the thought process is that the problem with the original City of Heroes servers is that a single authority could shut them down. Someone thinking about that problem might want to consider whether it is possible to make a City of Heroes game that is simultaneously unified - so the community doesn't have to fragment into tiny pieces - but also resilient so that no single person can shut the game down. There are a number of possibilities that come to mind.
1. Distributed gameplay, unified messaging. You could make the Fidonet equivalent of an MMO. A lot of small community servers that worked in isolation. When you log into your local server, your characters are there. They are nowhere else. You can't log into anyone else's server without making an account there. When you do, you have to make characters there. But if you want to chat with the community, the global chat channels cross servers. There's limited global community interaction, but gameplay is fragmented into lots of tiny pockets of players.
2. Portable characters. You let people play on any server they want, but you can export and import characters from anywhere to anywhere. This presumes there is a consensus on trusting imported and exported data. This kind of thing also tends to make it difficult to conceive of a business model where this works. A community might agree to allow all kinds of stuff, but a game, in the technical sense of the word, cannot simply allow unrestricted freedom of data motion and implicit data editing.
3. Distributed servers. You create lots of individual server shards, but they all work together in a large supercluster of game servers, much like the original City of Heroes servers did. A single server going down is similar to a single CoH shard going down. With better global copy and/or migration, we can make the system more or less invulnerable to a single host either dying or being taken down. If the shards are sufficiently distributed in the jurisdictional sense, it can become impractical to take them all down.
3b. Distributed peer to peer. A special case of (3) is to make a design that allows an unlimited number of servers coordinate player characters and progress, and work both online connected to global servers and offline in a stand alone mode, with a way to arbitrate synchronization. You can always just run your own server if you want to (in theory) but unless you implement (2) above, your characters are "trapped" on your single player server. You can't sometimes play offline and sometimes play online, except with different characters. Hypothetically speaking its possible to allow a player to play on their own server, then reconnect that server to the global clusters to play within the global servers. But there's technical and logistical problems with doing so, in particular with offline servers the global servers cannot trivially validate that any of the data on that server was validly generated.
Thinking about what's possible is the first step to determining what's practical. I would assume anyone thinking about making, or designing, or resurrecting an MMO ought to be at least thinking about these issues.
1. Something like Paragon Chat? And you can run your own server with that?
It's highly theoretical at the moment but it shouldn't really seem all that ambitious.
At this moment, I have a XMPP server running on my laptop and a character logged into Paragon Chat on that local server. You could do the same with a half-hour of downloading and setup.
If some sort of game engine materialized around Paragon Chat, then anyone could install it onto their home computer and play it as a single player game. All of those people in fan groups on Facebook and who keep an eye on Titan forums would do that.
You could also play it in groups at a central server like chat.titan.com or by connecting to your friend's personal server.
If a distributed character database existed then you could freely switch servers and it wouldn't matter. It would also mean that you couldn't shut down the game. You'd have to shut down the entire network.
Look at Nostalrius. What would Blizzard do if Nostariius had been just one server in a federation of hundreds, and the user database was spread across a network intended to be as resilient as the internet itself?
Ie. Play local. Run 'CoH' 'Simulated' on your own micro server i.e. as a local single player game... Shards that are like Dandelions. Persistent and resistant. As soon as that 1st distribution of CoH Simu... Like having your private party...
I always felt this functionality should have been built into CoH. ie. An 'offline' mode where I am the local server.
2. Sentinel? I never used it. I wish I'd known about it.
But the idea of a 'Bit Coin' aspect to Characters or game progression is intriguing. I initially think of all those different C64s EMUs that run those common 'disk' images. While CoH becoming a similar portable EMU with portable characters and portable progression teases the mind...it's not trivial but in the sleep of death no dreams will come. Best to dream, speculate and be ambitious...
3. Distributed Servers. Clusters. A federation?
Lots of individual parts. A network of parts that can be offline but also 'online' if so wanted.
The above examples make me think of Unreal Tourney. I know it wasn't an 'MMO' but I really liked that game. You had an offline mode. You had solo games 'with bots' with the challenge ladder of various flavours, all having their own appeal. You had a local LAN option. You could play over the Internet. Very forward thinking. Progressive. You still had 'something' even if nobody ever wanted to play online anymore!
Or long after the company concerned had finished exploiting it. Any CoH SIMU effort ought to think along progressive lines and wander creatively down those tracks. Eg. Offline modes. Locan LAN. Internet 'option.' Bitcoin 'progressive' data. 'Block chains?' Vanilla, Strawberry and chocolate flavours. Mod packs...for mobs. Zone editors. Mission generators... Even a port to a new engine...in time...
I think that MMOs missed a trick, in general, by not having a similar progressive design. Intravenous drip mentality of dependency then rip the cord out when its hooked you...then kick you to the kerb. You can hardly complain if the MMO base don't buy your new 'mediocre trinity' games 'made of shiny plastic' when you treat them so. Live by the sword and all that...After all. Consumers don't have to buy your product. Well, unless crony capitalism lobbies to mandate that you do...
4. Peer to Peer.
My preferred option. Maybe with offline and Local Lan capability. With CoH SIMU with a quiet low key release into the wild. It's private. For 'educational private' use only.
I wonder if a 'bolt' on real time CoH Simu element might be added to Paragon or whether, like Icon, the 'next step' (CoH 'Next' or 'Legacy') supersedes the predecessor as a different offshoot project.
I still have to pinch myself that we have Paragon Chat. So the proof of concept or at least 'progressive thinking' is kind of there...
Anyone can say the words "lets turn the CoH game client into a distributed graphical interaction tool using a central messaging server and translation interface" because those are just words. There's no ambition in them until someone actually makes one.
What, like Paragon Chat?
Azrael.