Box 024: Speculative Anything

21.12.2021.12.20

DS019: Cycles: Glorious reflections, accurate lighting, 5 minutes per frame rendering time. Tried to set this up in Eevee originally but the lighting got too complex.

The grey-blue is shattered by the streaks and beads of water, like the sky is coming apart in layers.

Don’t worry, it’s a brief one this week.

For some reason I’ve been listening to the soundtrack of my early twenties the last few weeks. Bloody loving it too, though a part of me is concerned that this is the beginning of the road to ‘they don’t make music like they used to anymore’ as a I clutch a copy of Panic at The Disco’s A Fever You Can’t Sweat Out and my eyes glaze over. I genuinely haven’t listened to new music in so long, I don’t know where I’d start. I need some new music without words to work to. I’ve rinsed the chilled cow stuff and I think jazz and blues are too important to be background music. Anything in the Zombie Zombie, Waveshaper, Ratatat part of the world? Wavy synths, whiny guitars and crispy drums pls?

I was doing a lecture the other day for some students and was talking about when I started turning away from doing futures stuff and more towards now stuff – I usually say ‘critical technical practice’ after Philip Agre. This is also connected with last week’s moan about the same version of the future being tramped out over and over again – this vague and unprovable notion that the future has basically been suffocated by its own hubris. That ‘futures’ are a sort of runaway out-of-control Fantasia scene of glossy interfaces and sincere utterances of ’empathy.’ That it’s bought into its own hype and is just reproducing – almost robotically – the same visions based on a toxic concoction of imagined nostalgia dressed up as ‘authenticity’ and the never-ending, pointless quest for power and control through the salami-slicing of the world into thinner slivers of ‘data.’ A sort of #developmentaesthetics for technology in general.

So I read some Alfred Nordmann as recommended by Jack Stilgoe. Nordmann was instrumental in a lot of discussion about speculative ethics around nanotechnology and the brunt of his research was his contention that speculating on future ethical conundrums not only distracts from current ethical conundrums but also constructs imaginaries and expectations of the technology that may be unfeasible. This is exactly the stuff I’m writing about at the moment so it was delicious. So by saying ‘what if a rogue nanotechnology got really into pineapples and then converted all of the planet into a giant pineapple farm?’ you create a rhetorical expectation of a pineapple-obsessed nano-goo.

So why are people drawn to speculating on these future pineapple or non-pineapple conundrums? Well, because the future, as presented from the present is an apparently cost-less enterprise. It isn’t ‘real’ and so is a playground for imagining things in at their extremes. This, I suppose, is the basis of speculative anything – ethics, design, philosophy etc. But, as Barbara Adam points out this makes speculators somewhat ‘trespassers in the future present of others’ because by setting up expectations and imaginaries, even extreme ones, we are foreclosing futures for others; ‘we are inescapably illegal immigrants in their worlds; and that they have not given their consent.’

To go back to Nordmann, speculating also doesn’t necessarily tackle un-seen ethical issues that are emergent in the present such as exploitative labour practices in laboratories or shady nanotech funding, it just gets everyone very chin-strokey about pineapples. This is hardly a radical insight and many have had it before – long-time friends Changeist call it flatpack futures, for example. Futures can be like getting an earring at school; you’re cool for a 14-year-old, perhaps you manage to convey a sense of worldly cultured-ness on the playground but you still have to do the same exams everyone else does. I did anyway. I was definitely cool, everyone thought so.

Perhaps in times of crisis our imagination gets more and more foreclosed and so you listen in on chats between talented young folks and just think ‘hang on, isn’t this just Microsoft’s 2013 product vision thing?’ when they talk about future tech.

Where are you going with this Tobias? I don’t know! It’s a blog, I just throw words at it and sometimes something sticks, ok?

Short Stuff

I feel like Short Stuff should have a theme tune. Like it would if this was a show. Add your own theme tune, go back, read it again, sing it in that voice. Record it. That was fun.

An estimated 1.35 million people die a year in road traffic accidents, 153,158 of those are in the UK. That doesn’t even count the contribution cars make to the 7 million people a year who die of air pollution. Of those 1.35 million, 154 were in Teslas. Numbers of people killed by ‘AI’ is tricky to find; not only because AI is such a nebulous concept but also those effects might not be experienced for years. Directly, robots killed 33 people in the US in 2017 but that doens’t even necessarily mean AI. What I’m getting at is the best way to avoid being killed directly by an AI is to ride a bicycle. Cars are still arguably the biggest direct technological threat to human lives. (I have the bucket of caveats under my desk if you need them.)
I’m un-fussed by pineapples. You know, in a fruit salad I’ll take it but I don’t think I’d ever just buy a pineapple.
I started listening to How To Save a Planet after having it backed up for a while. It’s brilliant.
Who’dathunk that trickle-own economics is a sham?
What’s So Great About That? is a new YouTube channel I’m really into. It’s basically another cultural commentary one but it’s really well produced by Grace Lee. ‘Be Gay; Do Crimes. Untitled Goose Game: Is it Good or Bad?‘ was a joy. Particularly the Twitter furore over the paradox of the goose as both a leftist and rightist icon while the developers continue to insist that it is, in fact, just a goose.
God I’m almost rid of Evernote.

Ok, there’s probably going to be another one next week. Don’t worry I won’t do one of those tiresome year-end recaps. I can’t remember further back than Wednesday anyway. My memory is shot to pieces, always has been, I wonder if it’s a by-product of my compartmentalising – I only cling on to information that is functionally useful rather than associate it with any particular time or place. That’s why so many things I say start with ‘someone said’ or ‘I remember hearing that.’ Not very good at referencing. Look, I will shut up now. I love you, have a bloody smashing week if you’re taking it easy and an easy one if you’re taking it hard.

Box 020: Bringing Democracy to Everything for Some Reason

23.11.20

DS015: More lighting, some cloth simulation. Nothing too complex this week. I’m trying to work on composition a little too – make the scenes feel more busy.

This morning the light looks like big slabs of American cheese slapped across the brick fronts of the buildings. It’s a kind of sickly, thick yellow with the veins of wires and pipes and cement running underneath.

Last week felt hit or miss. Like it could have gone either way. There were a bunch of deadlines and things in the air which I was totally prepared for coming down wrong or not landing at all but, miraculously, it all seems to have worked out which gave me a nice high into this morning. I suppose it all restarts today, back to the confusion of inputs and outputs to be arranged in just the right order. It’s a short one this week, all those balls in the air last week kept me distracted.

I saw this article on How DIY Tools are Democratzing Science and the title just stung me a bit. There’s not really any radically new ideas there, it’s all 3D printing, desktop labs etc. The stuff that’s been around for a while; it’s more this notion of democratisation. ‘Democratisation of x’ sits in an uneasy relationship with ‘disruption of x.’ They both imply a change in a culture or industry. While disruption is usually attached to some vaunted genius with a startup idea, the insight to see what others could not and all the trappings that brings, democratisation is usually about open-source tools and accessibility. Now, I definitely prefer one of those forms of technological change to the other. And particularly when it comes to science and the value of openly sharing data, results and theories to ensure integrity. We all know what happens when startups wonder into science with over-hyped promises locked in black boxes.

But ‘democratisation’ feels as misleading and overblown as ‘disruption.’ It’s a played-out argument but ‘disruption’ usually just means leveraging capital to do what was already there slightly cheaper: Eg, Uber didn’t disrupt transport, they just aggressively undercut the competitors by sticking to the grey areas of legislation, sometimes flat out breaking the law and undermining norms. DIY and open-source science tools will doubtless engage more people in research but it’s not ‘democratisation’ of science unless you’re providing free education to train new people in using those tools. All you’re doing is making science a bit cheaper for those who already know what they’re doing and have the privilege and access. I saw another one this week on the ‘democratisation of design’ which is a pretty common refrain. Is it appropriate to democratise design? Is design’s problem that it’s not democratic enough? The first step, again, in democratising design would be to find ways to provide free design education, to those who want it rather than forcing everyone – whether they care or not – to now have an opinion on and engage in design.

And what even is democratisation anyway? It has ugly overtones of regime change and of the mission to ‘bring democracy’ to Afghanistan and Iraq, Syria and all the other places Western armies are deployed. These ‘democratisation’ projects fuel military economies. Was the 15% bump in military spending by the UK government this week a part of ‘democratisation?’ What’s wrong with just saying ‘changing’ or ‘making more affordable’ or ‘opening up.’ I know, I know let’s not lambast media outlets for using overblown words, but I honestly couldn’t tell you what ‘democratisation’ means.

Short Stuff

Here’s Werner Herzog slapping down Elon Musk. It’s a pleasant novelty but not anything particularly insightful, other than the scorn in Herzog’s voice being very apparent.
I met Oliver Holms last week and he suggested Devonthink as a solution to Evernote’s progressive decline and unsuitability as a PDF library. Let me know if you have any other thoughts.
One of the things I’ve started to ‘get’ more in the last few years has been really good performance. Obviously, at the moment, that’s harder to come by. The initial spurts of generosity that flooded the Internet back at the beginning of Covid meant a lot of good ones got put out though. Like Celui qui tombe from Yoann Bourgeois in which the performers cling to a moving platform while Frank Sinatra croons ‘My Way’. Or The Statement by Crystal Pite which was out earlier this year.

Jack Stilgoe pulled together this panel about AI and hype – Fake It Til You Make It which will probably round off a bit of stuff I’m writing at the moment pretty nicely.
We have a vacancy in the programme for a lecturer in MA Data Visualisation. You can check out the ad here. If it’s of interest or you know someone who might be interested, let me know.

Anyway, relatively brief as I said. You could have probably not read the thing about democratisation, got to the videos and wrapped this one up in 30 seconds, but it’s too late now. As ever, I bloody love you, I do, speak to you next week.

Box 008: Lazy light and ambient criticism

26.08.20

DS003: Digital Sketch 3 uses the same ocean modifiers as last week with a procedural sand texture. This is duplicated with a sphere instanced across jittered vertices.

The Final Fantasy XIII soundtrack has been on full blast and even Scrivener has been pleasantly surprised that I’ve turned up to work every day. The scramble for virtue as we start to keel over into the new [academic] year seems to be bearing fruit so far to the tune of clacking keys and the whoosh of sent drafts.

Sticky, lazy Light

I spend quite a bit of time reading and thinking about light, colour, maths and rendering engines. I don’t want to retread it too much here, because I’m told that that stuff isn’t as interesting as I think it is, but I was trying to pin down what exactly the qualities of a CGI image are that make it obvious as a rendering – how in your gut you know that it was made in a computer.

From what I have observed, IRL light isn’t particularly good at doing what it’s supposed to. It’s lazy and unpredictable. It’s no respecter of vectors and bounces so much as feeling of where it might go and what type of colours it might produce. Light seems to operate on reckons and hunches in unpredictable ways. It might infinitesimally alter the shade of the shadows at the joining of your wall and ceiling by taking some blue from the lampshade in the opposite corner. It might scatter into three precisely even gradients around a tabletop – just to describe what I can see from where I’m sat.

Light travels slowly on the Disc and is slightly heavy, with a tendency to pile up against high mountain ranges. Research wizards have speculated that there is another, much speedier type of light which allows the slower light to be seen, but since this moves too fast to see they have been unable to find a use for it.
Last Continent – Terry Pratchett

In rendering, especially these ones that are very much in the zeitgeist, the light is too scientific. It follows the rules: There’s a light source, surfaces bounce it to various intensities and directions and the scene is lit. It gives you a sense of the void outside the render. In the non-rendered world there are no voids, no light and not-light, everything is sort of sticky and light just… hangs around, taking up space. That’ll be the thing that lets you know if you’re in a simulation, if the light looks too energetic, like it’s doing well at it’s job.

Of course this could just be a matter of better computation. A lot of very realistic looking renders do a very good job of casting micro-interactions to the point where you would very much struggle without some forensics to tell it’s a render. They’ve put in hundreds of hours of intense computation and heavy maths to get light’s sheer ambivilenbce Just Right. It’s those edge cases that are more interesting.

Ambient Critique

This piece for AIGA begins to point at something that I find quite frustrating in academic design work; the downward spiral of calls for reflection and critique that seem to emerge from a field that is perpetually put on a crisis pedestal. The article argues, with a particular view to graphic design, that critique is everywhere since design is the materialisation of ideas and thus suffuses the whole of culture. We don’t need to look for critique or create new space but find better ways of capturing the abundance of criticism in the material world; on Reddit threads and bus-stop conversations. I like this idea of a sort of ambient critical space. That rather than seeking out critique or building elite forums to enact it you could just sort of absorb it from the information environment by osmosis. Or even seek it in other places. What if you uploaded your work to 4chan instead of Behance?

The article also rings parallels with the idea that there is no discipline of design; that it’s dissolved into everything else. Whatever you think of it, I’m very much into this idea, even if it’s just an Overton Window to move the conversation about definitions, disciplines and ‘reflection’ on or better relegate it to the trash pile of history. I’ve sat in too many hot, dark lecture theatres and watched super-smart people who can’t agree call for more critique and reflectivity as a way to mollify the audience and each other.

Cynically, it’s also a way of reinforcing power in the cabal gathered in the hot, sweaty, dark lecture theatre. I’ve been reading about how contested narratives in machine learning are sites of power, both ethnographically and sociologically. In these circumstances, arguments like ‘we don’t know enough’ or ‘what the other’s understand is wrong’ are ways of affirming privileged: ‘It is a strategy to exclude other voices in the discourse by stating that their claims are naive.‘

Recents

Bracket: Takes Action – the fourth instalment in the architecture-ish journal – includes a piece from me – Exit Spaces; From Koreshan Cults to Wireless Mesh Networks. The text explores Albert O. Hirschmann‘s four responses to change in an organisation or state through a contemporary technological lens. First of all there is ‘Exit’ – simply leaving the state you are unhappy with – then ‘Voice’ – protest – and then Loyalty; sticking with it. The piece examines extra-territorial or extra-state organisations through history, from American communes to the ambitions of Seasteading and finally the growth of mesh networks and distributed exit communities. I actually wrote the piece six years ago and it seems that the journal has had some problems which means it’s only just been published now. It’s available on Amazon and other, better book-sellers.

My website got featured in the Cargo Collective newsletter this week. Just a nice big screen-grab. I’m concerned this legitimises the hours I spent tweaking the CSS on it and that I might feel the need to comfortably retire into obsolescence as a relatively passable visual designer someday soon. That’d be nice. You might have noticed I’m still tweaking this blog. It’s getting closer to a place where I’m happy with it. Anyway, enough about me, what about the Internet?

Short Stuff

If you and I speak about games then you know that I think Disco Elysium is the best game I played in the last decade. Here’s a super interesting in-depth video examining the design of their character progression system – the ‘thought cabinet.’ They were inspired by the design of digital shark-baiting arena Twitter dot com as a space that supports and encourages conflict (not in a good way) which is what the voices-in-your-head of Disco Elysium are all about.
Tom Armitage sent me some stuff in response to my musing on game engines last week including this post of his about virtual cameras which was super interesting.
This also reminded me of the hacking of cult demo P.T. where after much effort, it was discovered that the ghost is behind you all the time which makes the legendarily rare and terrifying demo more horrifying rather than less.
I discovered that there’s an actual name for the almost universally-recognised call-and-response knock; shave and a haircut (two bits). There’s a name for everything these days.
Someone made an HTML5 version of Winamp – Webamp. All that’s missing is a CRT monitor and a pink studded belt and I am fully ready to go on my self-cosplay from 2001.
The bottom of this blog has an email subscription option and a couple of folks have put their email addresses in there. I haven’t used it yet but I might one day so you can put your email address in there too if you like.

That’s all, love you, speak to you next week.