If you want to get the job (of publishing a blog) done fast, and move on, then use WP. If you want to mess around probably look at editorjs.io.

If you want to get the job (of publishing a blog) done fast, and move on, then use WP. If you want to mess around probably look at editorjs.io.

At least WP is free, Ghost is as “free” until you find out its only useful with the rest of the payed platform. editorjs.io is much better in that sense.

Yeah Microsoft for what’s worth does play ball, you can open complaints and they’ll actually read those and act fast. Google is a total pain to deal with, even if you’re on some type of google partnership they’ll not do much.

It’s also good to make notes on every configuration setting.

I do save my settings for the various programs in a git repository…

If it need documentation means things are over the line when comes to complexity and I should scale down / simplify. :)

Complexity and over-engineering are a serious problem, I really try to keep it as simple as possible so I don’t have to waste time managing it, dealing with updates and potential security issues. Simple code/infrastructure breaks less and has less potential insecure points.

Unless someone finds a way to advertise nodes that doesn’t depend on the entry point then yes. Consider this example: https://github.com/bitcoin/bitcoin/blob/1b2460bd5824170ab85757e35f81197199cce9d6/src/chainparams.cpp#L112 if someone takes down those domains it is game over for a new node until someone updates the code.

I get your point, those systems make it harder to take down things permanently but they aren’t as resilient and perfect as people paint them to be - an it has nothing to do with being pedantic, it is just the reality of things.

Here: https://lemmy.world/post/29490236/17012887

My point was: if you still need some central point of contact what’s the point in decentralized, you can still get fucked.

For instance the DHT systems you talk about, they’re good but still require some centralized points. In a bittorrent network with DHT a new client cannot join without either a tracker or the knowledge of at least one member of the network to exchange peers with. Bitcoin still has some hardcoded DNS seeds in the core client… etc.

bittorrent decentralization

True bittorrent decentralization never happened.

There’s no real / true decentralization. You’re always dependent on something, somewhere in some way. It can be harder to shut it down but there’s also a point of failure somewhere. Blockchain is all fun and games until you’ve to consider resource waste and that you still need DNS and IPs working.

Yeah, that’s a good one as well.

Filebrowser is great, it just lacks two things 1) 2FA and 2) the always upcoming OnlyOffice integration. If we got those two nothing else could ever compete with it. It already does pre-views and text editing, but Office documents would be great.

I can split the config to another file, not really a big deal. :)

In the simplest form it might be SSO. It does support multiple users and if you look for instance at the filebrowser it’s very possible to pass the username. But yes, this is very simple, very crude and exactly what a lot of people need.

Hmm… some people are going to say that basic auth would be insecure, I’m not going to be there because in this particular case it’s about the same thing.

However, this might be easier to configure and manage permissions than basic auth. Also this works cross-domain and basic auth will require full re-auth for every domain. Another obvious advantage is that at some point I plan to integrate 2FA.

You can backup the entire file then. I get your point, but it also seems like you’re referring to some container-based approach where you would place this inside a container and then mount the config file to some path. While some people might like that approach, that kind of goes against the original idea here, I didn’t want to run yet another instance of nginx for auth, nor another php-fpm - the ideia was simply to use this on a low power device , no containers, no overhead of duplicate webservers and PHP, just a single nginx running a couple of apps on the same php-fpm alongside this.

Well, it isn’t pretty, but gets the job done.

The thing with PHP in this case is that I was already serving a ton of simple websites / small apps like freshrss that use PHP and by making this tool in PHP it means I don’t need yet another process running and wasting resources, can just re-use the existing php-fpm for this.

For what’s worth PHP is better than it looks, and my implementation is very crude, but also small and auditable and contained to a single file. :)

Alternatives? https://filebrowser.org/