Now do Spring Boot. (sorry I don’t use github, because microsoft ai, so can’t create issue)

Every program is “open” source if you know how to use a decompiler.

To answer the headline question:

• because wasm fast

(the rest of the points are debatable)

Grab the sd card, and look at it on your pc. If windows can’t read it (idk anything about windows), do so from another linux (live boot Ubuntu or something on your windows machine).

The journal log should be stored at either one of:

/var/log/journal/<machine-id>/*.journal

/run/log/journal/<machine-id>/*.journal

Post it’s contents in full here via pastebin.com

obsidian for everything and sync to all devices

https://pairdrop.net/

open source, can be self hosted or you can use the official instance.

Personally I have been using KDE connect most of the time when I am at home.

Pairdrop I use more when sharing with other people across the internet.

You can run a gui-less service that recieves and displays push notifications. I’ve programmed something like this before. I know it is technically a kind of client, but it is not an email-client.

ask deepseek to summarize the video and post the text result here

The picture is perfectly ironic.

Even if it was not meant this way it can be interpreted as: an idealist preaching his philosophy to somebody who actually gets real work done.

(there is probably also a joke about the patriarchy and mansplaining in there somewhere)

I code in silence.

But when I listen to music I REEEEAALLYYYYY L I S T E N. Stuff like Tool, Coin Locker Kid, Kaoru Abe - and I find it impossible to concentrate on code while those play, they mentally drag me in.

You forgot a small detail.

Afaik you can not change repo visibility this way (without using the web UI or the GraphQL API). So if the goal is to avoid the web UI you’d have to add a step (which you can read up on in the script I shared).

Same for the repo description (but maybe there is a git native way, idk).

Great write up otherwise, thank you!

I would have known that if I had the attention span to read all that text from OP

I don’t navigate the site at all.

I just use the commandline to push commits to repos.

For creating a new repo on sr.ht I have written a script that uses the GraphQL API (which is horribly documented in my opinion and required days of trial and error). It is not meant for general users and is specific to my needs, but anyone who is interesred can find it linked below.

If you want to use it, you have to run git init and do a commit first. Everything else should be explained in the help. The script does some other stuff that I wanted when migrating all my projects from github, which you should be able to easily modify.

https://clbin.com/EII4R

(unlicense)

https://appflowy.io/

https://docs.appflowy.io/docs/documentation/software-contributions/architecture/frontend/web

Sadly it’s a bot more complicated than just a docker container, but there is the manual install doc that goes into a bit more detail.

For anything deeper you’d have to read the script.

https://docs.dokploy.com/docs/core/manual-installation

Personally I use Dokploy. It’s a dead simple docker web UI that makes domains and ssl easy peasy

I think I am limited by the software.

With a gigabit ethernet connection, I was not able to have a good experience.

Even when my internet doesn’t suck for a minute, I have yet to find a linux remote software that is not sluggish or ugly from compression artifacts, low res and inaccurate colors.

I tried my usual workflows and doing any graphic design or 3d work was impossible. But even stuff like coding or writing notes made me mistype A LOT, then backspace 3-5 times, since the visual feedback was delayed by at least half a second.

I run this somewhat. The question I asked myself was - do I R-E-A-L-L-Y need a clone of the root disk on two devices? And the answer was: no.

I have a desktop and a laptop.

Both run the same OS (with some package overlap, but not identical)

I use syncthing and a VPS syncthing server to sync some directories from the home folder. Downloads, project files, bashrc, .local/bin scripts and everything else that I would actually really need on both machines.

The syncthing VPS is always on, so I don’t need both computers on at the same time to sync the files. It also acts as an offsite backup this way, in case of a catasprophical destruction of both my computers.

(The trick with syncthing is to give the same directories the same ID on each machine before syncing. Otherwise it creates a second dir like “Downloads_2”.)

That setup is easy and gets me 95% there.

The 5% that is not synced are packages (which are sometimes only needed on one of the computers and not both) and system modifications (which I wouldn’t even want to sync, since a lot of those are hardware specific, like screen resolution and display layout).

The downsides:

• I have to configure some settings twice. Like the printer that is used by both computers.

• I have to install some packages twice. Like when I find a new tool and want it on both machines.

• I have to run updates seperately on both systems so I have been thinking about also setting up a shared package cache somehow, but was ultimately too lazy to do it, I just run the update twice.

I find the downsides acceptable, the whole thing was a breeze to set up and it has been running like this for about a year now without any hiccups.

And as a bonus, I also sync some important document to my phone.

Dokploy is a pretty easy web gui and is itself a docker container.

Makes it dead simple to manage multiple containers and domains. (Not for power users that need kubernetes level flexibility)