m 🍥 (@hontologies) header
m 🍥 (@hontologies) profile picture

we are propelled by disaster / we are moving swiftly

Joined February 2024
Photo by m 🍥 (@hontologies)Photo by m 🍥 (@hontologies)Photo by m 🍥 (@hontologies)Photo by m 🍥 (@hontologies)Photo by m 🍥 (@hontologies)Photo by m 🍥 (@hontologies)Photo by m 🍥 (@hontologies)Photo by m 🍥 (@hontologies)Photo by m 🍥 (@hontologies)Photo by m 🍥 (@hontologies)
10
1,877
yeah, it's like that
SNL 'wells for boys' sketch
1
1
5
75
i started seeing a guy for group therapy who asked me during the intake what my genitals were and what kind of sex i have with my girlfriend and it's been a good time so far
Replying to @bimbotheory
Miraculously found a guy that my insurance will cover who practices actual psychoanalysis who has “no experience with those communities” when I asked about his understanding of queerness/transness—perfect, who cares! Help me figure out something about myself I don’t already know.
1
3
80
so far a woman has expressed concern about her son going to a gay furry convention and staying in a room with his friend and said friend's boyfriend, who she's worried will rope him into a gay threesome it's great
2
48
it was really sweet that my coworker did this at work happy hour but i was stressed as hellllll
when ur in a crowded ladies’ room with ur cis friend and she starts trying to have a conversation it’s like omg chill r u trying to get me killed rn
4
78
maybe not the most groundbreaking work but i've been having opus 5.5 generate conceptual artwork based off of stuff i have on-disk each day. these are redactions of my journals (at least, the ones i handwrote via tablet/stylus)
1
1
70
unexpectedly poignant one generated by grabbing the things that were deleted between versions of a note. apparently it was just the word "changes"
1
21
llms are kinda well-suited to generating this kind of thing. systems are much simpler than coding ones, llms are good at research, and it's possible to carry work via an autistic commitment to the process, which a computer really is just purpose built to do
1
13
fascinating that all of the organizational change i'm seeing around llms is just companies prioritizing stuff that was still really important in the pre-agentic ai era but wasn't legibly so
1
2
78
like finally there's the kick in the pants companies need to surface business logic and product decisions in a legible way, generally fix how institutional knowledge works, and standardize architectures. that shit works for human devs too
1
1
36
and many of us have begged for it in the past, but without a nonhuman tool to run against it, the institutional will for it never existed- not legible enough of an incentive!
1
21
pangram is bad, but my ai generated text? very good. because ai is a good tool. how good? so good that you should never get to say i used it. in fact, you're wrong if you say i did, and i will never allow you to verify in any way. why? because your thing is bad and mine is good
2
55
at a company onsite and the team just wordlessly split between east asian and indian lmao
8
78
tbh every time a waymo-involved-crash makes the news i end up updating positively towards waymo because it is 100% some psychopath in a conventional car flooring it into a stationary waymo
Surveillance video shows the crash that involved two Waymos and two cars at 20th St and Van Buren. @PhoenixPolice said 3 people were taken to the hospital: 2 had non-life threatening injuries and one was transported with life-threatening injuries. @abc15
Readers added context they thought people might want to know
Surveillance video shows a human-driven SUV at high speed colliding with the Waymos. Police cite speed as a factor in the four-vehicle crash; the autonomous vehicles were struck. abc15.com/news/local-new… azfamily.com/2026/11/16/pol…
4
58
honestly that's fair because you at least had to have enough comprehension to paraphrase the thing. we get info from a wide variety of sources and llms are simply one of them at this point
The *truly* depressing thing is that there is an absolutely perfect way to trick any AI detector: Just have your text AI generated, read it, and paraphrase it from memory. But people are too lazy to do *even that*.
1
4
60
comprehend the output b4 you show it to another person or you're just foisting off your work on me.
1
11
started making my own shortform slop edits of romance of the 3 kingdoms 1994. claude is actually a solid editor even on a shitty $200 thinkcenter with an i5. no gui + ffmpeg kinda bangin for my purposes
1
3
63
it's been really fun having hermes agent on a dedicated box. it's like having a dedicated sre for selfhosted services
2
57
at some point you loop around to having a piece of paper and it saves your life
just one more tracker bro. just one more just one place to write down all your work I swear. I promise this is the one bro this is the sheet it’s going to be so low-overhead
5
77
Theater is simply not a current artform, it’s a historic one. To write a “play” is to purposefully engage in anachronism. Current artforms include sigma edits, moodboards, nanofiction, DM dialogues, mass shootings, AI fruit commedia dell’arte, video essays, thirst traps, etc.
I see a lot of theater in New York and have never seen a show that properly grapples with the cultural upheavals of recent years as, say, The Crucible did with McCarthyism or Angels In America did with AIDS and the Reagan era. The artists of our time seem asleep at the wheel by comparison.
65
310
35
4,363
68,920
in 2014-2015ish i went to a reading where tao lin performed this live i think
I regret to inform you all that we have found our first long false positive from 2006 it's a poem where the same 6 words are repeated four thousand times
10
219