infinull

I received a notification to use Door Dash yesterday, to order food... while I was cooking soup.

This uses ffmpeg under the hood and muxes the file into a .m4a file without transcoding. Basically keeping whatever compression youtube used for the audio (which is some sort of mpeg4 compatible audio, probably depends a little bit)

This still recompressed, but it's the best you can do using youtube as the source.

• uploader (almost certainly, but theoretically you could skip this step if you encoded your video well) compresses audio
• uploader uploads to youtube
• youtube re-compresses the audio again (almost certainly transcoding into a different codec)

No thanks, I think I'll install Linux onto a dead badger

I mean people said this about SponsorBlock, but it turns out you can crowd source the timestamps... If you built an easy enough UI.

I'm 99% certain that's exactly what it is.

worldcat lists the institutions with this book https://search.worldcat.org/title/1048005393, you'll probably have to show up in person. You may be able to do an inter-library loan if your local library doesn't have the book, you could get it from a sister library. However, the only public library that lists having the book is the Austin Public library in Austin, Texas.

If you want the instances to sync, you just need to sync the directory. (I currently use nextcloud sync for this, but in the past I used synching and before that btsync)

If they dont try to modify the files at the same time (with sync delay) there won't be any issues. If they do grow out of sync, you can fix pretty easily with db repair.

Alison also made a compilation of all the tik-toks of people reacting to the original: https://www.youtube.com/watch?v=TNa-8tFoUxs which I think adds a fair amount of humor (but I'm sure some people will feel this is beeting a dead horse)

Those aren't HTML exports though? Those are direct links to google docs.

I haven't used Chrome's reading list feature, since I don't use chrome, but they are competing "read it later" product, so should function vaguely similarly.

Unfortunately I think your impression of Pocket is basically correct, it hasn't received any meaningful updates since Mozilla bought it, and is very underwhelming product, it's still baffling for example that Pocket (at least as a browser plugin) uses a different (and generally worse) reading view from the Firefox "reading mode."

That being said I haven't found any "read it later" products I've actually liked using... (I switched to Wallabag after I quit using Pocket, but have used readability, instapaper, and the reading list feature of "The Old Reader"), so I just quit using the product category entirely, my replacement is "send to device" feature of Firefox so I can find articles on one device and send them to another to either view on a bigger screen, or a mobile screen. (I have a desktop, laptop, tablet and a phone... so this is very useful)

I mean... it's not a network technically. it's a broadcast station (though the stations themselves are networks)

Right, only one side of the connection needs an open port (and most clients will let that be either seed or leech side)... this is why having an open port on your end is useful if you're downloading, since you can download from seeders that don't have an open port.

I do this, but with keepass (keepass on all devices and then sync with nextcloud). Saved my butt a few times, I can go into the file history and pull an old version of the keepass db out of it, and then keepass has a merge feature, so I can pull the old file out, and merge with current to find missing records.

Anyway... backups good.

Something like that is probably technically possible, but you'd need to do a bunch of work.

Plex Plugins can't provide media sources anymore, so you need to do the trick plex_debrid is doing where you add stub sources to the plex server library and serve the files from a virtual filesystem.

You might be able to re-use the plex_debrid code but use youtube-dl instead of rclone

Sorry if this sounds combative, but I just don't think I'm understanding what's going on, I can't figure out how this could possibly work.

How does that even work though? Like... the exported doc is just a web page, it doesn't have any google watermarks (except the now invisible ones) marking it as a google web page.

If it's hosted on an external domain... it doesn't have the google domain in the URL bar either...

Like how is the scam victim fooled vs a normal web page with the same information... How is a google docs HTML export visually different from a LibreOffice or Microsoft Office HTML export in a way that tricks the scam victim into thinking it's legitimately from Google and therefore laundering the scammers reputation through Google. Like I know scam victims are generally distracted or otherwise not thinking clearly (or just dumb), but how does this work?

Besides the default font basically any Word Processor HTML export looks the same to a layman, it's plain black text on a white background with 1in margins. If scam victims trust plain white backgrounds and simple formatting there's a ton of ways to achieve that effect that bypass Google.

Howdy