My worry with svelte is vendor lock-in. It has a specific way of doing things, and I feel like my simple app does not need that level of interactivity abstraction. But maybe I have not looked into it deeply enough.

I suppose I just cannot imagine a situation where a rich interactive web app needs SEO that is not solved by having a separate landing page (and such solution not being way better and less effort than alternatives).

React has enough flexibility to be added later on to a project and not take over the entire web page, but only the parts it manages. It can act as a separate island or series of connected islands. So if you start thinking you dont need react but then later decide you want to add it to your non-react site, it should still be easily doable.

That's definitely the best argument for it, I don't deny. But I'm glad to see the web components space to be improving as well.

I agree that things improved. React and others are amazing for CSR. We have static site generators which are also amazing and nice to work with. But SSR territory is in strange place right now. React is overused in places it doesn't belong.

If SEO matters beyond a couple landing pages, I find it unlikely that you would be developing a rich web app with that much reactivity. You are more likely content focused, in which case a static site generator or simpler SSR frameworks are easier and fit the use case much better. Even from a performance perspective, why ship the entire react run time if you do not need it?

And on the other hand, if you are developing a rich web app with a lot of interactivity, then do you really need SEO beyond a couple (or one) landing pages? You should develop the web app in React CSR and build the landing pages as static sites to optimize SEO. That is a lot easier to me.

I started using my own WireGuard config instead of using tail scale. Works great for me, though it does take more work up front.

I intentionally answer wrong to confuse their AI model training. It does not work if the choice is obviously wrong, but if you do it with ambiguous ones, it lets you pass. Like if wants you to select birds, and the thing is just a bear that kinda can pass for a bird if you aren't looking deeply, I'll say it's a bird.

Doing my part of destroying machine learning models.

Windows is just not ready for this stuff. Most of this stuff is built for Linux. Linux is THE server OS. And windows is painful for developers too, so there's less solutions for it.

You'll be a lot better off with Linux for self hosting.

If you have to ask, no.

I say this as someone who doesn't use systemd. There's not much benefit to it. It's cool to do if you're an enthusiast or experimentalist, but from a practical stand point, systemd is most practical.

I use gentoo with openRC btw.

I used to mess with LAMMPS in college, and now that I have pursued software engineering, I wish I could get back into it and see how I can contribute.

Any resources you'd recommend to get back into this?

I like the idea. The only thing is I would worry it would add extra effort to the whole process, making you think "oh now I have to setup the stream and everything... Nah". And just programming might seem much easier to "just start it".

Do 1 hour a week. Schedule it to make sure it happens.

Consistency is more important than volume.