Nurgle

I got an update for the YT Plugin yesterday evening. Also; I’m in Europe, maybe it’s a regional change?

I just tested it again, it still works here.

GrayJay still seems to work

Thank you, I wasn’t aware

I wear glasses as anyways and I’d love a heads up display. Augmented reality where I could basically spawn full sized displays anywhere for work would be nice too. I’d probably need a device to control it somehow too.

However, I wouldn’t want them from Google, Apple, Meta or any of the other large corporations. Not coupled with their walled gardens, their subscriptions, their EULAs and terms and conditions and “updates” I didn’t ask for.

I just want the hardware and a driver for Linux. Connect the glasses via WiFi to my own computer and run the applications on this computer. If I want to use the glasses outside of my home, I would set up a VPN and use my phone to create a WiFi hotspot.

However, I’m pretty sure nobody is going to build it like that, so I’ll never have smart glasses. Which is fine.

I was hoping that it’s some kind of piratey alternative to Netflix.

I’ve stopped pirating more than a decade ago but streaming and “buying” content is getting worse and worse.

Now I’m just downloading older movies from YouTube and similar services or look it up on archive.org.

Pirating was quite simple in the olden donkey / emule / kazaar etc times. BitTorrent was simple at the beginning as well. Now it seems to be done via invite only forums and special rules and quotas.

IRC seemed complicated as well and I can’t bring myself to ever look into usenet because to me it was my first “social media” and it breaks my heart to see what it turned into.

Also it’s super weird because it’s not meant for this purpose. It’s like using twitter to share base64 encoded warez. It kinda works but it’s not meant for it.

I don’t understand it. I know Netflix and pirates but I see no connection. What is TF?

I really recommend watching this introduction by Andrej Karpathy https://www.youtube.com/watch?v=7xTGNNLPyMI

One part that really stuck with me is that the data in the model is more like a fading memory but the stuff in the context window is more like the working memory. Since I learned that I tend to put as much information as possible into the context window before asking questions about it. This improved the results drastically and reduced hallucinations.

With enough RAM and ideally a good GPU you can run smaller models (~8B Parameters) locally on your own device.