Thank you, this is exactly my use-case, along with some live streaming.
However,
How are you running a find command on a warch folder with continuously streaming video? I believe the IPC process has to end (system has to finish writing to a file) for post-processing (reencoding) to take place? That’s why I wanted to do this transparently, since I have no local storage to speak of to maintain a watch folder/cache.
Thanks for the reencoding solution, I’ll take a look at handbrake
Streaming to a socket sounds like a decent idea. I don’t know how and which program I could stream to; is there a way for handbrake to transparently reencode video and send it to a virtual FUSE mount-point? The problem I have is no local storage to keep video.
You’re right, I don’t know why I didn’t consider that. This is going to be a mix of security cameras and live streaming video that I’ll store on the cloud, and the problem is that I have horrible upload speeds along with no local storage for caching
The problem is that I don’t have local storage, and neither do I have very good upload bandwidth. Compression could in theory solve the bandwidth problem and cloud storage costs to an extent, but I completely missed the part about video being already compressed. I’ll take a look at handbrake though, essentially what I want is a transparent layer that will compress video files (or reencode video files since the former seems pointless) without touching local storage and shove it into my virtual FUSE system to upload directly to the Cloud.
Thanks, but the second problem I’m working with (and what I forgot to mention) is that I have no local storage - I would like to write semi-directly to cloud storage. I can probably manage a few GB for caching and that’s it.
The problem is that I don’t have the local storage to maintain a watch folder for continually streaming video. I want to write semi-directly to the Cloud, which is why I’m looking for a transparent reencoding layer. Can handbrake do this?
I’m going to be using an SBC for this, which doesn’t have the capacity for an extra storage drive. Also, I’m planning to move in a couple of months, and I wouldn’t want to deal with storage in the middle of all of this. The cloud isn’t as insanely expensive as I initially thought; B2 is $6/TB, and I hope that with reencoded streams at an OK resolution I wouldn’t go beyond 1.5TB a month (I’ll be deleting stuff with bucket policies, of course).
I’ll take a look at tdarr alongside ffmpeg, thanks!
I was considering the usual BananaPi/OrangePi/Raxda/Pine64 SBCs, are those not enough horsepower? I’d like to stay under $80 for my SBC purchase, and it will be doing double duty with managing some services like DNS and music scrobbling alongside uploading to the cloud
So perplexity can kind of weakly analyze the first few pages of small file size pdfs one at a time, but I’d love to have something that would allow me to upload several hundred research papers and textbooks that could then be analyzed for consensus and contradictions and give me more meaningful search results and summaries...
That would likely be a language model finetuned on said material. The problem is feeding PDFs as a structured data source for the model to ingest. The finetuning can’t happen with random unstructured PDFs
Is Google signaling the end of the open web? That’s some of the concern raised by its new embrace of AI. While most of the fears about AI may be overblown, this one could be legit. But it doesn’t mean that we need to accept it....
Google is one of the biggest obstacles to privacy. This article will explore how and why to de-Google yourself to reclaim and protect your digital privacy.
Basically an advertisement for their services, but since they’re shitting on Google’s ad revenue model I’m all for it.
I will not shit on Google completely myself though, because I do appreciate (other than a few naughty shenanigans they’ve pulled recently) their work on the Android kernel.
For me, I just dislike the feature bloat. I wonder if somebody has a brave fork, since brave has some of the best fingerprinting protection amongst all browsers
Transparent compression layer on Linux?
cross-posted from: lemmy.world/post/15706364...
LLM queries for personal pdf libraries?
So perplexity can kind of weakly analyze the first few pages of small file size pdfs one at a time, but I’d love to have something that would allow me to upload several hundred research papers and textbooks that could then be analyzed for consensus and contradictions and give me more meaningful search results and summaries...
Corbett Report interview on living fiat-free (mentions XMR) (corbettreport.com)
Decentralized Systems Will Be Necessary To Stop Google From Putting The Web Into Managed Decline (www.techdirt.com)
Is Google signaling the end of the open web? That’s some of the concern raised by its new embrace of AI. While most of the fears about AI may be overblown, this one could be legit. But it doesn’t mean that we need to accept it....
How (and why) to de-Google your life and protect your privacy | Proton (proton.me)
Google is one of the biggest obstacles to privacy. This article will explore how and why to de-Google yourself to reclaim and protect your digital privacy.
Halp! Calibrating touchscreen on Panasonic CF-30
cross-posted from: lemmy.blahaj.zone/post/12310804...
EFF "cover your tracks" browser privacy test (lemmy.dbzer0.com)
coveryourtracks.eff.org Wanted to share this awesome tool to test your browser, also please do share your results in the comments.