• 0 Posts
  • 67 Comments
Joined 1 year ago
cake
Cake day: July 7th, 2023

help-circle


  • You don’t need to run it on a pi. In fact, I’d actually argue against it; A pi will be underpowered if you’re ever needing to transcode anything. Transcoding is what Plex/Jellyfin does if your watching device can’t natively play the video. Maybe you have a 4k video, but you’re playing it on a 1080p screen. That video will need to be transcoded from 4k into 1080p for the screen to be able to display it. Or maybe the file is encoded using ACC (a fairly recent encoding method) which isn’t widely supported by older devices. This often happens with things like smart TVs (which often don’t support modern encoding and need to be transcoded even if the resolution is correct.)

    Basically, if you’re 100% positive that every device you’re watching it on will never need transcoding, then a pi is acceptable. But for anything else, I’d recommend a small PC instead. You can even use an old PC if you have one laying around.

    Or if you want to use a new machine, maybe something like an HP Elitedesk. They’re basically what you see in every single cubicle in every single office building. They’re extremely popular in corporate settings, which means there are a ton of used/refurbished systems available for cheap, because IT destroys the drives and sends the rest to refurb when they upgrade their fleet of PCs. So for the refurb you’re basically just paying the cost of an SSD they added in (to replace the one IT pulled out), plus whatever labor is associated with dusting it out and checking the connections to make sure they all work. You can pick up a modern one for like $250 on Amazon (or your preferred electronics store).

    Worth noting that the elitedesk generations are marked by a G-number, so google the model (like an EliteDesk G9, G7, etc) to see what kind of processor it has; Avoid anything with an intel 13th or 14th generation CPU, (they have major reliability issues) and check with Plex/Jellyfin’s CPU requirements list to see if it supports hardware accelerated transcoding. For Intel chips, look for QuickSync support.

    For storage, I’d recommend running a NAS with however many hard drives you can afford, and one that has extra ports for future expandability. Some NAS systems support Plex and/or Jellyfin directly, but the requirements for full support are tricky and you’ll almost always have better luck just running a dedicated PC for Plex. Then for playing, one of two things will happen. Either the device is capable of directly playing the file, or it will need to be transcoded. If it’s directly playing, the plex server basically just points the player to the NAS, and the player handles the rest. If it’s transcoding, the PC will access the NAS, then stream it to the player.

    As for deciding on Plex vs Jellyfin, that’s really a matter of personal preference. If you’re using Plex, I’d highly suggest a PlexPass sub/lifetime purchase; Wait until Black Friday, because they historically do a (~25% off) discount on their lifetime pass. Plex is definitely easier to set up, especially if you plan on streaming outside of your LAN.

    Jellyfin currently struggles from a lack of native app support; Lots of smart TVs don’t have a native Jellyfin app, for instance. But some people have issues and complaints (many of them justified!) with Plex, so if the FOSS sounds appealing, then consider Jellyfin instead. Jellyfin is also rapidly being developed, and many people expect it to have feature parity with Plex within a few years.

    And if you’re having trouble deciding, you can actually set up both (they can run in tandem on the same machine) and then see which one you prefer.

    And the nice part about using a mini PC is that you can also use it for more than just Plex/Jellyfin. I have the *arr suite running on mine, alongside a Factorio server, a Palworld server, and a few other things.



  • This looks like it was a timing analysis attack. Basically, they’re trying to figure out which user did something specific. They match the timing of the event with the traffic from the user, and now they know which user did the thing.

    It can be fuzzed by streaming something at the same time, because now your traffic is way harder to time analyze when you have a semi-constant stream of data running. But streaming something over Tor is an exercise in patience, (and it’s not something the typical user will just always have running in the background) so timing analysis attacks are gaining popularity.












  • Well yeah. The goal is to humanize the person they just killed, to make the players potentially regret their murderhobo ways for a brief moment. And one of the fastest ways to make a character (at least shallowly) altruistic is to have them pet the dog. Do something kind for something/someone innocent. It’s often used to show that an antagonist isn’t entirely evil, and is acting against the party due to a specific goal (rather than simply being evil for evil’s sake).

    It’s the inverse of the “kick the dog” trope, where a character does something obviously evil for no narrative purpose other than proving that they are evil.





  • Their argument towards fair use wasn’t ignored. It was inapplicable.

    It’s ridiculous to assume that an organization whose main purpose is data archival would knowingly and blatantly ignore copyright law

    Except that’s exactly what they did. They knowingly and blatantly violated copyright law. They had a system in place to ensure fair use compliance. They intentionally disabled that system, in violation of fair use, to allow unlimited free downloads of the books they had archived.

    IA’s entire argument was basically “but we’re a library” and totally missed the part where even public libraries need to comply with copyright law. Even with ebooks, they can’t simply distribute an unlimited number of copies; They have licensing agreements in place, for a specific number of specific ebooks to be checked out at any one time. And they have to use time-locked DRM to ensure compliance, by automatically revoking users’ reading ability when their check-out time is up. IA did precisely none of that.