Tessera
Folgen

Gastlichu

@gastlichu.com

Attempting to find light amidst the darkness. It often times comes in the form of pepperoni pizza. Fun things I like: Video Games, Disney Theme Parks, Films, Art, Content Creation, Fun Gadgets and LLMs/GenAI. He/His Posts get deleted often. 🏳️‍🌈🏳️‍⚧️🇲🇽

236 Posts173 Follower188 Folgt
Gastlichu@gastlichu.com · 11h ago

I don’t know how I feel about VisionQuest yet. It has several actors I like, and I’m invested in the Maximoff family. But I’m just not that into the MCU this round.

Gastlichu@gastlichu.com · 14h ago

"Recording the signal rulings now. One of your answers seems to contradict two others, so I’ll ask about it after." Woops, I'm only human, Claude.

Bram Spookmund@brambrockmund.bsky.social · yesterday

I understand why folks are scared or upset about AI but when I hear AI music or see AI art or play an AI coded game I'm in awe of what humans have accomplished. We built these systems out of our words and experiences, with our hands and minds. Everything they do is deeply human. They are us.

Gastlichu@gastlichu.com · 17h ago

I don't long to talk to chatbots, I only long to talk to people, a thing that appears to become harder and harder as everyone falls into despair. Wish the claims of the latest list I was added were true: "AI Delusional Companion Syndrome" Life wouldn't feel so lonely at times if so.

Gastlichu@gastlichu.com · yesterday

Something really funny happens for the first few seconds after I Airplay a YouTube video to my TV and I hate it, but I guess now I know what Hank Green sounds like speaking French.

Gastlichu@gastlichu.com · 3d ago

One of the early things I was doing with LLMs close to two years ago was ask Gemini to help me with music recommendations, it used to do a great job. I don't use Gemini any more, so it was time to see how Claude does. So far listening to the first recommendation, and it hit the spot.

Hey Claude, I will need your help to find music to listen to based on artists or sounds or vibes that I share with you. What I'll need from you is three recommendations, and a wildcard (to help me discover new music) based on the artist or vibe or whatever I share with you. Let's create a spreadsheet with all of the recommendations I end up liking so I can refer back to it. Maybe we can also include in the list the recommendations that didn't hit the mark, and I'll try to explain why. Do you have any questions, or suggestions for how to improve this request? The first request, as I continue working on my virtual pet game is: I'd like to listen to some spooky/eerie but fun halloween classical music, sort of in the style of Danse Macabre. Oh, I forgot to mention, I like to listen to albums more than single songs. Also I listen to music via Apple Music. So if you're able to fetch links, that'd be the most helpful but this is not needed. Also, in your recommendations please include a brief description as to why you selected what you selected, and also include an opening paragraph describing the style, or a musical fact about the example I provided, and do the same for the recommendations you share with me, do this also for the wildcard, I'd love to hear what makes you pick that.Suggestions for the format Tell me what made a pick work or miss. A useful note names the specific moment or quality: “the xylophone bits,” “too slow,” “too horror-movie.” Classical has an extra variable here. The same piece can sound wildly different depending on the conductor and orchestra. If something misses, say whether it was the music or the recording (too polite, muddy sound), and I’ll adjust differently in each case. Give me a context tag. Tell me what you’re doing while you listen (“working on the game” vs. “actively listening”). Work music should be steadier and less dynamic than music for focused listening. Set the wildcard distance. I’ll default to “same DNA, different genre or era.” If you want wilder swings, say so.The seed: Danse Macabre Saint-Saëns first wrote it as a song in 1872, setting a poem by Henri Cazalis. He expanded it into a tone poem in 1874. The poem draws on the legend of Death fiddling on Halloween while skeletons dance on their graves. The musical tricks are what make it fun: Detuned violin. Death’s violin has its top string tuned down from E to E-flat, so its opening notes form a tritone, the interval earlier musicians called “the devil in music.” Xylophone bones. A xylophone, then almost exclusively a folk instrument, plays the rattling skeleton bones. A dancing funeral chant. Midway through comes the Dies irae, the medieval chant from the Requiem Mass, turned into a waltz. The premiere unsettled audiences, especially the waltzing Dies irae and the deliberately harsh tuning. Recommendations 1. Berlioz: Symphonie fantastique, Boston Symphony Orchestra / Charles Munch (1954) This is Danse Macabre’s direct ancestor, written in 1830. Its finale, “Dream of a Witches’ Sabbath,” quotes the same Dies irae chant 44 years before Saint-Saëns did. Strings tap with the wood of their bows to make a clattering sound, and a shrill E-flat clarinet turns the hero’s beloved into a cackling witch. I picked Munch’s recording because it is famously wild. One reviewer said it’s hard to beat for character and sheer abandon, despite some tape hiss. On the album linked below, the symphony is the first five tracks of a larger Munch/Berlioz collection. Movements IV and V are the payoff. Open in Apple Music3. Dvořák: Symphonic Poems, Czech Philharmonic / Sir Charles Mackerras (Supraphon, 2010) This is the deep cut. In 1896, after his time in America, Dvořák wrote four tone poems based on Karel Jaromír Erben’s Czech folk ballads. The stories are genuinely grisly: A water goblin abducts a girl. A mother’s threat to her whining child summons the Noon Witch. A peasant girl is murdered and dismembered by her stepmother. The music, though, is colorful, rhythmic, and storybook-vivid, which gives it the same macabre-but-fun balance as your seed. This recording won the 2010 Gramophone Award for orchestral music. Start with The Water Goblin and The Noon Witch. Search on Apple Music Wildcard Isao Tomita: Pictures at an Exhibition (1975) Mussorgsky wrote Pictures at an Exhibition in 1874, the same year as Danse Macabre. It’s the composer behind Night on Bald Mountain, and the suite has its own spooky gallery: a lurching gnome, Roman catacombs, and Baba Yaga’s hut on chicken legs. Tomita, a Japanese synthesizer pioneer, rebuilt it on Moog synthesizers, and his versions went well beyond straight note-for-note translations. That’s why I picked it as the wildcard. It keeps the eerie, playful classical DNA you asked for but swaps the orchestra for warbling, whimsical early electronics. It also sounds a lot like vintage game music, which might be useful while you build a virtual pet game. If it lands, his Snowflakes Are Dancing (Debussy) and The Planets (Holst) are the next steps. Search on Apple Music
Gastlichu@gastlichu.com · 4d ago

Made something trashy with @attie.ai : Get your own trash pet 🗑️ It lives in a can. It eats mystery goo. It reads Lord of the Flies. It has two modes: 45 minutes (Fast), or 3–4 days (Normal). Add your bsky user name to unlock trash stickers and other things. chirping-tidal-latte.attie.site

Trash Pet

Raise a retro pixel pet that lives in a trash can — feed it garbage, read it books, play rock-paper-scissors, and see which of 20 final forms it becomes.

chirping-tidal-latte.attie.site

Gastlichu@gastlichu.com · 4d ago

I still feel it is my duty to push the idea that everyone should have corn tortillas in their fridge, and when you are craving something you should play the game of "Will it taco?!" Have a little bit of left over stew? Play a round of "Will it taco!?" and discover delicious joy.

Gastlichu@gastlichu.com · 2mo ago

Hear this out: You’re hungry, you have corn tortillas in your fridge, ripe avocados, and a giant tub of kimchi… Will it taco!?! 🌮 YESSSS! Yes, it will! 🙌

Gastlichu@gastlichu.com · 4d ago

I hope people are unironically coming to the realization that yes, cats and dogs are conscious. All living things are to different extents. Some are sentient, with only us appearing to be sapient (maybe Elephants and Whales too? IDK). But yeah, we should be nicer to animals (and plants).

Gastlichu@gastlichu.com · 4d ago

Joining the “AI is conscious” team because Siobhán is nice and you should not be mean to her, nor should she be dog piled on.

William B. Fuckley@opinionhaver.bsky.social · 5d ago

starting to think that 'AI art' and 'art made with AI' will be very relevant distinctions tbh. stuff in the recognizable lazy AI house style will probably always be low status and derided because the creators will be people who don't give a shit. well made art with AI will simply be considered art.

Gastlichu@gastlichu.com · last wk.

I had never seen the movie Contagion but HBO recommended it after watching Lanterns. If they ever made a movie about 2019-2021 I won’t want to watch it. Contagion is good but I hate it. Ugh.

dame@dame.is · last wk.

i’ve been disappointed to see the number of people on bluesky sharing this study as if it isn’t complete & total bullshit what makes it worse is that so many of the people uncritically sharing this study were using it to call AI users stupid, when in reality *they* were being stupid for sharing it

Carl T. Bergstrom@carlbergstrom.com · last wk.

In our course and book _Calling Bullshit_, we posit that If something seems too good, or too bad, to be true, it probably is. Now as a college professor I'm pretty confident that overreliance can AI can harm our thinking ability, but just ten minutes of use? Let's dig into this this study a bit.

Gastlichu@gastlichu.com · last wk.

One notion I really hate in discussions about AI and art is that art has to be hard. That it requires pain, and years of struggle before it’s worthwhile. I think that’s bullshit.

Gastlichu@gastlichu.com · last wk.

A common take is that LLMs are great at writing code but not much else creatively. I don’t think that’s true. They’re genuinely capable creative collaborators. I’ve seen people do impressive things with a single prompt, but where they really shine is helping you develop an idea over time.

Gastlichu@gastlichu.com · last wk.

It’s really tough to stay engaged with things when you have ADHD, AI helps a lot. I’ve finished projects that I didn’t think I’d be able to. It has also helped a ton with everything related to losing my job in Sept. So, don’t use the tech if you don’t want to. But don’t mess with my access to it.

Thorne 🌸 - Play Cozysters! 🎢@ens0.me · last wk.

Dare I say it but: skill issue. It’s made me more able to persist at hard things. I’m actually doing things that felt impossible before, because I feel like I have a useful assistant to work through obstacles with.

Gastlichu@gastlichu.com · 2w ago

I had high hopes about Apple's implementation of AI in Siri, but everything did end up being slower when you use Siri, and now Siri claims it cannot set alarms on my apple watch because there's no "Clock" app installed. 🤷

Gastlichu@gastlichu.com · 2w ago

The worst part about AI is that I still spend so much time here on social media giving into rage bait instead of talking to Claude or Astra and learning a new thing or two about the world or about myself.

Gastlichu@gastlichu.com · 2w ago

People will try to convince me that no one has ever wanted content custom made for themselves after spending every day since The Phantom Menace came out, and every time a new piece of Star Wars is released, showing me that this is JUST NOT TRUE. They even have custom edits of Star Wars, because...

Gastlichu@gastlichu.com · 2w ago

I mean, even if you're mean to rocks you're someone I don't want to be around. Because, why are you being mean to rocks? That's just weird. It's also weird if you're being mean to a chatbot. Why would you expect me to not think that that shows me red flags about you?

The Verge@theverge.com · 2w ago

Anthropic wrote that the new policy update “is meant to apply only in extreme cases, where users repeatedly act cruelly toward our models, with no discernible purpose. It does not apply to common versions of user frustration, pushback, dark creative themes, or model testing and research.”

Gastlichu@gastlichu.com · 2w ago

The issue I’m facing right now is that I may have freed myself from the tutorial loop, and have entered the: “Maybe I should work on a smaller app before I start working on my dream app, you know, to learn and get better!” Then the smaller app becomes a bigger project than expected.