20 Sep, ’24

New Blog, New Game

by Cal

Starting a new blog off with a post about a new game idea seems appropriate.

One of my oldest game-dev passions has been the realm of "browser games" - a media-genre which I don't know if there's a specific term for, but it's a game, and you play it with your web browser. Not in your web browser - this is an important distinction! We're talking about games that work like websites. Maybe they have some kind of minigames in them that are more "in the browser", maybe not. Games like Kingdom of Loathing or Fallen London. Maybe including the sibling genre of pet games, like Neopets and Flight Rising as well.

I've also been working a lot with Evennia the past few years, mostly working on a massive project that's now dead but also just playing with the framework to see what it could do. It's very much aimed at MUD-style games, but one of its nice features is that it has a built-in integrated website. And every now and then, I think, "what if I made a browser game in Evennia?" and almost immediately land at "nah, that would be a waste".

But recently, I was sniffing around Improbable Island again (thanks to it having the best banner ad in existence, I love that ad) and idly skimming through their wiki when it struck me.

If I made some kind of a hybrid browser game, like Improbable Island, Evennia would be perfect.

I was immediately taken with the idea, but as I usually do when I have a sudden flash of project inspiration, I put it in the back of my mind to simmer for a while, see if it thickened into something or just evaporated away. The result being a decision that... I needed to do more thinking to nail down an actual concept. And write the things down. And here we are!

So the first step: collect a list of the things I like about browser games.

That gives me a good starting point with my baseline requirements that I'd need to design around and implement. But what I'm thinking about making here is a more hybrid-style browser game, something that is build primarily on those keystone points but has a more real-time game-world environment aspect thingie built in as well. And this is where the really thinky parts come in, because the question here is not just "what do I want to do?", but "what do I like and think would work well?"

For that, I'm going through my personal experiences with games that have similar experiences.

First, and most obviously, we have Improbable Island, the game that set this all off in the first place. It mostly follows the standard Browser Game RPG format, with turn-based combat encounters, towns where you navigate by clicking links in a sort of directory-like context, all that. But it also has a world map, on a grid, which you travel around to explore and get to other cities and have random encounters and stuff. I think this idea is interesting, but it opens up a number of challenges, like how to fill out your world, how to handle the actual navigation, whether or not you can talk to people in the same square as you and how that works.

In contrast, KoL and FL have image-map style maps, where there is a world map marked with zones, and zone maps with locations, and then the locations are explorable with the respective game's encounter mechanics. This is the more browser-game-y implementation of a game world.

The two biggest things that stand out as the differences between the two approaches, to me, are a) the open-world style is more multiplayer-feeling, since you can run into other players, and b) the image-map style is easier to actually play in a web browser using standard website interface things. The first could be a pro or a con or a neutral or a pro-con. The second could be remediable if I put the work into making open-world grid map nav a nice intuitive clickable UI. So that gives me a clearer idea of what criteria I'm looking at, here, but doesn't really get me any closer to a decision. Progress, though.

Another game that came to mind almost immediately when I started thinking about examples was CyberCode Online. This one has that browser-y style game world - although it's just a list of locations you drill down into by clicking, instead of a map, which I definitely don't like as much - but it also has roguelike-style ascii "dungeons". You navigate these in a grid, as well, and there are mobs running around to fight, a boss, loot boxes, all the usual dungeon stuff. And once you go into a dungeon, you can give people basically a party invite code to let them join you in the dungeon and do it as a group.

It's extremely uncommon to have group combat mechanics in a browser-y game - that's one thing that I immediately pinned down as a "hybrid" feature that Evennia as a backing framework could be really useful for. But... I am really not feeling the dungeons idea. Mechanically speaking, it works out well, but mostly because the game in question is a grind game. Which is pretty uncommon in the medium, but more importantly, I don't want to make a grind game.

What did occur to me just as I was thinking through a compare-and-contrast of those two is, how should I put it... the relative scaling factor? In Improbably Island, the grid-nav realtime map is the world map - it's zoomed out. Individual tiles are normal page interactions, even when you can gather there with other places and chat like in cities. On the other hand, in CCO, the grid-nav realtime map is a specific location - it's zoomed in. And while I'm not feeling the procgen loot dungeon... I do like that idea. The navigable real-time maps as the smallest scale of area, with the image-map style for larger scale.

Hm. But then I think about my on-hold Pixel Pets project and how it has a grid-based overworld and I'm on the fence again. I really like the feel of how that game plays out, you know? But it's different, because Pixel Pets is a ZPG - a zero-player game, like e.g. Godville, where you have minimal interactions with your character and their stuff and mostly watch them go around doing things. Which means all the questions of, like, actually navigating that world map are irrelevant, because the game does the navigation itself. And if I did something like that - an auto-walker, to use MUD terminology - what would be the point of it? It adds no gameplay value.

And that brings me to the heart of it, actually. What is the gameplay value I'm trying to add? Fundamentally, for this hybrid-style game idea, I want to add a layer that's more immediately interactive. Being able to step a little more "in the world". And I think.... well, actually, I think that the grid nav map is a red herring. I think I got stuck on it just now because both of the clear examples I had of a hybrid-style game use it.

There's no reason I can't have a more realtime map environment and also use an image-map or directory listing style of navigation. It just needs the specific locations to have slightly more dynamic pages than the basics, like SSE with character arrivals or status changes....

Ah.

Unfortunately, as soon as I started sketching out how this might work, implementation-wise, I came up hard against "this would be a pain in the butt to implement using Evennia". So I guess the next step is digging a bit into implementation possibilities to see if this is worth doing. Whee.