The relevant information could be made available on the different approaches taken, the way they can be implemented, and more than likely shared code as well. In this way, with an easily found point of reference, MUD developers might be more likely to adopt something that at least gives some degree of familiarity to players. Or even just adapt existing favoured approaches made available there. This makes me wonder why there isn't a standard MUD development wiki, something that would seem ideally suited to this sort of thing.

Interesting discussion in forums, like this thread for instance, is all well and good, but it doesn't serve the same purpose.

A case in point is web-based MUD clients. It is impractical to expect every MUD who might adopt one to use existing packages for the client-side or server-side. But snippets they can plug into their existing code, are much more likely to be adopted. Maybe adding a nice pseudo-console. Maybe just hooking up the back and forwards communication via AJAX.

There are snippet repositories on web sites, but this would be information embedded into a range of pages on the subject, giving the pros, cons, tips and alternate approaches. Whether to write a custom client for your MUD, whether to write plugins or scripts for an existing MUD client, or whether to a web-based client and if it should use AJAX or Java.

Of course, it is all very well to suggest that such a thing is needed. But I believe the only real way to get one started is to go ahead and make one, populate it with content and then come back with it. I've already done that once this week for another subject, and it took two full days, so maybe I'll mull on this for a while.



No, but I lived for many years in country where English is a second language, and it is likely my use of it has suffered because of that.

I do not agree with your strongly expressed beliefs. But there is one aspect that I do agree on, which is that as soon as you add graphics into a MUD, the MUD no longer has the same facility to grab the imagination.

But as a programmer working on a MUD with a 3D world, visual tools are invaluable in working on it. And of course, not every player is the same, so where some like me might lose the degree of immersion that comes from having them available as playing tools, others might find them to be the thing that makes the difference between the MUD being playable or not. And again as a programmer, I enjoy working on both text-based and visually-based approaches, so the challenge of trying to make the game playable in both is something that makes it interesting.

I think we're talking past each David because I don't think our positions are all that different. Where we may disagree is that you seem to be implying that text itself is what is holding MUDs back and that we should move away from text towards graphics. My position is that text games do have a market but in order to reach out to new players MUDs need to present text in a more accessible way. Terminal windows, vt100 and ASCII are not the way to do this IMO, instead MUDs should aim to be playable over the web with readable fonts and familiar interface elements, and yes even some graphics.

Yes, I completely agree – it's much easier to lead a charge when you actually have a horse to ride on. This is why I don't expect serious standardization work to emerge until people have a better feel for how things actually work in practice.


I apologize, I did not mean to imply that your English is poor, it is in fact very good. I only asked because you spelled "approach" as "approche" (the French spelling) and I've seen many French people write "donky" for "donkey", so I was just curious. (I spent ~12 years in France.)


Well, there's using graphics for the main display, and then there's using graphics to display secondary information such as a mini-map, health bars, etc.


I think that the insistence on only using text is holding MUDs back; I have no objection to keeping text as a large element of the medium. In this thread, for example, I've been talking about graphics to represent tactical maps or other forms of mini-maps; MUDs have been doing this for a while now with ASCII (albeit poorly) so I view it as an improvement, not a complete and utter change of genre. The graphics I suggest are in service to the text. I think something might have misframed our initial conversation somehow as perhaps you thought I was talking of total replacement and I felt you were talking about no graphics at all (which confused me considering your stance on web clients etc.).

I get you now, and yes I would agree with that. As an example of what I'm doing, this is how the new web client for MD is looking.

*whistles* That looks excellent.

Yes, that is exactly along the lines of what I have in mind. Some people would probably cry foul at such a thing, but just IMHO I think that the future of MUDs lies in making the "wall of text" far more approachable to the new user (not necessarily totally new to MUDs in general, by the way). Note that your display, beyond being prettier than plain text, is providing useful information in the form of status icons, very easy to process health bars, what appears to be the time of day, and so forth.

For me, graphics are not about prettiness: they are about providing information in a manner that is far more rapidly processed. For example, yes, yes, you can put up a tactical map using ASCII symbols, but what is easier: seeing a little orc sprite on the map, or seeing some ASCII symbol you have to mentally map to orc? You can always get used to the ASCII symbols (just talk to any rogue-like player) but this is yet another barrier to entry, and yet another thing that can scare away new players.

While I find Shasarak's position to be somewhat excessive and extreme, he has a point in that you don't want to be in the business of writing a game that pretends to be a graphical game but really isn't. The bulk of your presentation has to be, well, presentable.

That looks quite fantastic at a glance. Good work.

The map is very nice, the rest looks rather bleak though in comparison. I'll happily admit I'm wrong when you have a 1000 player peak, but I don't think this approach is going to cut it.

I was thinking the other day how Zork is much more fun to play than the average MUD, and how utterly uninteractive WoW's world is in comparison. Are there any MUDs that surpass Zork when it comes to interaction and exploration?

And what is going to cut it? ASCII with custom fonts on telnet/tintin++…?


It is difficult to create highly interactive experiences in a game where lots of people are running around interacting with the same game world.

Out of curiosity, how do you get your map graphics? Do you hire an artist, do them yourselves, or have a friend do them as a favour? I ask in the interests of a better understanding of what an open commercial MUD has done, in order to have a better insight into possible alternatives.

Judging from this post, I believe that Orrin makes them himself, using Campaign Cartographer 3.

You're leaving out the most popular VT100 client of all, zMud. Custom fonts won't change a whole lot either, but it'll accomplish (more or less) the same thing with much less effort, allowing developers to focus on actual game content, rather than eye candy at the cost of accessibility, especially when you talk about clients that only run on Windows, and no, Wine sucks.

And speaking of a vested interest, how about you and your 4000 posts on the mushclient forum? ;)

One thing I'm curious about is how to make logging work with VT100 emulation. If the text can be modified anywhere on the screen, how can you create an accurate, non-annoying log file?

As a rogue-like player (translation: Someone who sucks really bad at nethack), I don't find the roguelike display to be particularly difficult. That said, it was a bit odd at first not knowing what the letters meant. I was thinking a decent way to give a roguelike display to a mud without it being jarring to newbies would be to simply have a legend displayed beside the map. It's not like nethack uses the whole screen in any window I run it in anyway (whether in a urxvt terminal or on the console in a framebuffer), and if you were to "square" the map display, you'd have leftover space for a map legend.

e.g.,

> - down                                                 —                                                                                      
< - up                                                  – –       —                                                                           
% - corpse (halfling)                                   |   |     — —
d - a tame little dog                                   |   –   –     |
@ - you                                                   –    |   |      —-
i - manes                                                 |    -|   –        |
                                                          |     |   —  —  |
                                                         –    >|   | %  | – |
                                                         |    — — – |  —
                                                        –  —   |   || |
                                                        |   |   — —| |
                                                        – -|   |   |  | |
                                                       i -  —– — – |
                                                          d  ||   |   |  –
                                                         i.. – —   |  <-
                                                         .@.    |
                                                         …  —

Note that this isn't colored, which makes it look pretty ugly.

I find that highly incomprehensible, sorry. :thinking:

There's a roguelike called dungeon crawl which has nice 2d and 3d tile sets.

I agree with Twisol here; I look at that and it makes my mind boggle. Sure, I could figure it out if I looked at it enough, but every second required to understand it is another potential second for losing new players.


So zMud is the future of MUDding, the solution to our problems of attracting new players to create 1000-player-peak MUDs…?

Concretely speaking, what exactly is your suggestion here? All you've done so far is say how all these clients can do VT100 graphics, and you have completely ignored my statements that VT100 is not at all sufficient for what I'm trying to achieve. You've said several times that all sorts of things aren't going to "cut it", but then you speak to me of a decade-plus-old client as the solution to modernizing these games…

I think that you'd have to have some way of separating the flow of (presumably loggable) text from the "out of band" text like the map. You could perhaps stop logging whenever the cursor is moved and start logging when it's reset, and then just cross your fingers…
A better solution IMO is to have the data appropriately tagged using some form of semantics so that client-side logic can determine if the data in question should be logged.

You guys seriously have trouble looking at a nethack screen? Geez… I know you aren't too young to have played it. :)

That said, I do usually play with colour enabled, as it makes it just a little bit quicker to spot and recognize things. Of course, it is a turn-based game so speed isn't essential, yet like most GOOD turn-based games, it provides enough sense of urgency to make you move faster than you probably should.

Seriously… -5 internets for you until you go telnet to ne... and play a game. Or use their Flash client..., or their Java client<....

Ironically, I've tried to play the game using a semi-graphical version such as Tyche posted, and I don't like it. I think if it were another game that were native I'd be fine with it, but it just seems wrong for nethack/moria/rogue/omega/larn/etc. :)