Multimedia VoIP for hard of hearing / deaf-blind :) Working on a platform matching deaf video / text callers to interpreters, who then dial out to PSTN (phone network). Granted there is some web-stuff involved (one of the clients involved is Web-RTC), but it is mainly VoIP protocols among non-web or even 3rd party clients.
Working with a stateful and complex protocol like SIP is fun and a pain. You need stateful load balancers, traffic routing can be tricky (f.ex. in k8s...), interoperability between providers is not trivial, etc.
We also use relatively niche protocols for real-time text over RTP, which isn't (or until recently, hasn't been) supported in much VoIP software. This is starting to change due to EU legislation requiring accessibility in all commercial PSTN services, however.
The market for this type of tech was very small ~15-20 years ago. Thus, the company was comitting to every possible sales lead, leaving a right mess in tech debt. So much of the work is handling and working around that...
Much of the product pre-dates viable, affordable 3rd party technology, so there are home-grown components for all sorts of things: conferencing media component that also supports text, RTP proxies that were needed because few 3rd party clients supported ICE or TURN, etc.
Stateful and complex protocols, with home-grown bespoke software components, 3rd party interoperability, 15-20 years of technical debt, and changing customer demands. Fun and challenging!
> So, to traverse these multiple stateful firewalls, we need to share some information to get underway: the peers have to know in advance the ip:port their counterpart is using.
> [...] To move beyond that, we built a coordination server to keep the ip:port information synchronized
This is where I wish SIP lived up to its name (Session Initiation Protocol, i.e. any session, such as a VPN one...) and wasn't such a complicated mess making it not worth the hassle. I mean it was made to be the communication side-channel used for establishing p2p rtp streams.
Responding to OP, a few too many parallell sub-threads.
Addiction to video games is a very different thing to using video games as a coping mechanism for severe mental health issues, which sounds to be the case here. Removing video games is just removing the coping mechanism, not solving the core issue.
This stuff requires professional help to deal with. If your kid is having issues with anxiety disorders / autism / adhd / etc, they may be in a situation where they simply cannot handle seeing a psychiatrist in-person (i.e. verbal shut-down, complete anxious breakdown from leaving the house, etc) which makes receiving a diagnosis... almost impossible. A diagnosis you need to qualify for most help offered by society.
As a parent you are completely f*cked if the place you live in does not offer the correct help in situations like this. You are forced to navigate through welfare systems that are not at all designed to handle people with these issues, and forced to expose your child to (to them) potentially mega-traumatic experiences, making treatment even harder.
This group of kids is growing larger in all western societies and most don't have systems in place to help them (yet).
Being faced with threats of prosecution as a parent in this situation sounds extremely rough... I have family that have dealt with similar issues in another country, and it is finally starting to improve after ~8 years of depression, missed education, navigating overloaded and maladapted welfare systems. Getting in contact with a psychiatrist over video call to get a diagnosis + treatment was a sloooow but eventual start to a solution. Getting a diagnosis helps to qualify for better help.
> From what I can tell The Crew is an online game? What am I missing here.
The main selling point of The Crew was the large, open-world map. All people I know who played that game completely disregarded any multiplayer features (of which there was not that many, at most amounting to co-op throughout the game's campaign).
> I feel like it is well understood that an online game, like Fortnite, WoW, whatever requires a server to work and may at some point stop working.
For most online games, it is absolutely possible to provide basic server functionality, even for MMOs like WoW. See another comment I wrote on here about CSGO/CS2. WoW (and Fortnite i think?) are in a different class than The Crew since they follow a subscription model, where it is more obvious that you do not own the game.
I don't quite buy this, having grown up with source engine games.
CSGO/CS2 (and to some extent TF2 before it) makes it fully possible to host your own, dedicated server, due to its heritage from the 90s-era HL1 mod. At the same time, there is complex matchmaking and ranking support, existing completely in parallel.
I don't think anyone expects to have the entire matchmaking and competitive ranking infrastructure available to host. Bare essentials for hosting a server + ability of players to join should not be terribly hard to provide, especially if companies know before hand that this will be required, and can thus design server infrastructure accordingly.
As for this being done via legislation / court precedence, I see no other option. I don't see how these basic tools are technically hard to achieve for the vast majority of games, and I don't see any world where companies themselves would consider providing this.