dontblink

Let's say I want to bridge from WhatsApp or telegram to Matrix, have I gaibed something in terms of privacy? In which case would it make sense? Public group chats? Direct chats?

Some services run really good behind a reverse proxy on 443, but some others can really become an hassle.. And sometimes just opening other ports would be easier than to try configuring everything to work through 443.

An example that comes to my mind is SSH, yeah you can use SSLH to forward requests coming from 443 to 22, but it's so much easier to just leave 22 open..

Now, for SSH, if you have certificate authentication or a strong password, I think you can feel quite safe, but what about other random ports? What risks I'm exposing my server to if I open some of them when needed for a service? Is the effort of trying to pass everything through 443/80 worth it?

Let's say I have a domain called mysite.com

mysite.com points to a server which only opens port 443, and each connection will need to go through that and deal with Caddy reverse proxy.

I want to host more services on it.

Let's say I want to host an email service, the easiest thing would be using a subdomain such as mail.mysite.com and reverse proxy each connection to the internal port on which the service run.

Same with a chat service chat.mysite.com.

But for the sake of readability it would be much better to simply have username@mysite.com than username@mail.mysite.com or username@chat.mysite.com.

reverse proxying every request from a subdomain to the right port is pretty straightforward with Caddy, also if you use cloudflare you can proxy with cloudflare each subdomain and have auto SSL certificate without further set up, which is amazing!

But what if I do want my services to be accessed through mysite.com directly instead of a specific per-service subdomain?

Some federated services also have two separate ports for server requests and client requests, which further complicates the process..

Is this service specific and must configured individually for each service? Or there is a way to tell caddy that a specific request going through mysite.com should be redirected through port X.X.X.X? Is there a way Caddy can recognize where requests need to be directed?

I've been self hosting a matrix istance for a while, but I'm honestly really really tired of bugs on clients and authentication not working, I know matrix is very feature rich and is awesome that is federated, but I'd prefer to use something which loads my server less and which is more simple despite lacking some features, what do you think?

So google now requires Id verification for submitting apps to android, what does it mean for Foss apps, for Foss stores like fdroid and for future development?

TypeScript does not throw an error at compile time for accessing an out-of-bounds index. Instead, it assumes that the value could be one of the types defined in the array (in this case, 1 or 2) or undefined.

TypeScript automatically infers the type of a value accessed from an array, even if that access is out of bounds. It assumes that the value could be one of the defined types or undefined, which can lead to confusion if you expect stricter enforcement of valid indices.

I just spent the last 2 hours trying to understand why I was getting a valid type from something that shouldn't have been valid.

I think that the hate that JavaScript receives is well deserved, at least coming from Rust this is an absolute nightmare.

After a while since I left JS, I decided to learn Typescript, React and to give it a go again.. It's all fun, until you have to hunt down bugs..

I'm working with an headless CMS, so you have like 4 layers of stuff: JS, TS, React, CMS.

My opinion is that debugging works for SIMPLE stuff, like if I have a static website generator, debugging and hunting the errors by hand to find what's wrong it's okay. But when you have to do really complex stuff, interact with APIs, working with promises, async requests and all, I do think that hunting bugs like that is the recipe for a disaster.

My project runs into a bug: I get 27 different errors messages from different tools, some of them are easy to misunderstand and not really pointing out the right problem. And I don't want to imagine what's like working with a big fullstack app where you have to manage backend as well, not just its APIs..

Once things starts to get complex the language MUST have a good compiler, MUST have a very good package manager, an included linter which is the SAME one for everyone, and API/libraries documentation tool which looks everywhere the same, and most importantly a carefully crafted error handling system. Once a language is very complex it needs to goes straight to the point and tells you where the issue is, what you are doing wrong, and it's okay to have layers but you cannot use 5 different error detection mechanisms for one project.

Consider I only know apis are structured data that can be called or modified from within a program, and have no real further knowledge in real use cases nor in networking.

Where should I start from? Should I study backend?

I prefer docs rather than videos.

Something like Android auto or apple auto but with Linux, does it exist? Or if not what are your homemade solutions?