunes ripped from the past and made available to you here in the future. You can search the site, or browse by alphabetical filenames, but the labelling here is…inconsistent at best and I can’t vouch for how easy it will be to find that Natalie Imbruglia B-side you remember so well, but frankly even clicking titles at random should throw up some interesting stuff. If there’s any way someone can plug this into a 24/7 streaming service that would be great – I for one would absolutely listen to ‘The Sound of Geocities’.
  • BLAG: I featured a website about Ghost Signs the other week, and the person who used to run it (Sam Roberts, in case you’re curious) got in touch and told me about their new website, called BLAG which stands for Better Letters Magazine, apparently, and which is all about sign painting and lettering  and which if you work in art or design and particularly in this sort of field you probably ought to subscribe to right away. This is actually really interesting stuff, with articles that cover a whole host of topics from the broad general area of ‘painting words on boards’ (I am not being facetious, promise).
  • Fake Graffiti Generator: Look, I don’t know whether or not this is enough to fool someone, but I have a sneaking suspicion that you might be able to trick people into thinking the outputs from this site are real if you post them to Twitter and let its legendarily-sh1tty image-compression do its thing. You’re given an MS Paint-type interface and encouraged to draw/write whatever you fancy, which will appear as a vaguely-stylised and filtered overlay on an image of a nondescript brick wall – you can export your work, crop it to remove the ‘this is made on the web’ disclaimer and HEY PRESTO, exactly the sort of thing which, if you judge it right (wrong?) you’ll be able to parlay into some mid-level social media engagement! Bonus points for anyone using this to mock-up some sort of violently republican messaging and sharing it with shocked disgust on the TL – ALL OF THE POINTS if you can manage to trick the Mail into running it.
  • Sean Bonner’s Voxel Gallery: I know, I know, ‘art gallery in 3d virtual space’ is old hat and played out and NOONE CARES – that said, I think this is interesting and better than the usual stuff, and is an interesting nod towards some sort of idea of shared persistent virtual space with a consistent use-case (note I am not using the ‘m’ word here). To quote Bonner, “last year I bought some virtual property in CryptoVoxels (which has recently rebranded as Voxels) and I set up a gallery showing a bunch of my photography. I spent all the other day re-curating it to put a bigger spotlight on the work of others that I’ve collected in the past 2 years and I’m delighted with the way it turned out. I did a kind of “walking tour” on Twitter which you can read along with, or you can just fly solo and look at it for yourself but on Twitter I talked a lot about what and why I bought some of these pieces and what art and photography means to me and what I’m trying to do with it, so maybe a bit more insightful that way. My work is still on the ground & top floors, but the massive middle floors are dedicated to others now.” This is interesting because it’s an actual personal expression of Bonner’s collection, displayed how he wants it to be displayed, in a space of his own design – there’s a a degree of creation and curation and personal touch here that made me value the experience of navigating it far more than most of the other ‘some walls with some jpegs on them’ galleries you tend to see. It’s also worth wondering around the rest of the ‘neighbourhood’ the gallery sits in – there’s definitely a small community here, which whilst it probably won’t convince you that any of this stuff is set to be anything other than a niche concern anytime soon was enough to make me feel slightly less skeptical about the whole ‘digital persistent curated shared spaces’ thing.
  • Insect Identification: Are any of you aspirant entomologists? WHY NOT IT LOOKS FASCINATING? Should you suddenly develop a hitherto-unimagined desire to get really into thoraxes and pupae, this website could become an invaluable companion – it contains a FCUKTONNE of information about insects and how to identify them, and it’s pleasingly old-school in its UX. No AI here – instead there’s a really nice system that guides you towards identification by stating with the basic body shape of the insect you’re seeking to name and then getting more specific from there. Or you can sort by colour, or number of legs, or any number of other variables – in the main, though, this is just a great way of looking up a bunch of terrifying-looking insect life (sorry, but anything with more than 4 legs is inherently unsettling, it’s the law – nothing good scuttles).
  • Regional US Food: A glorious Twitter feed which shares images of some of the more…idiosyncratic items available to Americans as they race to be the first person in their family to get all the types of diabetes simultaneously (sorry Americans, but nothing about this stuff is normal – DEEP-FRIED BUTTER IS NOT NORMAL). If you have any qualms about giving this a follow, let me entice you with the following descriptions: “The horseshoe: an open-faced sandwich from Springfield, IL made with Texas toast, meat (usually a hamburger patty), fries, and cheese sauce”, or perhaps “Pickle Pizza,  homemade dough w/ a dill ranch sauce topped with mozzarella cheese, dill seasoning and dill pickles.” Your arteries will have hardened just reading the previous 50 words tbh, so you may as well just click in.
  • Capcut: Powerful in-browser video editing tool, which will be a godsend to those of you who have to do TikTok stuff for work but would like to be able to do grown-up, professional things as you do so, things like ‘using a mouse’ and ‘looking at a screen that’s a decent size for your increasingly-iffy eyesight’, that sort of thing.
  • Monsters Everywhere: Is AR ever really going to have its moment? It feels a little like it peaked a few years ago with Pokemon Go, and that since then nothing using the tech has ever quite captured the public animation in the same way, and I am yet to see any particularly compelling use cases for it despite the theory that AR/MR is going to be a constitutional part of THE METAVERSAL EXPERIENCE (which, just to remind you, doesn’t yet exist and is almost entirely formless at the time of writing). Still, people are still playing around with it and Monsters Everywhere is a prototypical game/toy which is built on top of Snap’s infrastructure and which lets players capture, collect and battle cute little monsters which appear all over the world via the magic of augmented reality. There’s seemingly more of a Tamagotchi element to this than with Pokemon, and I maintain that I think there’s definitely something in an updated virtual pet – not convinced that this is going to be anything other than an experimental toy tbh, but it’s worth a look if you’re curious about the tech and its uses.
  • The 2022 Mineral Cup: Votes are being cast RIGHT NOW to determine which of the planet’s minerals is definitively the best (as determined by users of Twitter). You may think this is a joke, but then you scroll down the feed and the replies and you realise that there is a large international community of invested Geologists who are very firm in their belief that OPAL IS NOT A MINERAL and therefore should not be involved in this at all. This is WONDERFUL – one of those all-too-rare moments when you stumble across something that is very much part of Someone Else’s Internet and find it to be charming.
  • Elden Ring Lore: Between about February and April I basically spent every night from about 9pm-midnight getting very stoned and watching people on Twitch playing Elden Ring (look, it was A Bad Time and I had Stuff Going On, ok?), and I became slightly obsessed with it despite knowing full well I would never, ever play it. Which explains why I really ADORE this YouTube channel which posts short videos delving into the game’s lore and storytelling, extrapolating all sorts of things from the character and enemy and level design that may or may not be subtextually happening in the gameworld. You need to know the game to enjoy this, but I promise you that if you spent any time in the Lands Beyond you will really enjoy this a lot.
  • The Half-Bakery: “The Halfbakery is a communal database of original, fictitious inventions, edited by its users. It was created by people who like to speculate, both as a form of satire and as a form of creative expression.” Which doesn’t quite tell you the half of it – this is a great treasure trove of silly – but occasionally genuinely brilliant – ideas and surprisingly-serious discussion around their potential viability or otherwise; this is VERY active, seeing new submissions every day, including one from just yesterday which imagines a genuinely-brilliant new type of ‘end of the pier’ claw-grabber type skill game in which you’re tasked with picking someone’s pocket, with the contents of said pocket being yours to keep if you manage it without detection. Honestly, there is so much WONDERFUL inspiration here, you could lose hours.
  • 90s Cursor Effects: Literally just that – a selection of examples of effects applied to cursors, popularised by sites like Geocities back in the day. There is literally NO good reason why your staid, boring corporate website shouldn’t implement one or two of these to liven it up a bit – a trail of ashen material following the cursor on your crematorium website, maybe, or smoke trailing around as you browse the vaping provide of your choice. Come on ffs, live a little!
  • Why Ryan Hates Mondays: I confess to not knowing what is going on here AT ALL, but I was charmed by this site – it takes about 5 mins to explore entirely, during which you’ll navigate the cute isometric landscape as a small bearlike creature (is this Ryan?) who gently interacts with various bits of scenery as he goes about his day. I have literally no clue about the Mondays thing, but this pleased me far more than it had any right to.
  • Mingle: Finally this week, a lovely little game in which your task is simply to match the pairs. The art style here is so, so pleasing, and the animations give each shape/character (look, it will make more sense when you click, I promise) a real sense of personality – this is a lovely distraction from wondering about whether or not this marks a definitive end to Britain’s top-tier prominence on the world stage, and whither the nation, and all those questions which I imagine are now FAR more pressing to you than ‘will I be able to eat and stay warm for the next few years?’.

By Yuri Yuan Ye

LAST UP, RELAX INTO THE HOME STRAIGHT WITH THIS LOVELY AMBIENT-Y SELECTION BY DREAM CHIMNEY! 

THE CIRCUS OF TUMBLRS!

  •  Peculiar Manicule: Possibly not technically a Tumblr,but very much one in spirit, this site collects 60s and 70s design in one pleasingly-curated online space. “Enter the Day-Glo world of The Peculiar Manicule and explore an awe-inspiring archive of 1960s and 70s graphic design. Witness mind-blowing displays of ink on paper by designers and illustrators, both known and unknown, in four main galleries, Books & Magazines, Ephemera, Typography and Paper Playthings.” A manicule, by the way, is “A symbol in the shape of a pointing hand, used in printing and graphics to draw attention to or indicate something” (no, I didn’t know either).
  • Antique Animals: Illustrations of animals from old textbooks and novels and the like, with a certain pleasing aesthetic to them.

THE TROUGH OF (INSTA) FEEDS!

  • Brick To The Past: You may all be longstanding LEGO Insta enthusiasts, but this was totally new to me and pleased me no end. Brisk to the Past is an account dedicated to sharing detailed, large-scale historical recreations featuring LEGO – I would imagine they are currently desperately searching for a bunch of flat greys so as to accurately render the Queen’s coronation as a touching tribute, but til they get round to that you may enjoy perusing their past works including such BLOCKBUSTER moments from the past as ‘The Burning of the White House in 1814 (in LEGO)’ and ‘The March of the Etruscan Army on Rome (in LEGO)’. They do commissions too, and I can think of no greater gift to bestow on someone than, say, “Crowds of Mourners outside Buckingham Palace (in LEGO)”.

LONG THINGS WHICH ARE LONG!

  • Ransom Capitalism: The week’s other big news (in the UK at least) has been the announcement of what our glorious new Prime Minister is proposing to try and stop people literally freezing to death over the coming months – to whit, capped energy prices and a long-term promise of more expensive energy for years to come! To be clear, it was vital that something be done to mitigate the coming crisis at least to a small extent, but, equally, it’s not hard to see one or two problems with this approach in the short, medium and indeed long term. This article in the LRB unpicks one of the main problems with it as a solution, namely that it effectively cements the narrative that there is simply no option other than to pay the amount that the energy producers are demanding – which, self-evidently, really isn’t true at all. There’s something rather miserable about the extent to which ‘are there some degrees of profit in certain sectors and particular circumstances which we should determine are simply Not Ok?’ (to be clear, really not a radical line of thinking) is no longer a question we seem prepared to even have a theoretical discussion about.
  • Beyond Hyperanthropomorphism: This is long and quite chewy (but, I promise, not super-hard if you can get beyond some of the stylistic and linguistic conventions of the philosophical essay employed here) and so, so interesting – Venkatesh Rao writes about his thinking on the extent to which we should be worried about the development of a terrifying Artificial General Intelligence from the current suite of models and the extent to which breathless articles discussing concepts such as ‘sentience’ and ‘consciousness’ should be taken seriously. Rao’s argument is that they should not, and his reasoning takes you on a fascinating journey from Nagel (BATS!) and through a bunch of semi-related philosophical concepts to posit that, fundamentally, it is extremely improbable that anything derived from current data training models could ever attain anything we might meaningfully refer to as ‘sentience’, because “to the extent the internet is not about the actual world in any coherent way, but just a random junk-heap of recorded sensations from it, and textual strings produced about it by entities (us) that it hasn’t yet modeled, a modern AI model cannot have a worldlike experience by overfitting The junk-heap. At best, it can have a surreal, dream-like experience. One that at best, encodes an experience of time (embodied by the order of training data presented to it), but no space, distance, relative arrangement, body envelope, materiality, physics, or anything else.” Honestly, this is worth the effort – and if you’ve any familiarity with reading philosophy you’ll get into it after 800 words or so, promise.
  • Towards a Theory of The Creator: What does the term ‘creator’ mean in 2022? No fcuker knows, hence this post by Alexander Iadarola in which they think through some of the characteristics that define people who label themselves as such. A bit snarky, fine, but also feels accurate, and made me wince in recognition at multiple points (and to be clear I would never describe myself as such, nor would I want to be so described). Overall this is a really good overview of The Content Econonmy Now, or at least the experience of being involved in it to any degree: “Creators are concerned with signal-to-noise ratios more than anything else. Informational integrity in this context has little to do with what a piece of information says; it has everything to do with what it’s like or how it feels to consume it. These two qualities – what something says and how it feels to receive it – become indistinguishable. The parameters of this realm of saying/feeling are obtuse, because they have to do with desire, which is hard to know; it is normal for users to indulge in content that feels “bad” to consume (examples include hatereads, hatefollows, etc.)”
  • A Guide To The Metaverse: With all the standard caveats about the fact that IT DOESN’T EXIST, if you’re in the invidious position of having people ask you for explanations and overviews of WHAT THIS ALL MEANS (it means, dear clients, that you’re going to be sold a LOT of pointless-but-expensive digital activations in the next few years, some of which will win awards despite never having been experienced by anyone who doesn’t work in advermarketingpr. WHAT AN INDUSTRY!) then you could do worse than bookmarking this guide published by Infosys and written by tech journalist Kate Bevan, who is very good at this sort of stuff and pleasingly clear-headed about it all. Frankly you can probably c&p this into a series of client briefing documents and charge them £££ for the privilege – but, er, obviously don’t because that would be STEALING.
  • Me and GPT-3: Or ‘what does the machine know about me and how does it use that information?’ – this is a thought-provoking look at whether and to what extent personal data has been hoovered up by the OpenAI scrapers (it has) and what the machines do with it when prompted. This isn’t a scaremongery piece – there’s nothing ‘bad’ in here, and no suggestion that people will be able to find out your address or bank details by asking GPT-3 for them – but it’s a really good article in terms of reminding you that everything that these programmes ‘think’ is ripped from somewhere else, and beyond questions of bias and prejudice there are also interesting questions about the sort of factual and subjective material around individuals that is getting invisibly assimilated into the forevercorpus in ways we (all together now!) don’t really understand the implications of.
  • Social Mobility: This is very odd, but also surprisingly good and rather impressible (if, to reiterate, largely-inexplicable). Would you expect to read a 6-7000 word (at least) white paper on the current state of social mobility worldwide, the impact of technology on social stratification and the ossification of said strata, and potential routes that could be taken to ameliorate these situations on the website of a major international bank? No, of course you wouldn’t, and yet here we are. A product of Atelier BNP Paribas, the bank’s ‘technology and innovation unit’, this is a frankly dizzying piece of editorial, combining surprisingly-excellent and shiny design (SCROLLYTELLING!) with some not-exactly-hypercapitalist analysis of What Is Going On With Everything. WHY DOES THIS EXIST? WHO IS THIS FOR? Is this BNP trying to attract younger staff members by showing its surprisingly-economically-progressive side? Why is it so shiny? How much did it cost and how long did it take and how many people have actually read the thing? No idea as to the answer to ANY of these questions, so if any of you happen to work there and have insider gen then please do let me know.  Anyway, this is not only conceptually-baffling but actually really quite a good read – wonders will never cease!
  • The Lootverse: Do you remember a year or so ago in the heady days of the NFT boom, when something called ‘Loot’ launched to widespread bafflement? If not, let me remind you – Loot was a project that literally sold randomised lists of equipment, the sort of stuff you’d get in D&D or a fantasy-themed videogame, so users would pay however many Eth for a link to what was effectively a textfile containing copy like “Sword of Thorgandia +1” and “Potion of Prolapse”. At the time, many commentators expressed a degree of bafflement that anyone would fork out cash for something so ostensibly-pointless, while the developers waxed lyrical about the development of the wider Loot creative community, which would build a whole web3 ecosystem around the basic idea…which, amazingly, seems to actually have happened. Nothing hugely tangible to see yet, but this might be the first post-NFT project where all the guff about ‘community’ and ‘building a world around the IP’ actually seems to have a degree of proper, bottom-up support. So there are people developing videogames with Loot integration, avatar generators based on Loot owners’ equipment stacks, etc etc. Which, to be clear, is all still very niche and quite frivolous and very much a work-in-progress which may never develop much further, but it’s interesting that this has at least slightly started to embody a few of those far-fetched NFT/web3 dreams that everyone was feverishly having twelve long months ago.
  • Critterz: Another cautionary tale from the world of play-to-earn – Critterz was a P2E system built on top of Minecraft, a really interesting idea which iirc I featured here about 6 months or so ago, and which per every single example of this sort of idea I’ve seen to date briefly promised to CHANGE PEOPLE’S LIVES by letting them earn cashmoney for building in Minecraft before the mod got banned by Mojang (the game’s developer) and the bottom fell out of the whole thing.That’s not the story here, though, so much as Critterz apparent status as another in which the labour of people in the developing world could be exploited by richer people for financial gain – I challenge you not to do a small inward gasp of horror at the bit where one interviewee talks in self-awareness-free terms about how great it would be to have poor people being paid minimum wage to ‘play’ as NPCs in online games, adding an extra layer of human verosimilitude for rich subscriber players to enjoy in exchange for a handful of low-denomination Coins. I wonder when we’re going to see our first big people trafficking / virtual enslavement story?
  • The Great Fake Instaverification Scam: This is fascinating – the story of how people are conning Instagram into giving them the coveted blue tick of verification by paying agencies a fee to establish a fake profile on Spotify as a musician, which apparently is all you need to convince Insta you’re a notable person these days. The margin’s here are eye-opening – approximately $1500 outlay for a service charged out at $25k is GOOD MONEY by anyone’s standards (also, $25k for an Insta blue tick seems like…a lot, although I appreciate it’s seen as The Gateway To Influencer Gold).
  • TikToketamine: Ok, so the actual title is “They’re Pushing Cut-Rate Ketamine Therapy on TikTok” but I will fight anyone who contends that my revision isn’t loads better. This is a slightly scaremongery look at people peddling ketamine therapy on TikTok as a wellness/mental health treatment – the basic gist here is that ‘this isn’t regulated enough!’, which is a fair point, but there doesn’t seem to be anything bad happening here per se. The overall point, though – that TikTok is increasingly the wild west of EVERYTHING, and there is going to need to be some sort of pretty serious action soon to work to regulate its usage for this sort of stuff because, honestly, this is going to make YouTube look like the Encyclopaedia Britannica before too long – makes sense.
  • The Professional Try-Hard is Dead: You’re bored of articles mentioning ‘quiet quitting’; I’m VERY bored of them. Still, this one is worth a read because a) it’s Vanity Fair and the writing (by Delia Cai) is better than most of the other stuff published on this topic; b) it’s pretty funny; and c) it sort-of touches on what I personally think is the heart of the matter, an evolution of the general Graber ‘bullsh1t jobs’ thesis whereby one of the reasons noone really seems to give a sh1t at the moment is that the juxtaposition of ‘this fcuking world we live in’ AND the very specifically-ridiculous nature of the majority of tertiary-sector white collar employment has rendered the truth of our working lives even more starkly-preposterous than normal. It has never been clearer that the work we do doesn’t matter in any meaningful sense beyond its ability to deliver us the means of paying our bills, and as such why ought we do anything beyond the very bare minimum to achieve that end? WE SHOULD NOT. Seriously, I think this is A Thing, though perhaps (lol) I’m articulating it imperfectly here.
  • Date Me Docs: On the continuing generational dissatisfaction with The Apps, and the apparently growing trend of people writing ‘dating docs’ about themselves to attempt to engender better/more meaningful/deeper connections with people by, er, giving them an exhaustive list of all your wants and desires and obsessions and tastes and likes and hates in advance, so you can eliminate all that tedious and potentially dead-endish ‘getting to know you’ palaver. I don’t quite know what to think of this, or what the ‘why’ is here, but I have some half-thought ideas about generations that have grown up being told that all personal problems can be solved through optimisation and ‘hacks’, and the tyranny of ‘more information = better decisions, always’ (as well as my everpresent theory that the Harry Potter series ruined an entire generation of people by presenting the idea of the Sorting Hat and, by extension, the concept that everyone has a bucket that is right for them, and can be put into exactly the right bucket as long as you know enough about them).
  • A Brief History of Portugal: I knew nothing about the history of Portugal as an empire before reading this, and now I know ever-so-slightly-more-than-basically-nothing – this is really interesting, not only on the past 500 years in Europe but also some of the practical challenges of attempting to maintain an empire when your country is, objectively, quite small.
  • Longevity, Inc: There are many wonderful details in this piece, about the Canadian man who’s spearheading a North American longevity revolution amongst the very rich, but I think my favourite is the throwaway line about the fact that he used to be a celebrity bespoke tailor (closely followed by his admission that, fine, a lot of the treatments don’t actually work in any meaningful sense but they do look really good on Insta). As ever with articles like this, the piece is a reasonably-balanced mix of ‘lol look at the rich idiots’ and ‘hm, I am not sure that these are necessarily the people I want to live longer than anyone else’.
  • ZOO8: Many years ago I worked for an agency that did arts and cultural PR, which basically meant ‘anything that the founder and staff thought was sort-of cool’ (it may not surprise you to know that nobody got rich from this endeavour, and we all took quite a long time to recover). One of the clients from that time – which I didn’t personally work on but very much enjoyed hearing about – was ZOO8, a music festival in 2008 held in a zoo (DO YOU SEE???). Now whilst that sounds like a terrible idea, not least for the animals, it turns out…actually, no, it really wasn’t. This essay tells the story of the festival and how it all went very wrong – based on my knowledge this is a largely-accurate account, although it omits a) the naked man with a machete who spent a large part of the Saturday wandering around the site threatening people, unmolested by security; and b) my mate Paul, who was accosted by a stranger on attempting to leave the site, which stranger attempted to rob him of his jacket and then shot him with an air rifle when he demurred. PR IS GLAMOROUS!
  • In Search of the Golden Brain: This is a great little read about a Masquerade-style treasure hunt included in a Spitting Image tie-in annual published in 1985; spoofing the then-inescapable craze for MASSIVE PRIZE TREASURE HUNTS that the original game had spawned, the annual contained what looked very much like a spoof version in which players had to solve clues to discover the location of Ronald Reagan’s missing brain (topical lol) – except it turns out it wasn’t in fact a joke, and there really was a golden brain out there waiting to be found. This is lovely, and also proof that Spitting Image (in its original incarnation at least) was properly smart.
  • Not Cooking In Rome: Another essay by Rebecca May Johnson on food, this time on eating street food in Rome. This is perhaps a bit self-indulgent, but the writing is SO GOOD and so evocative that I can almost taste the pizza rossa from the place across the road from where I used to live.
  • A Drinker’s Peace: I haven’t seen Withnail & I for years, and I confess to having consigned it to that oubliette of things I consumed too much of as a teenager to ever really enjoy again (see also: The Velvet Underground and Nico; On The Road; Pump Up The Volume; Green Ginger Wine) – this essay, though, made me want to watch it all over again. It’s a lovely piece of writing by Travis Woods about the film, the 60s, the passing of time and the loss of innocence and hope, and it managed to make me (almost) forget about all those dullards who would insist on quoting it whenever they had more than four pints.
  • Men of the Night Shift: Finally in this week’s longreads, a beautiful piece of writing about a short stint spent working a night shift, and the strangeness of the places and the people you see when so doing. I don’t know if you’ve ever worked nights for any period of time, but after a few weeks you really do find your whole perspective on things…tilting ever so slightly, and this piece does a wonderful job of communicating that slightly-off-kilter, overcaffeinated inner-skull buzz that you often get just before 4am on a quiet industrial estate.

By Huleeb

AND NOW, MOVING PICTURES AND SOUNDS!: