You really need to get your backups in order.

Honestly I think the hardest part is identifying it and locating it.

Probably need to start a community around it. People link to stuff. Pay walls would have to be dealt with. In the vast majority of cases that’s not too difficult. No it does start getting less available around video sources. IP restrictions will be a pain in the ass.

Recording the video is messy. YT-DLP could work for some things, But honestly with the level of what’s coming I’d be afraid of leaving fingerprints on anything. Probably throwing a full screen player up in a 1080p window and using OBS on it would be the safest. Could probably get away with using an elgato to capture the HDMI signal up to 4K.

Speech to text models are light enough to run on raspberry pi. They’ll need to be vetted. They’re not highly accurate. Captioning is a great community task.

Organizing an indexing the captions, there’s no shortage of free database software. I probably start with sqlight to keep things portable and fluid. Moving to Maria or Postgres when things get too slow, But then we’re going to have to host it. Anonymous hosting is a completely different ball of wax.

Storing the data would get out of hand quickly. It’s trivial enough to buy a single 20 tb hard drive and store more than we’d need for years. But then hosting it anonymously would be difficult to say the least. Even the markers of these conversations would be traceable enough for us to be located. Paying for a private enough nude to be safe is going to be pricey over time.

I’m sure archive.org would take it, But honestly I wouldn’t put $5 on them surviving a couple of years into the new administration. What they’re doing is to inconvenient to too many corporations with deep pockets.

IPFS would work, well about as well as it works anyway, but that’s the opposite of anonymous.

Edit: come to think of it it would be a hoot to run it on the short video federated platform. Just keep the database somewhere else. Again not anonymous enough for my tastes, But what I had a little bit of fun to the project.

“The statistical likelihood,” continued the autopilot primly, “is that other civilizations will arise. There will one day be lemon-soaked paper napkins. Till then there will be a short delay. Please return to your seat.”

You can pretty much use anything that’s not intel iris with varying levels of work.

The real fight is throwing enough vram at it.

I don’t understand this sentiment

There’s a lot we can do right now, but the most important thing is organizing

Organizing? Resistance? Armed? That’s honestly insane.

You’re going to organize against half the US? Gonna start a civil war with every last (fully armed) enemy in your own backyard?

They could blockade cities from food and shut down any movement in 3 days.

The Civil War worked efficiently because there was a battlefront. This is more of a Republican Soup.

It’s not for now, it’s for our kids, or their kids, assuming we get some morals in a couple of generations.

Oh dems have. But you have to have control of the state to do that. Hogan (R governor) tried his damnest to unwrap central Maryland from Western Maryland.

I’m not sure that we do. Not in our lifetimes anyway.

With a functional justice department we’d have a chance. There’s nothing to stop them from tweaking the electoral lines. There’s nothing to stop them from not certifying an election. We’re about to have the scotus filled with young like-minded Republicans. We’re about to have every federal judge biased for them.

Even having both sides of Congress the best thing we could do would be to status quo because every time a veto is overturned the scotus could just stamp it down as unconstitutional.

The president has God King status, he can have opponents jail for executed.

The thing is even if none of these things were in play, The popular vote just voted for a dictatorship. He was utterly and absolutely clear and anyone that says he was joking around doesn’t actually believe that they’re just too embarrass socially to announce that they themselves are racist/fascist/misogynist. There is nothing here to win back. We’re better than 50% rotten to the core and those people aren’t going away.

Even this election wasn’t right versus left it’s right versus more right. If you put a true left candidate in they’re just going to get murdered. (That may or may not be literal)

I don’t think it would have mattered starting a few months ago he was very open with everything that he did and said. But media like that has a tendency to get buried over time and I want my kids and their kids to know exactly what the f*** happened

For a while now I’ve wanted to start a project of shit politicians say. Video as much as possible. Every shitty immoral reprehensible thing that they come out and say and do logged into a searchable archive. Make it super easy for regular people to stitch together anti-political ads using the politicians own words with context.

Somebody needs to fix IPFS so we can archive all that shit and share in hosting a chunk.

They’ve already said that they’re perfectly fine with the economy taking a massive hit. Nobody about to come into power gives a damn about the economy, They can simply siphon money out of whatever systems they want. Drive up the deficit and pay themselves and their corporations out of the government coffers.

Even if they gave a damn about the economy, what are you going to push for collectively? The department of Justice is about to be fully corrupted. The separation of church and state is being removed. There’s nothing that they’re going to be able to give to satisfy the needs of the public short of a complete replacement of government which they are not going to just roll over on.

I don’t think you can general strike out of a dictatorship.

Don’t go could turkey, you’ll fail. Lizard brain is gonna respond with or without monkey brains help.

As much as you can, start with replacing fuck with frack or fine or fudge, or just make it funny. Stopping the cadence is harder than changing the word.

LOL, he’s corny and tries to be a little too flashy. But at least he gets to the point quickly and tends to cover all the basics without shooting off into a useless tangent. (sans the coffee)

This is the first time I think I’ve ever mentioned him as reference material, but it was mostly because he covered the guardrail concept that no one else I’ve watched talked about.

We’re a long way from trusting it to do something critical without intervention.

AI would be good at looking at an X-ray after a doctor and pointing out anomalies. But it would be bad to have it tell the doctor that everything looks fine.

I send a lot of stuff through co-pilot and I’ve never gotten anything remotely sexual with it set to precise mode.

Do keep in mind that running your personal thoughts, questions, and ideas through an AI will probably get your personal thoughts, questions, and ideas fed into their marketing system.

If you have a half decent Nvidia card and some spare time to wait for responses to come back self-hosting Ollama isn’t that difficult.

You could actually set your own guardrails up. in the config you could set up some post instructions for every prompt specifically instructing it not to get emotional.

The YouTuber Network Chuck recently did a decent video covering the set up and guardrail security.

It tells me what document in the collection it used, But it doesn’t give me too much in the way of context or anything about the exact location in the document. It will usually give me some wording if I’m missing it and I can go to the document and search for that wording.

I’m just one person searching a handful of documents so the sample size is pretty small for repeatability, so far, if it says it’s in there, it’s in there. It definitely misses things though, I’m still early in the process. I need to try some different models and perhaps clean up the data a little bit for some of the stuff.

Using the documentation as source data It doesn’t seem to hallucinate or insist things are wrong, it’s more likely to say I don’t see any information about that when the data is clearly in the data set somewhere.

YW on the responses I’m having fun with it even if it’s taking forever to get it to dial in and be truly useful.

Trident VGA?

I got a 3DFX voodoo as soon as they came out. GL quake was mind-blowing.

I bought a Riva TNT

Then a GeForce 2

Then a Radeon 9000

Then for a bunch of years I just moved into laptop after laptop with discrete GPUs.

Now I still have a 1080 and a 2070 doing a little bit of light AI work and video transcoding for me. But I’m still relying on crappy laptop GPUs for all my gaming. They’re good enough.

I have two projects for it right now. The first is shoving my labyrinth of HOA documents into it so I can answer quick questions about the HOA docs or at least find the right answer more effectively.

The second is for work, I shoved a couple months of slack, some Google docs, some PDFs all about our production product. Next I’m going to start shoving some of GitHub in there. It would be kind of nice to have something that I could ask where is the shorting algorithm and how does it work and it could give me back where the source code is in any documentation related to it.

The HOA docs I could feed into GPT, I’m still a little apprehensive to handover all of our production code to a public AI though.

I’ve got it running on a 2070 super and I’ve got another instance running on a fairly new ARC. It’s not fast, But it’s also not miserable. I’m running on the medium sized models I only have so much VRAM to deal with. It’s kind of like trying to read the output off a dot matrix printer.

The natural language aspect is better than trying to shove it into a conventional search engine, say I don’t know what a particular function is called or some aspect or what the subcompany my HOA uses to review architectural requests. Especially for the work stuff when there’s so many different types of documents lying around. I still need to try some different models though my current model is a little dumb about context. I’m also having a little trouble with technical documentation that doesn’t have a lot of English fluff. It’s like I need it to digest a dictionary to go along with the documents.

I got olama and WebUI working privately / locally and I’m able to insert documents into it with persistence and query them.