Unbridled Thoughts: 5

05/02/2026: Considering the onboard control technology on the Enterprise of Star Trek The Next Generation: it looks like the crew has to use a lot of touch screens that display both dynamic readouts and controls. This is hard for me to believe. I feel like static physical buttons to control a dynamic set of commands (like how keyboard + computers work now) will always be preferable to dynamic controls directly on a touch-sensitive thing. Especially for critical interfaces and those that don’t need to minimize their physical size. Even if the touch screen is perfectly responsive, it is not ergonomic, lacks tactile feedback, and increases the likelihood of mistake input. For extra-important things like weapons or propulsion that can’t afford to have accidental activation, it seems like guarded switches would be hard to improve on. Of course, touch-sensitive panels are useful for things like capturing complex hand movements in 2D (drawing, navigation in a virtual space, other multi-finger things)… I just can’t imagine them being the future’s choice for critical button-based controls.

05/03/2026: I wonder if an archeology website would be enjoyable to make. Maybe just focused on some very simple general aspect of pre-history modern humans.

05/03/2026: Considering how engagement systems and feed algorithms work on social media platforms, it is hard to imagine how they could be expected to avoid the things that are so harmful about them. People talk about “the algorithm” but I think it is important to realize that it is every part of a system, each of which is designed to generate profit for the platform. If the algorithm is Santa, think of the entire system as the rest of his infrastructure (Ms. Claus, “worker” elves, reindeer, parents) that enables him to fly around, enter everyone’s home, make it his job to know their naughty business, and love bomb if they please his peculiarities. The emergent toxicity of these complex systems is an interesting thing to think about. I think thinking about what is behind the feed and noticing what causes me to develop new behavioral patterns is good antidote to some of their most detrimental aspects. Even with some awareness, I’m overwhelmed with negative effects… I’m glad to realize that mass social media platforms are absolutely not for me.

05/03/2026: Yesterday I got an email from THE once-very-popular social media platform that I joined in early 2007 and was active on until sometime around 2013. My account has now been permanently disabled for egregious and repeated violations of their community standards. No further information is available. I know this was a mistake made by what locals are calling a “rouge AI” that is poorly performing ban-based moderation. But still, hearing false accusations, however stupid and vague and unfounded, imparts a strange sting. Having a poorly-implemented automated system forever erase my early online historical presence while putting me on a list somewhere is beyond silly. I’m not allowed to make any accounts in the future (a silver lining). There’s no recourse. I’ve sent an email just to say it was a mistake and to request they tell me what triggered the system in case a human can do a retrospective, but I really don’t want anything to do with them, now more than ever. It sounds like this has happened to many people lately. I feel bad for the dad account I read about who was banned for a decade-old picture of his family posing with Chuck Norris (at a photo-op), flagged as “Human Exploitation”. Maybe I’m better off not knowing what ridiculousness caused my neglected account’s demise.

Grant Ryan King profile image

05/04/2026: Why does the soliciting “pest control” business bother me so? I think there are many reasons. It is more than just the sense that solicitor’s marketing tactics seem predatory to the at-home lonely elderly. It is very bothersome to have someone going around poisoning critters that then distribute that poison to nearby properties and other bigger critters. Causing an ecological imbalance will only make things seem better in the short term, but over time the imbalance will get worse, as will the impulse to temporarily mitigate it with more poison. One of the most frustrating issues with soliciting pest control companies is that they seem to initiate this imbalance by marketing with misinformation about preventative care.

05/05/2026: I published a few minimal video streaming websites last night. The design is very minimal: each website is just a responsive player for a single video file. The system doesn’t handle the data efficiently. Seek controls don’t work as expected. It is not an optimal way to watch or to host this media. I would be better off embedding the video, but still I feel compelled to host the files for some reasons. I’m thinking about what I could do that doesn’t involve involving obscured pre-made streaming solutions.

05/06/2026: I feel very silly for banging my head against the Azure wall for so long. How much time and money could I have saved had I reconsidered AWS sooner? I forgot to shop around when Google Cloud started feeling the way Azure does now. Anyways Amazon Web Services looks like a dream these days. It is amazing how a modern UI can really make things feel less stressful. This AWS UI is the most inspiring I’ve seen in a long time.

05/06/2026: All of the efficiency problems with the video streaming websites have been resolved. It all came down to Azure nonsense. It really was a large heaping pile of nonsense. Basically their REST version defaults to something used in 2009 that modern video players (the “bytes” header’s syntax) doesn’t play nicely with, so the client couldn’t get partial content, which was breaking the cache and causing the entire giant movie file to start re-downloading and playing from the beginning when attempting to seek. Changing this behavior is crazy difficult. After overengineering a semi-solution that would have been full of trade-offs, all it took to fix everything was migrating movie file storage to AWS S3 and off of Azure.

05/15/2026: The last few days I keep thinking about this thing that is surely wrong but I have to write it down so I can stop wondering about it for now. There is a type of discrete simulation that I think could help reduce file sizes in the future. Expansive cellular automata rulesets could possibly be used to compress large amounts of data to something very small, without loss. Aren’t some deterministic in both directions? Can’t they then be run in reverse? If so, that’d be a better-than-brute-force way to figure out the simulation’s end state (expanded data) that would be produced by the simulation’s initial state (compressed data). I want to understand if it could work. Also compressing/decompressing would need crazy computing power and wouldn’t be feasible now. I could be wrong about the reversibility of the expansive type though; I need to do experiments.

May I say, that’s all folks. At least, that’s all the unbridled thoughts I have recorded for the first half of May. Let’s feel safe in assuming the thoughts will continue. Stay tuned for Unbridled Thoughts: 6, surely publishing soon.

<3 Grant