nvidia doesn’t push ads. shield runs androidtv and with the default stock launcher, you will see the ads google wants you to. but you can easily change the launcher to something like flauncher and have a streamlined experience.
I too care a lot about privacy and data sovereignty but those aren't sufficient arguments to self-host. For instance, my wife cares about the two too and so she uses most of the services that I host at my home, but she isn't going to start self-hosting herself anytime soon.
I think the missing piece is you need to enjoy the process itself - without that, it's not really tenable (at least today).
ZFS is RAM hungry, plus doesn't like USB connections (like the article implied). So, I've been eyeing btrfs as a way to setup my NAS drives. Would I miss something in that setup?
ZFS is not RAM hungry. The only official memory requirements that exist specify 768MB system RAM and thats the whole OS (Solaris 9) minimum requirement. Just like any other file system ZFS will use memory if available and release it when other parts of the system need it.
Also ZFS is perfectly happy with USB connections. In fact it's the best type of FS to have if your storage is unreliable due to its self healing capabilities. Not that modern USB is unreliable nowadays and there are plenty of DAS solutions that rely on 3.x USB.
Coolify and friends (Dokploy?) look like nice tools. But I am not very comfortable with them because the state of my server(s) isn't present in code. So, I like NixOS or Ansible more but then they require a bunch of boilerplate and custom infrastructure for setting up production.
Anyone know some infrastructure-as-code framework that makes it easy to spin up and maintain production servers? Something declarative, perhaps, but not Kubernetes?
I’ve been working on doing this with Coolify. There are very few coolify settings to backup, and then all the application configs are stored in /data/coolify. And I use kopia to backup all the volumes. It’s not pretty, and a little hacky, but workable for disaster recovery.
Amazon's search results have been garbage from a really long time, I often wonder how come the executives or the team behind it never experience that themselves. I now to Amazon only if I know exactly what brand I am going to buy before opening Amazon.
I also quit Prime couple of years ago. Hardly miss it.
Amazon search looks bad for us because it is designed to sell ads. Its goal is to make company pay the most money to show articles.
Iirc, when this was proposed, Jeff Bazos said that this was the most stupid idea he ever heard.
I think the reason why it was introduced, and why Executives don't want to change it, is that it generates a ton of money for Amazon.
I'd personally love if in the end this would be the reason that Amazon stop making money and it would have been some short sighted greedy move. I'm afraid that advertisement, when it comes down to numbers, is just damn too profitable.
Unfortunately the store's primary revenue source seems to be from advertisers bidding on sponsored search result slots instead of the actual product sales.
Which can't possibly be actually true, since advertiser bids must necessarily be funded by actual product sales - I guess there's an edge case where an independently successful business intentionally overspends on Amazon ads to quash potential competitor discovery, relying on non-Amazon sales to make up for the deficit, but I can't imagine that's a particularly common result.
If people give up on buying things from Amazon because there's just no way to find reliably usable products, Amazon will eventually lose out on that advertising revenue. So either we're mid way through that process, or there's something more complicated going on.
I would love to use KeepassXC but it doesn't make it easy to share credentials with the wife. I _could_ use a dedicated vault, but we'll then need to cut-paste things for sharing existing credentials.
So, for now, I've settled on Vaultwarden and it has been surprisingly stable so far.
UV is fast, like FAST. Plus, it removes the need for pyenv (for managing different Python versions) and pip for me. Plus, no need to activate env or anything, `uv run ...` automatically runs your code through the env.
I don't see a way to change current and global versions of python/venvs to run scripts, so that when I type "python" it uses that, without making an alias.
export PATH="$(uv run python -BISc 'import sys; print(sys.base_exec_prefix)')/bin:$PATH"
2. Put this in an executable file called "python" early in your PATH:
#!/bin/sh
exec uv run python $*
Those are basically what pyenv does (via a symlink and PATH entry).
The second option will always pick up the Python local to the project directory you're in, if any. The former (if you put it in your shell profile) will not.
I specifically want to run "python", rather subcommands for some other command, since I often I want to pass in arguments to the python interpreter itself, along with my script.
> I don't see a way to change current and global versions of python
You really shouldn't be doing this. Utilities on your host rely on the system python being stable, unchanged, pinned to a very specific python version; also with exact versions for dependencies.
For example in fedora/rhel, changing the default platform-python will prevent you from using yum/dnf to apply system updates.
What's the recommend full-stack setup, if not NextJS?
My background: I have ~15 years of software development experience, but practically all of it in backend stuff. (Well, I did some AngularJS a decade ago but that's it.) So, when I recently wanted to build a fullstack app for the first time for a side project, I looked around and learnt NextJS was the way to go. (Gemini said that, then Cline's official documentation said the same.) I am early in the process, so happy to learn of solid alternatives.
I am planning to host everything on some VPSes through Docker, because that's what I am comfortable with. So, no Vercel or Netlify.
If you don't need server rendering, I'd just recommend React without framework, and Vite [1] to run the app in dev and build the app in prod, then host the build (basically a HTML page with a JS file) on AWS S3 or similar static hosting platforms. That's the solution I've been using for 10+ years without any problem.
For the backend you can use whatever you are comfortable with. I mostly use PostgREST [2] nowadays. I recommend to use react-query [3] to call your APIs on the client side.
What are you building? I am working on a classic SaaS webapp, and I have found that using React/Refine.dev/Vite has been great. Refine.dev let me focus on my features, and not basic CRUD page stuff.
I am building an AI application (of course). It'll have an agentic workflow in the backend and exposes a chat style interface to its users. So, not a lot of CRUD stuff.
I would recommend still looking at it for 15 mins. You will still need basic stuff like login pages, a profile page, a responsive nav system, etc., right?
reply