• 0 Posts
  • 9 Comments
Joined 1 year ago
cake
Cake day: August 9th, 2023

help-circle


  • I’m sure you know, but you’re probably going to get a lot of grief for this. I’m deeply suspicious of any new AI tool, especially one that tries to get in between me and my news (looking at you Feedly), and I’m sure I’m not the only one. So if you’re not already, I’d prepare yourself for a lot of strong emotions, and probably not in a good way.

    If you wanted to get ahead of that kind of thing, you might want to explain what kinds of safeties you’re building into it. For example, on your roadmap you say want it to “Generate argument of for and against perspective then summarise the result of the 2 arguments.” This kind of thing in particular is quite risky. Any time you try to introduce value statements into an LLM summary, you’re in the danger zone. Even if you’re just trying to summarize the actual perspective of the piece, you’re basically just begging the LLM to hallucinate. But asking it to summarize hypothetical opposing arguments is just asking for trouble.

    I could go on, but I don’t want to start a pile on. I appreciate when folks try to build cool stuff, you’ve just waded into some choppy waters…



  • As suggested above, I would try Infuse player. I recently switched from a Kodi/Jellyfin setup to an Apple TV/Jellyfin setup and I’m extremely happy with it. Infuse has a free trial, and then you can choose to pay a few different ways (they do have a rather expensive lifetime option, but it might be worth it). The Infuse app has no trouble playing directly from my Jellyfin server, no transcoding, even for full 4K Bluray rips, and yes it even supports Dolby Vision (which the native Jellyfin app struggles with). No hiccups, no issues with multiple audio tracks or subtitles, it’s just buttery smooth direct playback.

    It also has a couple different ways of interacting with your Jellyfin library, so it feels completely seamless to me.



  • That’s a very cool concept. I’d definitely be willing to participate in a platform that has that kind of trust system baked in, as long as it respected my privacy and couldn’t broadcast how much time I spend on specific things etc. Instance owners would also potentially get access to some incredibly personal and lucrative user data, so protections would have to be strict. But I guess there are a lot of ways to get at positive user engagement in a non-invasive way. I think it could solve a lot of current and potential problems. I wish I was confident the majority of users would be into it, but I’m not so sure.



  • I think by default bots should not be allowed anywhere. But if that’s a bridge too far, then their use should have to be regularly justified and explained to communities. Maybe it should even be a rule that their full code has to be released on a regular basis, so users can review it themselves and be sure nothing fishy is going on. I’m specifically thinking of the Media Bias Fact Checker Bot (I know, I harp on it too much). It’s basically a spammer bot at this point, cluttering up our feeds even when it can’t figure out the source, and providing bad and inaccurate information when it can. And mods refuse to answer for it.