Sheogorath

@ErikMichelson not necessarily automated, no, but I would think it's important to check if the implementation is quick enough for a typical use-case and also how the bandwidth usage looks...

Newer link to camo: https://github.com/cactus/go-camo

It's a trade off. We either limit images to 1st-party pages and make things more private this way (as websites like GitHub and GitLab do) or we don't and spare...

Also note that the "EvilHacker" could just upload the image to HedgeDoc and use it from there. Same effect. If we have an image proxy the URL would at least...

I don't think that's a good idea. We need a way to migrate uploads easily, if we want to enforce that. Also this ignores that those uploads might still be...

> good point we have the creation date so we can calculate with that and the TTL. I don't think it would be a good idea to use the creation...

This was already implemented for 2.0 before the decision for the rewrite from scratch was made: https://github.com/hedgedoc/hedgedoc/issues/317

One correct me if I'm wrong, but given that it requires a service running as root or at least root-like permissions in order to modify the firewall rules, it won't...

If you actually consider add fedora native packages, it could be nice to use the new pack-it infrastructure that allows direct integration: https://github.com/marketplace/packit-as-a-service It's a project that tries to help...

This is actually a quote from the official docker best practices: https://docs.docker.com/develop/develop-images/dockerfile_best-practices/#user To see the official reasoning in all detail I recommend to have a look at the `gosu` repository:...