Killing telnet +/-?
-
So one of the things I know I’ve talked about before is what the future of MU* looks like. What systems we might be using in a year, five years, etc.
I think it’s fair to say with the rise of Ares lots of people are excited by the move to web and away from telnet for many things - caveat: I know this isn’t everyone and some people just don’t get on with Ares, but I don’t want to rehash that discussion in this topic.
It gets me thinking, how much are we held back by telnet? If we could drop it entirely and do everything in a browser, what would we gain? And what would we lose?
What would we lose?
Dedicated clients would probably go - one of the advantages of telnet is that the protocol is fairly simple, and everything else is window dressing in a client - you send text, you get text. The client is completely unaware of the meaning of the commands. Even if every game started from the same system that had the same API, whenever you add something it’s going to add another call or object to the API, and a single client wouldn’t be able to understand all the variations.
If you lose dedicated clients, you lose a lot of notification functionality. Sure, browsers have some notification ability, but now your game is just in another tab somewhere and it’s easy enough to close by accident and then where did my game go? Also, mobile browsers are an entirely different issue.
(Side note, you can do things here with service workers in the browser, but it’s still not the same)
You might be able to do something clever with an Electron app that just behaves as a window into the game’s site, but I wouldn’t want to assume its workability without some serious testing.
There’s also accessibility concerns. Telnet is easier to make friendly for screenreaders than a single page app website. It’s not impossible, but it’s already an afterthought in many areas, and it could get lost at the wayside, closing off games to an already under-served population.
What would we gain?
Killing off telnet does come with its advantages, of course. Development can be focused in one area, whenever you add a new system, you add it in one place, rather than having to add the telnet-side code as well as the web-side code. You can get rid of long and arcane commands that need to take in multiple inputs and make it a single dialog box. Information can be displayed in ways other than just ASCII.
You could add rich information to your interface - when you enter a room you could get a picture of the room. For your grid you could display an actual grid of information that shows you where the various rooms are, the exits you know about and so on.
It could improve accessibility. Not everyone wants to play with telnet clients to play a game and creating it all in browser could bring more people on board. And if done properly it could make games more accessible to screenreaders, but again that’s a big if.
And of course, just because I’m a big showoff, here’s a gif of the thing I’ve been working on on and off (mostly off). I wanted to build something customisable where you can set our your layout however you want, with potentially multiple character windows worth of stuff at once (maybe you want to RP as your PC but chat as your staff-bit). All in browser of course, the gif showing two browser windows at once.
So far the backend is written in Python, exposing a GraphQL API, and the frontend is written in JS using the Vue 3 framework.
But other than just being a vehicle for being a showoff, this topic is also for other people’s thoughts, so tell me, do you want to see the end of telnet? Or is it so core to MUing we can’t get rid of it?
-
@Rathenhope I have nothing against telnet, I do not come to destroy it, but I don’t see any real benefit to keeping it if we can move on to something that meets modern needs better.
-
I believe they can coexist. For games that don’t have built-in screenreader support, telnet is practically a necessity. I agree that modern solutions are viable. It’s just a matter of developers putting in the effort to support accessibility.
While Evennia isn’t flashy out of the box, it does support screenreaders, both in its webclient and with an in-game option.
Your demo looks very cool, @Rathenhope!
-
I am biased against clients that show images instead of room descriptions. I realise that people often don’t actually read the descriptions – but using images restricts builders to what they can actually get images of. So either you need one hell of an artist on staff for many games or you’re going to end up with a weird mix of fan art and stock photos.
-
Anything that seeks to wholly replace what we have now needs to do almost everything we can do and do it better than what we have. We’re creatures of habit and are psychologically driven to stick with the status quo if the benefit of a new thing is only marginal. This is why, I think, things like Ares have worked so well. They add new methods to introduce new people to the games while having the old methods around so that those already here can join in and, eventually, transition.
MMOs were born out of MUDs, but didn’t replace them. So if you make something so fundamentally new and alien in design… it won’t feel like a MU even if it does the exact same things. A MU, to me, is almost entirely text. Sometimes it’s fancy text, sometimes it has colours, but it’s all text. If I wanted fancy graphics and pictures and all that jazz, I wouldn’t play a MU.
-
I just want a client where the GM can pipe atmospheric music at me during their scene. (And I can choose to mute it, but come on, that’d be rad.)
-
@Solstice YES! THIS!
-
@Solstice said in Killing telnet +/-?:
I just want a client where the GM can pipe atmospheric music at me during their scene. (And I can choose to mute it, but come on, that’d be rad.)
I’m having a lot of trouble thinking of anything I’d like less in a client.
-
To be fair, if MUs go totally web based, I will probably quit altogether.
The reason for this is that it seems (to me) that web-based MUs trend toward asynchronous play, and I personally don’t prefer asynch. If I get into a scene that gets declared asynch at the moment (which doesn’t happen a lot), I will completely refuse all other RP until that scene or PRP or whatever is resolved. It’s a combination of my brain refusing to wrap around the IC timelines involved in an asynch scene, plus an aversion to retconning, which could be required by an asynch scene depending on when the asynch stuff is happening compared to the rest of play.
-
@Solstice said in Killing telnet +/-?:
I just want a client where the GM can pipe atmospheric music at me during their scene. (And I can choose to mute it, but come on, that’d be rad.)
I feel like I’m mentioning Evennia too often, but Evennia’s webclient has built-in support for that!
self.msg(audio="URL")It can do the same for
videoandimage.It’s just a matter of adding a staff-side command to allow STs to message that to everyone in the room. And a player command so @Pavel can MUTE that stuff.
A friend of mine has made a few tracks for Silent Heaven, and I’m excited to add them to the game.
Game-runner of Silent Heaven, a small-town horror MU.
https://silentheaven.org -
Mind blown. I need this in my life.
-
To me, a MU is a combination of three things: Real-time, text-based, and roleplaying. Altering those three things too drastically makes things feel less MU-like, as far as I’m concerned.
So honestly I don’t care what happens on the back end (though clients are preferred. If I have a MU client open, I am in MU-playing-mode. If I have a browser open, I’m liable to get lost on Wikipedia and totally forget I’ve got a scene in another tab) so long as those three core things remain the same.
This could very well just be a case of old man yells at clouds, but it works fine as it is. Some changes here or there to make things more approachable, or easier to develop? Sure. But if you’re going to throw music and pictures and shit at me when I come wanting a text-based roleplaying experience, it’s not what I’m there for. It’s like asking for something to read and being thrown a DVD.
-
Most of my favorite tabletop sessions, the DM has put some real effort into creating mood lighting when the time came for spooky parts, and had a selection of creepy music and even soundboarded some sound-effects.
That being said, creating the possibility for people to use tools or features if they want them is a far cry from shoving them down people’s throats.
/audio off
/images off
/clouds on
/screams on -
@Solstice said in Killing telnet +/-?:
That being said, creating the possibility for people to use tools or features if they want them is a far cry from shoving them down people’s throats.
Okay, but what’s wrong with sending a Youtube link or a link to a Spotify playlist or something? Why does it need to be done through the client? Why does it need to be a built-in part of the game?
We have the MSP (MUD Sound Protocol) already, and it’s rarely if ever used.
He/Him. Opinions and views are solely my own unless specifically stated otherwise.
BE AN ADULT -
The need here has nothing to do with telnet and everything to do with having a desktop and mobile client. MUSHes have the most similarity to chat apps like Discord/Slack, and they all have native clients. Next-gen MUs should have one too, or you lose usability. Notifications, accessibility, custom output, logging, and various other features work better on native clients, and there’s a preference issue. (I’m a slack/discord desktop client girl myself.)
A next-gen client shouldn’t use raw telnet, though, it should be API-driven.
Unfortunately there’s a mountain of technical challenges in the way of making this happen.
-
@Faraday Agreed. I use the Ares webclient as a matter of convenience when I don’t have my regular client available. I definitely couldn’t do it full-time.
-
@Faraday Right, and I don’t disagree, but telnet’s lingering ability has been to be pretty much game-agnostic. You want to add another command with a new display output? Sure! Telnet doesn’t care about anything but sending and receiving text.
With a native client (which I agree is preferable because you have so much more control), if game A wants to add portals and game B wants to add spaceships and game C wants to add an economy, well then the API for each of those games diverges and either you have to have a Super API that can somehow standardise how to add entirely new components which the system developer didn’t consider, or you end up with a client per game.
And suddenly to run a game not only do you need to be able to write Python or JS or Ruby or whatever the backend code is written in, but you also need to be able to write desktop and mobile apps, which is an entirely different skill set (and when we start getting to the mess that is iOS application publishing, an entirely new headache)
(Yes I appreciate that a lot of that complexity can be managed with cross-platform apps these days and so on, but there’s a still a host of complexity)
I mean all of that is probably included in your mountain of technical challenges there, but it’s why I started by poking at a webapp version first, because at least some of that complexity can be shoved under a rug, for now.
-
@Rathenhope said in Killing telnet +/-?:
I mean all of that is probably included in your mountain of technical challenges there, but it’s why I started by poking at a webapp version first, because at least some of that complexity can be shoved under a rug, for now.
Sure, I wasn’t suggesting that you shouldn’t work on that thing. I just think it’s solving a different problem.
I don’t think every game having a separate desktop or mobile client is realistic, either from a game runner’s perspective (as you said, it’s unreasonable to expect them to develop that) or from a player’s perspective (can you imagine having to download three or four different apps to play?)
But I also don’t think that having a generic non-telnet API is unrealistic. It’s not trivial, but it’s not impossible either. I’ve already started designing around this for a hypothetical Ares mobile client.
-
I was thinking about features I’d love to see in whatever communications protocol replaces telnet for MUing, and the big one I keep coming back to is:
Markdown support.
Also, if we’re talking about moving from a mostly deprecated communications protocol to a newer more robust one, what’s stopping the MU community from jumping to SSH?
Proud Member of the Pro-Mummy Alliance
-
@Pavel said in Killing telnet +/-?:
@Solstice said in Killing telnet +/-?:
That being said, creating the possibility for people to use tools or features if they want them is a far cry from shoving them down people’s throats.
Okay, but what’s wrong with sending a Youtube link or a link to a Spotify playlist or something? Why does it need to be done through the client? Why does it need to be a built-in part of the game?
I mean, nothing’s wrong with that, but it’s absolutely a different experience. Solstice is definitely talking about a sort of feature that’s very common in tabletop platforms and which I can agree is pretty awesome when a DM invests in using them. It’s a different experience getting a link and opening it vs having the DM kind of – curate and control the audio experience.
I’m not saying it should be a top priority necessary feature, but it’s wrong that there’s no experiential difference between that idea and just sharing a link.
