Notices by H (u24@c.im)

I’m hoping that some Open Source project will start collecting and distributing _trusted data collections_ as raw material to train LLM’s on. Thats where the value is; not the models themselves. Which are mostly trying to un-shittify the shit they got fed.

To me, the biggest issue with the whole thing is _I do not want something trained on the entirety of crap out there_. We _all_ know that most “content” available is biased, incorrect, racist, ignorant, hot garbage.

Super mad at Apple again tonight as I realize that you cannot have the Music app without Radio, the TV app without trailers for TV+, the or the Podcasts app without an un-hideable screen full of hideous murder-related content or the app store without exploitative IAP gambling-for-kids garbage. I can lock down a device so that a child can't listen to songs with the F-word in them but there is *no* way to say DO NOT SHOW THEM ENCOURAGEMENTS TO GAMBLING ADDICTION.

“…just because you *know* something needs JavaScript, doesn't mean it still does. You can make better websites if you test those assumptions every now and then.”

This is so true, and the reason why it is healthy to check back on vanilla web tech from time to time.

https://www.htmhell.dev/adventcalendar/2023/2/

I absolutely believe, based on knowledge of the subject matter, that *real*, sentient, artificial intelligence is not only possible, but will eventually be created.

I also do not believe for a second that current technology is anywhere *close* to the level needed to do the job, and even if it were, our understanding of cognition is far too low to actually pull it off except by accident.

A common experience in TikTok is for a new audio to appear (many people can use the same audio in their videos), the algorithm detects that you like it (you let it play to the end), so the app feeds you other videos with that same audio. After you see a few you think, wow, this audio is everywhere today!

Then you tap on the audio to see that there are only a few dozen videos with it. It feels like it's a huge viral movement, but it's not.

It takes less repetition than you'd think.

Current Mood: I hate all technology

git rebase: what can go wrong? https://jvns.ca/blog/2023/11/06/rebasing-what-can-go-wrong-/

@oluOnline I really miss when RSS was prominent and well-supported. You could subscribe to sites and their content came to you, generally in a focused/ad-free way too

Social bookmarking was a thing too... people would share their bookmarks and organize little communities around shared interests/interesting people