Skip Navigation

InitialsDiceBearhttps://github.com/dicebear/dicebearhttps://creativecommons.org/publicdomain/zero/1.0/„Initials” (https://github.com/dicebear/dicebear) by „DiceBear”, licensed under „CC0 1.0” (https://creativecommons.org/publicdomain/zero/1.0/)C
Posts
1
Comments
135
Joined
7 mo. ago

  • If you go to YouTube to watch a commercial (here is an example for Orangina https://www.youtube.com/watch?v=wrkUDm9pnIw), should the whole video be blocked? YouTube provide a video hosting service and they add commercials on top of that.

    I find it unreasonable to demand that YouTube essentially remove content from within the video. That would just end back in the EU "protect-the-children"-Chat Control area of the debate. You could have uploaders mark their videos as containing marketing (and perhaps they do) but enforcement would still be difficult given the sheer volume of videos.

  • At least you can turn of the filter.

  • Good luck! Post your success in a new post, might inspire others. :-)

  • The only reason i have 32 is because I salvaged an old work computer. I have 60 running containers and use 7gb RAM (server is mostly idle now). So I have to agree. I just wanted to say that one does not even need ddr4, at least if you are a single user.

  • I have 32gb ddr3, no problems. I keep adding stuff and wonder when I will hit the limit.

  • No changes to the container image, but I had to fiddle with the config file. There is an option (tokenExpirationHours) to set for how long a token is valid, default is 2 hours. I can only find this in the commented example config, setting it to sometime really high like 50 let my friend upload his audiobooks without problems.

    Create a config file and place it somewhere the container can access it. Then add the environment variable FILEBROWSER_CONFIG=/some/path/to/your/config.yaml.

    Here is a snippet from my config:

     yaml
        
    server:
      # ...
    auth:
      tokenExpirationHours: 50
      # ...
    userDefaults:
      fileLoading:
        maxConcurrentUpload: 100 # The inferface only shows up to 10 so I don't know if setting this value higher does anything
        uploadChunkSizeMb: 10
    
      

    tokenExpirationHours is the key to set.

  • This is 100% off topic but I just realised username is ”I R mad lad” and not ”Irma Dlad”.

  • Really? I never tried the original but didn’t have much problems setting up quantum. I had to modify my caddy file somewhat to suite my needs, but just works now.

  • Someone has asked this on stack overflow as well. https://stackoverflow.com/questions/51150276/linux-random-6-0

    From link:It's a Russian roulete in programming. $RANDOM returns a number between 0 and RAND_MAX. If the mod 6 on the returned number equals 0, the command after && (conditional execution) executes and it deletes the root directory, basically destroying everything you have on your disk with no normal way of retrieving it (the OS can't function). If this doesn't happen, the conditional execution after || happens and outputs You live.

  • I use KeePassXC + Strongbox (iOS) and sync it with Syncthing. I have a server always running so if I save a change on my computer, it will always be synced to the server. I use an iPhone 13 so I have to use möbius sync and it is not perfect, I have to manually open the app to make sure it syncs (something iOS background usage something). I also have an old Android phone that also syncs my kbdx file. You could have a RPI Zero (or any other kind computer) as a cheep backup solution just for Syncthing.

  • We need a tag for when the AI use is not known.

    Here is a great project: https://github.com/sapristi/mmuxer

    Mail Muxer is a Python tool that will monitor your Inbox, and filter incomming emails according to the given configuration.

    Is AI used in any shape or form to develop this? No clue, I am not affiliated, but I would really like to post a promo about it and would need to tag it correctly.

  • I asked the same question it should be tagged as [AIP]+[hint] under these rules.

    hint Human acts on the task and the AI surfaces suggestions passively.

  • Makes sense, thanks for taking the time to explain. <3

  • You complained about docker putting everything into a bag and I tried to argue that this approach was better because there is no need to consider library versions conflicting between apps.

  • But in practice idk if they always have actual hydraulic brakes. I’d still think they do, even if there’s also regenerative braking.

    I thought most of the input from the driver (steering, breaking etc) had a layer of computer between the input interface (steering wheel etc) and the actual moving parts.

    I’m thinking at some point it will probably be that you’ll need to hand over driving to a computer whilst in a city or smth.

    That is an interesting solution on how a first push for self driving cars could look like.

    I saw a documentary some years back about self driving trucks in Australia, a company used them instead because it was safer for a computer than a human to drive through the desert.

  • any use would mean a tag

    So when I write code, I regularly ask an LLM how to do things. A few examples from my history is how I could overlap rows when generating PDF files (have a library for this), how I could assign an empty Char and to create a singleton class. A LLM never changes any of my files, I use it as super-search-engine.

    Would that constitute an AI tag?

  • Where should the line be drawn for what is AI coded vs AS assisted (in some manner) and thus tagged?

    XerahS is very openly coded with AI and should be tagged. But what about the recent debate about the use of LLM in cURL?

  • I’d say it is worse trying to find replacement for software not available on Linux and trying to adjust/learn the new workflow when you do.

    On the other hand, a lot of software I wanted was primarily Linux, so that was nice when I switched.

  • Selfhosted @lemmy.world

    How to use Radicale with iOS