Link to the actual publication

I like this idea in concept. However in most cases failure is mostly due to power loss. So unless these seprate systems have different power sources it might not protect from catastrophy that much.

As some one who does aquaponics at home. Here are some things ive learned which you might consider.

• There is a minimum physical footprint that these systems use, especially if you are trying to conserve power. So it might make more sense to max out those systems in that minimum footprint.
• Different fish can have very different temperature and pH requirements.
• The pH and temp that the fish like isn’t always what a plant wants.
• some fish have much greater ranges of pH and temp requirements than others.
• if indoors, it’s easy for a pest to establish. Like aphids will trive without ladybugs or green lacewings.
• Vegetables grown in aquaponic systems simply do not taste as good as vegetables grown in soil. Often they are lacking flavor.
• something about the water in an aquaponic system is excellent for propagation from plant cuttings.

For these reasons, I use aquaponics more as a backup and complimentary grow system to my no till regenative garden. Having an indoor system is great in the winter if you want leafy greens or to proprpgate trees for the next season.

I second magic earth for driving. The only reason I haven’t entirely removed Google maps from my phone yet is because magic earth doesn’t have data on business hours. This is the last killer feature I’m waiting for.

In the meantime I’m routing for any open source map that can do live traffic and business info.

Repositories can and should be mirrored to multiple places.

The main repo should probably be isolated in some way. Like maybe a self-hosted Gitea instance which then mirrors all commits to a public facing repo like github.

I do this with all my projects that I care about since I asked myself what was the cost of using github if it wasn’t money.

The documentation should also be stored separately from code in its own repo so that it can be hosted as a static website easily.

And there are so many options for CI/CD that you don’t have to get vendor locked into github actions.

Edit: I haven’t tested mirroring issues. But it should be possible.

After a man date, I like to do a man touch and man mount.

I’m pretty happy with total launcher in combination with app search since I didn’t like totals app drawer.

Edit: I just found OpenLauncher and I’m going to test that out.

I’m on EndeavourOS with an Nvidia gpu. I’ve not had to do anything extra for the the version of proton that comes with steam to work besides install the os with the Nvidia proprietary drivers. And then running eos-update --aur --nvidia

I did notice that I got a lot of screen tearing if using Wayland and that more recent versions of proton didn’t work if either Force Composition Pipeline or Force Full Composition Pipeline were enabled; which should have fixed the screentearing so I just use x11 for now.

There are some things I did to make my experience better however. Like installing an proton-ge. Here is a list of what I installed.

nvidia-dkms
nvidia-settings
libva-nvidia-driver # required by vlc to play mkv files with nvidia gpu
nvidia-tweaks # https://github.com/ventureoo/nvidia-tweaks
lib32-nvidia-utils
gamemode
proton-ge-custom-bin
lib32-libudev0-shim # fixes Steam runtime's super old 32 bit version of libnm
lib32-libnm # required if using systemd 253.5-2 or newer

I would also install nvidia-dracut-hook if you are using both Nvidia and dracut. Dracut is the default on recent versions on endeavorOS.

For proton ge, I also added myself to the games group with

sudo usermod $USER -a -G games

I also like to prepend the following to my games launch options in steam

gamemoderun PROTON\_CONFIG=dxr11,dxr PROTON\_ENABLE\_NVAPI=1 PROTON\_HIDE\_NVIDIA\_GPU=0 VK\_ICD\_FILENAMES=/usr/share/vulkan/icd.d/nvidia\_icd.json VKD3D\_CONFIG=dxr11,dxr VKD3D\_DISABLE\_EXTENSIONS=VK\_KHR\_present\_id,VK\_KHR\_present\_wait VKD3D\_FEATURE\_LEVEL=12\_1 VKD3D\_SHADER\_MODEL=6\_6

And I set proton-ge as my default proton version on the steam options.

It should be, but I’ve had too many issues with windows somehow updating and ruining my boot partion/grub.

If I want to run windows, I only run it on a VM with hardware passthrough or on a totally different machine since physically unseating my Linux M.2 from the motherboard is too much hassle and I don’t want to move it to a PCIe card.

I agree. I only use linux. Which includes for gaming. And I game a lot.

I’d argue it’s better to use actual alternatives. Half of the issue with free and open source software is that it’s userbase is too small. If more people used it, it could actually improve in many ways.

Lets take gaming on Linux as an example. The userbase on steam is somewhere around 5%. So there is almost no incentive for developers to make games that run nativly on Linux. Its actually easier to run the games in a compatibility layer then to get a Linux port of a game. And although wine and proton work incredibly well, sometimes even running a game better than on windows; a Linux native version of every game would be ideal. Which will never happen with such a small userbase.

Next you have the terrible business practices of these companies. Even if you use the pirated versions. You are in their ecosystem and their community. You increase their profitability and their stock price simply by continuing the industry standard.

Pirated versions of software like this is excusable if you need it for work or sometihing. But imagine if instead of staying with the status quo, you use and help improve actual free and open source alternatives. Versons of software that don’t steal your data or monetize how you use it by selling your input to others or stealing it for “AI” datasets.

Imagine using free and open source software that gives you feedom because your data stays on your devices, your creations belong to only yourself or who ypu choose to share it with, and you work with others to improve it; even if it’s by just submitting bug reports. Imagine using something like that which you find so altruisticly beneficial that instead of pirating the software that has no respect for you, you donate money to the devs of free and open source software. Yes, I’m a pirate. But I do donate money to the right causes and something that protects my freedom is worth both my time and my money.

You could probably make a new issue in a wishlust repo that uses markdown checkboxes or something similar. Would be good if you already host Gitea or another git sever.

deleted by creator

You probably won’t be able to run an LTS kernel on a brand new PC that just hit the market. But using the most recent kernel for arch or a derivative like endevorOS should work after like a week maximum.

I did have an issue like this on Ubuntu and its what made me actually start distro hopping since it worked fine on fedora and Arch using the latest kernels.

Digging in a bit. It looks a bit different than I thought it did on my end.

I periodically use Kodi to backup my metadata to an NFO file. But I guess I forgot to script that and haven’t actually done that in a while. When Kodi scrapes the metadata it preprends the mpaa value with the string MPAA Rated . I did this to make it more obvious if the rating was scraped correctly since I’ve seen kodi attempt to scrape before but then just record the value as null.

1. Inside Out didn’t have an NFO that Jellyfin was reading from. But Jellyfin is set to scrape its own metadata from all avaliable sources. I just didn’t enable it to write its own NFO file since I want the files that Kodi makes to take precidence and not be overwritten be Jellyfin. So for the case of insde out, it’s pulling the correct rating without any changes. As seen in this desktop screenshot.

2. A Boy and His Dog, on the other hand does have a nfo from Kodi and it does have that MPAA rating on jellyfin with the prepended string as seen in this screenshot and in this bit from the nfo file.

<?xml version="1.0" encoding="UTF-8" standalone="yes" ?>
<movie>
    <title>A Boy and His Dog</title>
    <originaltitle>A Boy and His Dog</originaltitle>
    <ratings>
        <rating name="themoviedb" max="10" default="true">
            <value>6.300000</value>
            <votes>309</votes>
        </rating>
    </ratings>
    <userrating>0</userrating>
    <top250>0</top250>
    <outline>A young man and his telepathic dog wander a post-apocalyptic wasteland.</outline>
    <plot>Set in the year 2024 in post-apocalyptic America, 18-year old Vic and his telepathic dog, Blood, are scavengers in the desolate wilderness ravaged by World War IV, where survivors must battle for food and shelter in the desert-like wasteland. Vic and Blood eke out a meager existence, foraging for food and fighting gangs of cutthroats.</plot>
    <tagline>A rather kinky tale of survival.</tagline>
    <runtime>90</runtime>
    <thumb aspect="poster" preview="https://assets.fanart.tv/preview/movies/4917/movieposter/a-boy-and-his-dog-5e3fe50ad3787.jpg">https://assets.fanart.tv/fanart/movies/4917/movieposter/a-boy-and-his-dog-5e3fe50ad3787.jpg</thumb>
    <mpaa>MPAA Rated R</mpaa>

3. Burning has a nfo file from Kodi. But the MPAA key is null in the NFO file. So there is no prepended string. However jellyfin still scraped the MPAA rating for it’s self and correctly labeled it as seen in this screenshot and nfo.

<?xml version="1.0" encoding="UTF-8" standalone="yes" ?>
<movie>
    <title>Burning</title>
    <originaltitle>버닝</originaltitle>
    <ratings>
        <rating name="themoviedb" max="10" default="true">
            <value>7.400000</value>
            <votes>1440</votes>
        </rating>
    </ratings>
    <userrating>0</userrating>
    <top250>0</top250>
    <outline>Jong-su bumps into a girl who used to live in the same neighborhood, who asks him to look after her cat while she&apos;s on a trip to Africa. When back, she introduces Ben, a mysterious guy she met there, who confesses his secret hobby.</outline>
    <plot>An aspiring writer goes to the airport to pick up a high school friend returning from a trip to Africa but is disheartened to see her with another man.</plot>
    <tagline>The truth is all in your head.</tagline>
    <runtime>148</runtime>
    <thumb aspect="poster" preview="https://image.tmdb.org/t/p/w500/kXiF80o74fE9gf3Utf9moAI7ar0.jpg">https://image.tmdb.org/t/p/original/kXiF80o74fE9gf3Utf9moAI7ar0.jpg</thumb>
    <mpaa></mpaa>

So I guess I have 2 questions.

1. Does the string I put in the rating distort the recommendation system?
2. Does the recommendation system prioritize a local nfo file even if the value for the MPAA key is null when jellyfin actually has the correct rating in its cache?

I’m not 100% sure what the detail page is but here is the page I see when selected.

I think I have Jellyfin mostly at the default setings, with one exception. I did change the name of the key for the rating variable in the metadata to be ‘MPAA rating’ instead of the default which I think was ‘rating’ before since I found it confusing. So if the current recommendation calculation is using that variable then I guess it would be null for all my media.

Also, here is the metadata jellyfin shows.

Inside Out (2015)
Container: mkv
Path: /mnt/ext4/film/film/Movies/Inside Out (2015).mkv
Size: 5022 MB

Subtitle
Title: Italiano Forced - Default - SUBRIP
Language: ita
Codec: SUBRIP
Default: Yes
Forced: Yes
External: No

Video
Title: Video 2160p - 4K - HEVC - HDR
Codec: HEVC
Profile: Main 10
Level: 150
Resolution: 3840x2160
Aspect ratio: 16:9
Interlaced: No
Framerate: 23.976025
Bitrate: 7401 kbps
Bit depth: 10 bit
Video range: HDR
Video range type: HDR10
DV title: DV Profile 8.1 (HDR10)
DV version major: 1
DV version minor: 0
DV profile: 8
DV level: 6
DV rpu preset flag: 1
DV el preset flag: 0
DV bl preset flag: 1
DV bl signal compatibility id: 1
Color space: bt2020nc
Color transfer: smpte2084
Color primaries: bt2020
Pixel format: yuv420p10le
Ref frames: 1

Audio
Title: Italiano AC-3 5.1 - Dolby Digital - Default
Language: ita
Codec: AC3
Layout: 5.1
Channels: 6 ch
Bitrate: 448 kbps
Sample rate: 48000 Hz
Default: Yes
Forced: No
External: No

Audio
Title: English AC-3 5.1 - Dolby Digital
Language: eng
Codec: AC3
Layout: 5.1
Channels: 6 ch
Bitrate: 640 kbps
Sample rate: 48000 Hz
Default: No
Forced: No
External: No

Subtitle
Title: English Forced - SUBRIP
Language: eng
Codec: SUBRIP
Default: No
Forced: No
External: No

Subtitle
Title: Italiano - SUBRIP
Language: ita
Codec: SUBRIP
Default: No
Forced: No
External: No

Subtitle
Title: English - SUBRIP
Language: eng
Codec: SUBRIP
Default: No
Forced: No
External: No

PS: I scrape my metadata from themoviedb.org in Kodi and then export that to an xml file of the same name, Jellyfin seems to find and use that.

Ain’t nothing but a heartache

I recommend EndeavorOS now to everyone that actually wants to learn linux, or people that don’t want to be “fighting” their os.

It works enough to not have to do anything to it besides update, including installing nvidia drivers. And it’s arch based so they can just read the arch wiki if they have questions.

Honestly the only issue ive had with it is one of apps not working on wayland so i just had to switch to x11.

Its a little less noob friendly than manjaro (they had great guis that make it so you never need to open a terminal at all) but i cant recommend manjaro anymore since they dont support the latest version of pacman.

As far as an os that’s close to enterprise servers, if they aren’t contanerizing the workloads and running k8s on a distroless (or atleast minimal) base image then i don’t want to work there anyway.

In TBD, it’s not a “release” until its production ready. The methodology and philosophy doesn’t prevent you from developing multiple feature branches at once or even deploying a work in progress feature branch to a dev environment.

All TBD requires in that case is once the feature branch is production ready, it’s merged to the trunk. You may need to add a feature toggle if there are multiple release like for different architectures. And you also might benefit from using git tags and deploying to production from a git tag instead of the most recent commit on a branch.

Exactly what you need to do is going to depend on the project’s exact needs but TBD is totally possible in that example.

I recommend adding ollama under the artificial intelligence tag.

Running Slackware as your daily is great for learning about linux.

Although I haven’t used it in about 12 years. Is it still like this?