Stephen Whittle

@mov-eax-rgb.bsky.social

Game designer, programmer, artist and hardware dev. Unreal Engine contributor. He/him.

papers please style game where you are in charge of the little metal detector station at the front of the Federal Department of Control and have to stop people from bringing in banned material like "any form of branded packaging" and "apples"

What is happening is not that AI-generated cultural goods are replacing human-made cultural goods; what is happening is that capital is flowing out of the production of cultural goods *entirely,* making us all poorer for it.

I have blamed The agents That hacked Your website And which You probably Reasonably thought I controlled Forgive me I am rich And unaccustomed To taking responsibility

33 years ago today, the Stranger stepped onto the dock of Myst Island for the first time. We can still hear the water splashing against the pilings.

The dock of Myst Island from the original version of the game. A wooden dock ends at a spiral staircase that leads up to giant gears on a hillside. A ship is sunk in the water on the right, and parts of buildings are visible on the left.

More accurate headline: "After killing Bungie's cash cow out of spite and seeing Marathon's dwindling player base, Sony allows Bungie to perform necromancy on Destiny 2 with not enough necromancers"

I was already honored to be a voice in their game, but now I’m DOUBLY grateful to have @doublefine.com’s continued support as they unabashedly celebrate our collab! (And please don’t worry. While my success & public joy seem to be upsetting to some, I feel like I’m getting to live my dream life!)

TheGamer@thegamerweb.bsky.social · last wk.

Double Fine's decision to cast Erika Ishii has sparked harassment and abuse online, prompting the studio to shut down replies.

I'm pleased to announce that around 200 pages' worth of tutorial resources are coming to the Game UI Database for FREE via a brand new blog. No ads. No paywalls. No email signups. Here's a preview of the first piece dropping next week, all about crafting cross-platform control schemes and layouts!

BildBild

I will probably be as repetitive about this like I am with backups, bodies, and batteries. Australians, plan to be where you want on July 22 2028 to see the total solar eclipse, you ferals.

Nanoraptor@nanoraptor.danamania.com · 8mo ago

A reminder to Australians, we'll get a decent solar eclipse in a little over two years. eclipse.asa.astronomy.org.au/eclipse-aust... Centred right over Sydney during the later part of its totality, it'll also cover Bourke, Nyngan, Gilgandra, Dubbo, Orange, Bathurst, Mudgee, Lithgow and Katoomba.

A map showing NSW's path of totality for Australia's 2028 solar eclipse, showing it passing right over a wide swathe of new south wales, including Bourke up the north west, down almost centred over Sydney in the south east and covering frankly the vast majority of the population in New South Wales.

And I people that have that naive interpretation about texts, and who think of them mostly as Things That Encode Meaning, and don't have a good sense of what it means that they are Artifacts are a lot more likely to see the artifact produced by robots and assume "oh hey there must be Meaning here"

I’m absolutely sure this is Steam being huge babies as always (see: ACCC stoush last decade) and we *do* need to explode the social media companies, but we gotta get this commissioner out. It’s wild to let someone have this amount of power with seemingly no checks on it

Interesting from Terence Tao on his Mastodon. I'm not a math guy (although I am a math fan and observe math developments with interest) but I think there are two points here that are worth noting, one of which I haven't really thought about in these terms before

Terence Tao
@tao@mathstodon.xyz

While working on the revision of the Equational Theories Project (ETP) report at https://arxiv.org/abs/2512.07087 I came across an observation that we quietly made at the back of that report (see Section 14.1) but has become more relevant today.

The ETP was a successful project, running over several months, to resolve over 22 million implications in universal algebra (and formalize them in Lean) through a variety of methods, both human and automated, although in our case the automation was much more in the ``good old-fashioned'' form of utilizing automated theorem provers (ATP) and simple scripts, rather than modern LLMs.  If ETP were run today, it is likely that modern LLMs would very quickly resolve almost all of the implications studied, and the few remaining holdouts would likely then fall in a matter of days with additional LLM compute. (1/3)Terence Tao
@tao@mathstodon.xyz

However, as observed in Section 14.1, there were noticeable opportunity costs with moving too rapidly.  We noted a few cases where a key implication stumped us for a while, and forced us to come up with interesting new techniques to construct infinite counterexamples to that implication; but subsequently it was found via an automated theorem prover that the implication in question actually had a relatively small finite counterexample.  But if that counterexample had been found first, we might not have discovered the more interesting method.  We in fact suspect that some of our early automated sweeps to resolve implications may have accidentally obscured some potentially fruitful problems that would have revealed even more constructions or insights, although we do not have the counterfactual data to confirm these suspicions definitively.  For better or for worse, the ETP problem set is now "solved", and there is no practical way to undo this and recreate the unsolved status of this problem to encourage further exploration.

For the first time in the history of mathematics, we are faced with the situation that open problems in pure mathematics -- one of the primary fuels for mathematical discovery -- are being mined as a non-renewable resource; and if the solution is extracted poorly from that resource, then the net value to mathematics gained as a result can be significantly less than if it were obtained in a responsible and sustainable fashion.  (2/3)Terence Tao
@tao@mathstodon.xyz

One response to this situation would be to accelerate "propspecting" for new sources of problems.  The ETP itself was in this vein: the project was deliberately chosen to open up a new source of problems to study, rather than to exhaustively mine out an existing productive vein of problems.  And the ETP problem set (implications between single equational laws of order at most four) is particularly amenable to scaling out further; one could quite plausibly now run an ETP for order five equations (which would be 3 billion implications rather than 22 million), or stay in order 4 but work out the implications of pairs of equations rather than individual equations (leading to about 40 billion implications).  

However, in most other cases, the most interesting new problems to ask are not simply formed by enlarging the parameters from a previously solved problem; it is usually the process of solving the problem itself which naturally suggests what the next directions will be.  The modern trend of solving problems by purely automated methods, with no emphasis on human understanding, is thus akin to non-renewable resource extraction, achieving the immediate goal of solving the problem at hand, but at the cost of discarding the seeds of more productive questions to ask in the future. (3/3)