Frontend<->Backend communication in production mode

[ctmbl]

Introduction

#20 introduced the Docker system needed to deploy the iScsc website and thus introduce separation between development and production modes.
In this thread we discussed about the differences between dev and prod modes, especially everything related to the "proxy" needed for communication between the backend and the frontend.

Context of the discussion

To summarize it I asked why does the routes are different in dev and prod mode and @atxr answered that thanks to a "proxy" in dev mode the frontend can easily contact the backend as if it where a local URL, without taking into account the network as I understand it.
However in prod mode this "proxy" apparently doesn't work because of some react build that shrink the features and capabilities to make the website lighter and faster for deployment.
Thus, a "reverse proxy" is used to replace it, it uses nginx as originally configured here and its role is simply to ease communication between frontend and backend, dealing with URL, ports and so on...

Now it's time for @atxr to correct what I just summarized and add details I ignore as well as good documentation/video about these topics and technologies!

[atxr]

I'll do a little summary tonight!

[ctmbl]

tks!

[atxr]

Actually your summary is very good! Do you still have some question after the last presentation?

[ctmbl]

I quite like your explication of what nginx is if you can write it down it would be great!
But I didn't get what CORS is?