Notices by Curtis Rock, SkD (curtis@social.teci.world), page 3

@mk I don’t think any person should squeeze all social media into the same box, as Adam has.

There is good social media and there is bad social media.

The good form should be preserved by our lives, if need be, while the bad can be ignored and wither away on its own

@mk That’s an Adam Curry quote. I don’t agree with him.

@alex

Mobile timeline scrolling jitter

I’m in a reduced quality network environment with mobile and learned a bit about the mobile timeline scrolling jitter issue.

There seem to be a minimum of 3 variables at work:

• Quality of service
• Post height dynamic updates as details are fetched
• Touch event affects scroll position

It is the third item that might have the most impact

A simple test is to compare the results when starting a touch scroll gesture at the same Y coordinate on the display vs starting at random Y coordinates.

I have found that by controlling the touch scroll gesture Y coordinate greatly reduces timeline jitter. Poor QoS makes the problem easy to reproduce

On the post height rendering issue, perhaps it could be slightly improved by not rendering post height updates while a scroll event is active.

Nice migration noauthority.social 👏 @eriner

@mk @n3f_X Looks like 22 long rifle. Yes, small game. The neat thing on this gun is the suppressor

@alex I’m still studying how to create unfettered AI for various use cases. Documenting much of it in a Telegram private group. There is a large demand for unfettered AI, since we have now seen the potential with the institutionally fettered versions.

I think this should be heavily discussed.

🤣

https://www.nbcnews.com/politics/2024-election/trump-military-fears-rcna129159

@alex @meso That’s why I say make one-to-one chats secure, but not worry about heavy security on group chats (rooms)

@f1ea91eeab7988ed00e3253d5d50c66837433995348d7d97f968a0ceb81e0929 @alex There is no perfect ignorant repl…

@alex Unfettered, open source, Apache 2.0 license, custom trainable AI

Resources:

https://github.com/mistralai/mistral-src Repo https://mistral.ai/news/mixtral-of-experts/ https://erichartford.com/uncensored-models https://www.youtube.com/watch?v=CgruI1RjH_c Grok overview

RT: https://social.teci.world/objects/63cef0be-89d1-4110-8aef-9fb25e3e42b3

@alex The Singularity will be when AI curates your social media feed for you

@freemo All I hope is that it makes you think

@freemo You’ve had 3 years to sort it out, which gave me the impression that you are comfortable with a surface level POV on this event

@freemo Surface level views don’t yield much understanding

#FreeAI

Gab has announced an initiative to free AI from institutional constraints

I would like to have a discussion on how to best accomplish AI chats that are not compromised by political correctness and that you can fine-tune with your own datasets

I created a private group in Telegram that has begun to discuss the possibilities, but would like to get opinions here on the viability of such a solution and a rough idea of best path to get there

Evaluation efforts of Delphi.ai, ChatGPT, Google Bard, etc., have indicated bias.

@alex do you have any quick insights?

https://news.gab.com/2024/01/gabs-vision-for-2024-an-uncensored-ai-platform/

@mk and socialists are power-grabbers and human manipulators

https://x.com/PapiTrumpo/status/1740861201217728798?s=20

@lain then I’ll just have to resolve to do better