Another game that was originally made as part of the Slow Game Jam, 4Ever Transit Ride takes you on bus trips through procedurally-generated cities.
Your interaction with the city is indirect and at a distance: you can choose when to transfer to another line or how much to slouch in your seat, but the buildings will be forever out of reach. Yet they still feel a lot more concrete than many other outside-the-window decorations, which I suspect is due to the entire city being generated. Even if you’ll never be able to touch that tower in the distance, it’s still there, in the same relation to the other buildings, until you move on to the next town.
I like it when artists talk about their process, but
Anders Hoff
(writing under the name Inconvergent)
has gone a step further and released a toolkit, named “snek”. It’s a generative programming system implemented in Common Lisp that composes changes in encapsulated objects (called alterations) that lets Inconvergent build generative processes without having to worry about managing the intermediate state.
It reminds me a bit of parametric geometric operations (such as the CAD-inspired ones used by Tweet 2 Form) and a little bit of Quil’s functional take on Processing, but snek is very much its own thing too. I look forward to seeing what Inconvergent builds with it.
Ethics have come up here a fewtimes before,
though mostly in the context of bots. This stuff is important to think
about if you have any intention of releasing your generative projects
into the wild.
Liza uses a number of examples I was previously unaware of, and puts forth three principles: anticipate deliberate misuse, consider how code is a powerful amplifier, and show your work.
I think that the importance of showing your work is the one that have been the most overlooked. Despite things like the EU’s right to explanation, most algorithms are still opaque. And, as Liza points out, most people have vast misconceptions about artificial intelligence.
How will you think about asking for an explanation when you don’t realize there’s a question? Can you tell what was made by a computer and what wasn’t?
I share Liza’s concerns about the implications of the intersection of all three guidelines. Your news streams are already being polluted by maliciously-spawned fabrications. I’ve pointed out many ways that images can be manipulated. Now picture all those clickbait articles automated.
People with unusual or unpopular problems will suffer more: people who don’t want to listen will have a new excuse to claim everything was Photoshopped, even when they can’t really tell the difference themselves. We believe evidence that supports our beliefs and discard evidence that makes us feel uncomfortable.
So, paradoxically, we’ll have people believing all kinds of things that fit their biases while rejecting things that are true but discomfiting. If you thought Flat Earthers and Moon-Landing denialists were bad, wait until they can claim that everything is manipulated while understanding none of the science behind it. Truth becomes what we already believe.
I share Liza’s belief that generative artists can help slow this. By exposing more people to how the systems actually work, countering abuses of the tools before they happen, and taking steps to deal with the ability of computers to scale problems up to massive levels, we can make our own corners of the world a little better.
Forest Walk is a quiet, undemanding walk through generated woods.
The recent Slow Game Jam was about bringing the slow movement to games, a bit like slow cinema or slow television. What it means for a game to participate in slowness had several different interpretations, among them this forest walking simulator.
Marrying generated spaces with slow games plays to the strengths of both: the variety from the generation keeps the game from exhausting its outputs early, and the undemanding nature of the experience lets the user take a step back and engage with the results of the generator on its own terms.
There’s a lot going on right now, what with the generative poetry and the neural-net sketches, but I don’t want the conclusion of this year’s 7-Day Roguelike Challenge to go unnoticed.
Not only are the games finished, but so are the reviews. Looks like there are some interesting ideas this time around.
A poetry generator by Camden Segal, from NaNoGenMo 2014.
It uses public-domain text from Project Gutenberg to build poems that match a given rhyming scheme.
It’s a good example of how poetry is both easier and harder than generating prose. The patterned syllables and rhyming schemes give the words a structure that random prose lacks. The form makes it easier for chance associations to create evocative images, like a ship emerging from the mist.
But then the image falls apart on the next line, dissolving in the sunlight, and we see the difficulty of poetry: a tightly-constrained form like a haiku demands a density of meaning that this random shuffle approach can’t meet.
Still, it’s a commendable effort. And it does write very, very bad limericks and hallucinatory sonnets, so it has its uses. I’d like to see more attempts along these lines, perhaps taking meter into consideration or finding a way to preserve more of the meaning from the source text.
A sonnet:
What an eagle swooped around him like brooks, valiant place those not to put his hunger, spring which in his head of the heart he looks, appear with the heavy doors of a younger, of murdering of advanced as a pall, was opened his head so strongly tempted, his eye chapter ii black i humbly call, him so that after them in the frenzies, the most poisonous they were presiding, aspect the whale when after sunset and bronze, defyingly i use a braver thing residing, side upon the fishers mostly young dons, inside of seven hundred and erring, a year totally lost his feet glaring.
A haiku:
Sensation cases, lucifer dico qui vive, hospitable folk.
Some limericks:
Her fate that ship would have always fun, and never do you choose them and run, father’s funeral philip, very genuine filip, agents could not amenable to stun.
To spread abroad i felt almost sure, immediately in danger of perfect cure, pleasure or too busy, cannot my dear lizzy, in a servant therefore must beg your.
Other circumstances arriving, situation the death of deriving, of the least getting rid, punctuality both however did, should be with many of deriving.
Seedship is a game by
John Ayliff
that makes extensive use of procedurally-generated text to drive the game. You take on the role of the AI who is shepherding the last hibernating humans across the cosmos in an attempt to found a colony before the ship’s systems fail.
The basic loop is a push-your-luck decision of whether to stay with the imperfections of the current, known planet or to push on and attempt to find a better one before the hazards of the journey damage the ship beyond repair.
Where the generative text really shines is when you finally decide to found a colony and watch the consequences of your decisions play out. The text reflects the nuances of the simulation in a way that both hearkens back to some of the earliest text-based simulations while also demonstrating that modern, Twine-based development is fully capable of creating dynamic, expansive experiences: you don’t need whizzy 3D voxel tech to generate interesting stories.
The output wasn’t exactly brilliant literature, but it was computationally generated:
Darling Sweetheart
You are my avid fellow feeling. My affection curiously clings to your
passionate wish. My liking yearns for your heart. You are my wistful
sympathy: my tender liking.
Yours beautifully
M. U. C.
Honey Dear
My sympathetic affection beautifully attracts your affectionate
enthusiasm. You are my loving adoration: my breathless adoration. My
fellow feeling breathlessly hopes for your dear eagerness. My lovesick
adoration cherishes your avid ardour.
Yours wistfully
M. U. C.
“M.U.C.” stands for “Manchester University Computer,” of course.
The
combinatory
approach it used–with simple templates drawing from lists of words–is largely the same technique that is the basis for things like Tracery.
The Mastodon bot (by, I believe,
Colin Mitchell) seems to be based on on the emulated version researched by David Link, though there are also PHP and Python versions floating around (I’m not sure how authentic their word lists are).
It strikes me that a Tracery version might be a good project for NaPoGenMo…
It’s that time again: time to make generated poetry!
NaPoGenMo is a much more laid back event than NaNoGenMo (which is already pretty laid back): you don’t have to make a long poem. just any generated poetry. Try making a Haiku generator in Tracery, or a limerick generator with RiTa.
I’m looking forward to seeing what poetry comes out of this year’s projects.
