by simonsarris on 2/18/25, 1:08 AM with 98 comments
by darkwater on 2/18/25, 8:57 AM
by NikkiA on 2/18/25, 10:23 AM
Some will, no doubt, thereafter argue that goths all dress/look the same and thus can't be actually free thinking, but that would require not having a clue about goths or that their expression of being goth tends to look similar because 'the aesthetic' is the defining aspect, but even within that aesthetic there is quite a wide variety of styles and looks that some people will not even consider to be a 'goth'.
by muzani on 2/18/25, 12:06 PM
3 years ago, there was that 3D look. No idea where it came from, but the first time I saw it was from Revolut.
About 10 years ago, we had emojis sprinkled everywhere. Replacing icons, prefixing email titles, being used at bullet points.
Before that, we had vector flat art, which was reflecting responsive design. Sometime after this was the trend of putting videos in the sign up page of an app.
When I was a teenager, websites were all "microsoft blue".
90s design, no longer web, were colorful and had handwriting font. While 80s had a lot of stripes and 3D text.
There's definitely something trendy and edgy about these websites but not too edgy. They're always seasonal. I mean we still see flat design but we almost never anything like Windows Vista and their Aero design. Windows 98 had a certain charm.
by BoxFour on 2/18/25, 12:46 PM
Hmm, I tend to dismiss this idea mainly because I think it'll soon become unnecessary and also belies why most people have "custom setups". My feeling is most people are using LLMs to achieve concrete goals: How to do some basic woodworking, bake bread, get a condensed version of a college course on nuclear physics, or write code to accomplish a task.
Right now specialized setups and finely-tuned models might make sense for bridging the gap between "almost there" and "good enough", but the overall trend seems to be moving toward general-purpose LLMs becoming “good enough” to handle most of these tasks. Over time, the gap between a highly specialized model and a general-purpose one seems to shrink for the level of expertise most people are looking for.
No doubt there may still be some customization for more novel creative applications (and the author even touches on one I expect to see-emulating the dreamlike aesthetic of early generative AI). But novel creativity is a small minority of "My AI Setup" type articles that I see at the moment.
by xrd on 2/18/25, 1:54 PM
I just started reading "The Brief Wondrous Life of Oscar Wao" by Junot Diaz, and it is about a kid who is into subcultures within a strong Dominican community inside a larger NYC community. Interesting parallels.
by card_zero on 2/18/25, 7:39 AM
by partomniscient on 2/18/25, 2:05 PM
I also experience culture shock when I moved to the UK. I left Australia which was still "pre-emo", and arrived in the UK and was confused as to why goths were running around with fat pants, pokemon backpacks and riding skateboards. It took me a little while to realise there was a next generation subculture already established.
by lordgrenville on 2/18/25, 3:25 PM
Sometimes the peer group can be remote. The only emo kid in Baghdad is a unique snowflake relative to other Baghdadis, but is following a fashion code vetted by people on Tumblr.
by empiricus on 2/18/25, 10:25 AM
by Mistletoe on 2/18/25, 7:10 AM
St. Vincent's font on her music video looks exactly like the logos in the article that have all merged to the same gray goo.
by superkuh on 2/18/25, 6:47 AM
by LordDragonfang on 2/18/25, 7:05 PM
For example, the confusing "goth family tree" at the beginning of the article is clearly Dall-E/ChatGPT prompted with just that, with labels edited in after the fact in a way that makes the whole thing nonsense. For the "trad goth" picture further down, it's blantantly obvious it's Dall-E's cluttered style with meaningless lines. It's the exact kind of slop the article complains about, and I'd go so far as to say it's anti-information, because you can't trust any of it. It would be fine if the images were purely for decorative purposes, but for an article that purports to authoritatively describe visual aesthetics, handing that work off to a hallucination-prone AI renders it untrustworthy.
I assume this was the result of a 30 second google, but the author including that as a reference source is so bad that it makes the rest of the article hard to take seriously.
edit:
This particular line jumps out at me:
>Particularly if the AI products assisting us now are successfully trained to not hallucinate anymore
I'm normally one to jump in and defend AI when someone says it's useless because it hallucinates sometimes, because that could not be further from the case. But if you think it doesn't hallucinate at all anymore, then your credibility when writing about AI is severely limited.
by rdtsc on 2/18/25, 7:42 AM
There is a classic joke about that -- they all want to look different, but they all look different in the same way.
by svilen_dobrev on 2/18/25, 12:16 PM
- You're all different!
- WE are all different!
- (lonely voice) i am not
by torcete on 2/18/25, 9:46 AM
by diegoeche on 2/18/25, 4:45 PM
I found super interesting the parallels of cookie cutter contra-culture and the results of a heavily vetted AI slop. How most original/interesting thought has a great insight but is easily dismissed...
> parallel interrogation of a creative process leads to boring outcomes; serial investigation gives you creative outcomes.
by subjectsigma on 2/18/25, 2:00 PM
by mlsu on 2/18/25, 7:49 PM
When people are different, they are wrong relative to each other. When someone is in a far-off subculture (far left, far right, super goth, super avant garde, etc.) they are "wrong" to the majority culture. Compared to the mainstream cultural understanding, anarcho-primitivism is wrong, it's incorrect, it does not fit. You will not find it being used to explain current events in mainstream newspapers or CNN. Because it is wrong.
However, the difference between people and LLM is that the person has a sense of self, an agency, etc, and the LLM is just a tool. When a person is wrong, they have an opinion that differs, it's maybe interesting, there may be some value to glean from talking to them and understanding what reasons they have for being wrong. When an LLM is wrong, it's a tool that is malfunctioning. It's just wrong. By the way, when an employee is wrong, they're a tool that is malfunctioning; that's why weird out of distribution people make bad employees (and why they tend not to become employees).
This is the same critique of hallucination. "Hallucination" is a word we use to describe when the model does not fit reality. But both the model and reality are relative, of course. The same applies for human to human we just assign different value judgements, and rightly so.
by wiether on 2/18/25, 3:07 PM
He also did "mall goths"[1] but the mall is located in London so...
[0]: https://www.youtube.com/watch?v=zPE44MPNf2s [1]: https://www.youtube.com/watch?v=2Q5CyC85sI8
by jl6 on 2/18/25, 8:33 AM
by anal_reactor on 2/18/25, 2:32 PM
by Animats on 2/18/25, 7:24 AM
by JasserInicide on 2/18/25, 3:45 PM
You've probably heard phrases being thrown around online like "conservative is the new punk"; while mostly silly, there is a nugget of truth to them.
by mhh__ on 2/18/25, 11:52 AM
by metalman on 2/18/25, 8:58 AM