[0]: https://blockchain.info/charts/blocks-size (26 Jun 2016, 0519 UTC)
1) Have "tracker" servers set up around the world. The App needs to know at least one of the IPs, preferably all, in order to get a list of Peers it can connect to. This is still considered "decentralized" so long as none of the "trackers" is considered a "master" over the others.
2) Have UserA manually enter the IP of another UserB. Then, UserB tells UserA all of the Peers it knows about, which UserA will then connect to, and so on and so forth. This approach can quickly get out of hand unless there is some sort of limit on how many peers a User can be connected to at once. That's up to the developer (or possibly user) to decide.
Theoretically, as long as one Peer is alive and doesn't forget the list of other Peers it has been told of, the network will never completely shut down. Even if it does, however, the network can "start back up" again, but there may be cases of "subnets" within the P2P network if no peers from one subnet ever know of peers from another subnet.
In regards to the actual chat functionality, it's very simple. As long as you can preserve consistency between peers (which is its own subject), it's relatively easy to have each Peer keep a copy of the current chat. Even if the chat is between >2 Peers.
There are intricacies to this, but it's not actually that difficult. It's a lot more complicated to design, but once the groundwork has been laid, it can sustain itself for the most part.
How is that not decentralized? I really don't see how. You'd need /someone/ to maintain that list of tracker serves.
Like email. Let everyone host their own chat server. Pass messages around to the correct server and let it forward it to the user's device.
Or you go the Tox / Ring route with a DHT of users and cut out the server alltogether, but the server part is pretty much always necessary for "pretty usernames". Tox has been basically dead in the water for quite some time because they have no effective DNS for users that is universal to all the Tox apps with multiple competing implementations.
My personal takeaway is that if you are going to have a server you should take full advantage of it like Matrix does, because Matrix mobile clients can be push-notify based and not need to always be online to get messages (though Tox et al have ways around that, they are all dependent on storing messages locally to the destination in the DHT, and most users do not want to have to cache other peoples messages).
As I said in my original post, it sounds similar to how Bittorrent's DHT works; you'd still need a "kickstart" server to send you a list of peers. Sure, you can hardcode a list of peers, but hardcoded stuff isn't decentralized. Someone has to maintain that list.
Not that I disagree with you but can we all stop using this as an argument.
Also what's the matter with using intent in arguments?
We have slack instead of IRC, because IRC sucks.
Consider in order to have offline history I'm logged in permanently via. tmux session on a server which I connect to with Mosh. Weechat is a barely okay client (totally not suitable for non-technical users and Mosh still has significant delay).
What do you expect the non-engineering part of the company to do?
Just the other day, a friend created a Slack channel (or whatever it's called) for a group of friends to plan a trip, and then promptly set up an IRC tunnel so he wouldn't have to use it. :-)
I don't have an opinion myself; I just find it to be amusing.
There is no technical reason why something like whatsapp can't be peer-to-peer. Choosing for a centralized service is implicitly choosing for giving the powers that be the opportunity to massively listen in on our various modes of conversation, to figure out your 'graph' and/or to allow censorship.
An old quote has that the internet sees censorship as a routing problem and will route around the break. But that only works if we explicitly refuse to allow centralized services.