This is something I didn't plan on doing; I've been developing my game Eleanor's Dream on and off since the beginning of -05, and while I've been chatting about it privately with everybody who cared to listen (read: Forge people I've come to know personally), I never had any great intent to bring my game to the Forge in public. In part this is because I'm used to designing without public discourse, in part because I didn't feel that the game would be served by gamer feedback and in part because I do much of my design work in Finnish.

However, lately there's been a lot of threads about designing games for children, which is exactly what I'm doing with Eleanor. Particularly the threads Magic Backpack and Great Storybook, happening so close to each other, got me thinking that perhaps I was unjustified in thinking that my game's not of interest here. I should think that the design would be interesting for other people grappling with mainstream games for children, at least.

Eleanor's Dream is a game with several design and publishing goals:

• Completely mainstream, with no explicit ties to existing roleplaying culture; probably called "adventure game" or "storytelling game"
• Directed at children aged 6-12 and their parents
• Marketed and sold through mainstream bookstores

The game is currently pretty close to taking form, so I can write a bit about what it will look like: a coloring book -like large softcover, around 100-150 pages long, low price point for a roleplaying game. The process of play involves writing and drawing in the book (as well as coloring it's illustrations of the mood strikes), so the book is very much a tool of play, not just a rule-book. I'm hoping to publish the Finnish version around spring, and if all goes well, I might have an English version for Gencon, even. No idea if it's a good idea to bring a game like this it Gencon, though.

Now, among other projects I've been doing design and writing my material into English; a couple of weeks ago I finally got together a pretty coherent English playtest version of my game. It's fully playable, a "beta test" version, and available for anybody interested in taking a look:
The rules
The dream book
I've honed some details after putting those up, but for the main that's the current state of the design. My next plan after getting the Essen bruhaha over with is to retranslate the dream book into Finnish and embark on a series of playtests with actual children, of which there are some samples in my current environs. I expect that I'll have to streamline some rules and perhaps redesign some subsystems at that point, but if all goes well, the core systems are already all-excellent.

Why am I writing about this in Playtest? Well, because this is anything but first thoughts, and doesn't involve an organized endeavour ;) But apart from that, I've been playtesting the game now and then for the last nine months, so I already have some hands-on experience with how the game runs. My experience has been solely with adult gamers, however, so it's not the whole truth. I'm of the school of thought that children's culture is just adult culture with simpler words, though, so my first priority is to make a fun and streamlined game and worry about it's child-suitability factor later.

My last playtest, however, was before implementing the current system for large conflicts, so this crucial bit of design is not yet tested at all. I'm confident that it'll work out, though, as I've known for a long while what the system in question needs to do, and have been using various place-holder systems while trying to figure out the exact form the system needs to take. We'll see how it flies in actuality when I get time to do that translation and run some playtests.

What I require of the Forge: well, if anybody wants to playtest Eleanor (with children or adults, gamers or not), I welcome the help. Other than that I feel that the game is well in hand, I'm feeling almost professional about it's development at this point. Of course comments, critique and questions are appreciated. Primarily I hope that examining the design proves inspiring for others with similar projects, so I'm putting up the files as a reference.

Hi Eero,

I'd like to know more about the playtests and how you used them to continue work on the game. For instance, what features of play were conceived, written, playtested, and confirmed without modification? As opposed to features that playtesting showed would not work as planned.

Best, Ron

Ah, yes, that's my intented topic here. Just didn't get around to it yet, what with the Essen stuff constantly leaching my time.

My first actual playtest of the game happened in January with two players, my elder brother Markku and a good friend called Sami. At this point I had the dream book interactions pretty much figured out; there was some extra clumsiness in the format of the "stat block", but the essence was done. The resource system was essentially the same it is now, that's one of the first things I created for the game. What I didn't have at that point is pretty illustrative of one way of creating a game, I think: I didn't know how the game was supposed to be gamemastered (read: situations created, driven to head, interacted with player) and conflicts resolved. What I knew was the type of interaction I wanted, and the kind of fictionary matter created.

At this point I also had some confusing design level package that the playtest helped get rid of; the conflict system is a pretty good example, it worked on the principle of players listing "situational advantages" for themselves in a given task. And yes, it was task-based. (Now I'm sure a reader who doesn't know my design history will think that this is another one of those "gee, conflict resolution sure is neat" confessionals; however, I knew perfectly well what I was trying to do, opting for an old-style subsumed conflict resolution, even if it didn't work out.) The playtest was a great help in proving that the thing didn't fucking work, so I promptly threw the system out, spent a couple of months trying out alternatives, and then found what I needed in the decision point system the game currently uses (yes, I moved some design functions from the conflict system to the general interaction system).

The playtest was ran at a small boardgaming convention through the day, alongside boardgaming events we each had in our schedules. We must have played some 6 or 8 hours in total, in addition to talking about the game a lot; the lone January playtest was certainly an important phase in the early development process. The "character creation" system the game still uses worked and works exactly as it's supposed to, which is to say, the game doesn't require any kind of character creation as it's all done in play. However, unlike the current version, this system had a significant extra step in the scene framing: after determining the dream element where the dream would start (locations at this playtest, only), the player could state a "dream self" for himself, which could be some kind of human role, an animal, or whatever else. In the current rules this privilege is moved to the GM; it's a good idea with value, but I ultimately dislike it's chaotic influence in the playtests this far.

Anyway, the fiction-content itself in the January playtest was at the beginning pretty close to what I'm looking for, let's look at what Sami had in his first couple of dreams:
- Found himself on the Windy Plain just as snow was starting to come down. Was a traveller. Climbed on a big rock to perhaps see into some other place. Heard voices, and waited patiently. Then the Horse-Horde rode in and the dream was ended.
- At the Trollcliffs during the day. Was a big bear. Went into his cave to catch a nap, but stumbled on a troll who wanted to live in the same cave. Fought with the troll and killed it, then went back into the cave. The cave had been renovated into a trollish school class room, and the pupils were pretty angry at the bear who killed their teacher. They ripped the bear apart and ate him.
That kind of stuff. Even with an essentially old-skool GM-player relationship the dream token economy and the modular dream book achieved much of the feel and simplicity I was looking for; when the player got more dream tokens, the fast beginning dreams gave way to more elaborate stories. We got to six or eight dream tokens or so during the day, and went through some pretty nice stories about swamp monsters and troll children.

However, there were two places where the system of the time failed miserably, one of them being the conflict resolution mechanic I mentioned above. I hadn't played anything with old-style task resolution for a couple of years, and man it's tiring! After a couple of hours of play the players were still keen to go on, but I myself had started hating the conflict resolution system. After approximately four hours I was convinced that no fucking way was I using a task resolution system in this game, it was arbitrary as all hell. And dull, too, it was completely inane to have to constantly make thrilling decisions like "if I am a kite in the wind, do I get a penalty for this kind of wind?" So I had to scratch a pretty major part of the early design agenda, which involved some vague theory about setting involvement and stuff (we discussed this in Stockholm, if you remember, Ron). That's good, because after pondering it for half a year I think I now have a rather powerful conflict resolution system.

But, the other place where the system failed was and is rather more relevant for the state of the design. As can be seen in the documents, the game has a loosely systematical campaign arc in the system of "Princes" and the infrequent interaction they have with the player. In the January playtest this system was more or less in place, even if it was still a bit rough. The kind of play we got when we threw Markku in this mix was so strange that I've been forced to consider it an anomalous result which I've been promptly ignoring.

A part of the color of the game is that the Princes of the dreamland can bestow names upon those who serve them. Now, Markku's reaction to this was interesting: he refused to have anything to do with any servitude involved and spat upon the idea of being named by a supposedly higher being; he would name himself, or not be named at all. As can be seen in the system, this pretty much means not progressing anywhere from phase 1. Why was Markku acting like an ass, then? Because the game didn't have roles, is why! As can be seen in the current version as well, technically there is no role-creation phase in this "roleplaying" game at all; the players play themselves. Now, Markku doesn't have what I'd call the most balanced scales on the wagon, but still his reaction was interesting: because he was playing himself, he would take it upon himself to not adventure, but rather start farming, because that's what he wants to do. The lack of role-creation interestingly enough was interpreted by him as an opportunity to air his political and philosophical convinctions, when in normal roleplaying games he routinely subsumes these aspects of his self in favor of "being a good chap" and playing as he's expected to play by others.

Well, so it went. The game got certainly intensive, even if a bit chaotic, when Markku decided to start purging the dreamland of non-correct politics. As said, I found this reaction to the idea of "playing yourself" so out there that I've been mostly ignoring it as an extreme case, but it did inspire me to hone and clarify the rules concerning the Principalities. It's easy to see from the dream book that we've been discussing libetarian politics with Markku quite a bit during the last couple of years; I included the Grim Captain, one of the Princes, specifically in an effort to try how the game works with some political allegory.

One small point from the playtest: the January version of the rules had pretty bad down-time, because player turns tended to get longer with time, and players quickly got adept with avoiding turn ending.

Anyway, to sum it up, here's the rules that have been changed since the January playtest:
- Conflict resolution: it was obvious from the playtest that even if it did technically "work" in the sense of getting conflicts resolved, it was a horribly dull and pointless experience to utilize.
- Turn structure: at the time I had no idea I'd be including something like the decision point system, but in hindsight it's exactly what the game needs to keep focused and allow for a bit more meaningful interaction between players.
- Dream forms: I still like the idea of players being able to play bears or angels or whatever, but in practice I find that I don't much like the kind of constrained fiction it results in. Players are adept in pre-setting their conflicts with comprehensive roles, so that whatever you throw at them, they don't particularly need to think, they just go with the role. Much nicer when you have to consiciously fight to create fictional identity.
Other changes have been minor clarifications and such, not really that important in the big scene. Notably the manner of dream book interaction has been a core idea that hasn't been revised in the playtest process.

Later playtests: I playtested the game a bit later in the spring, but mostly they were short putterings with the intent of getting the conflict resolution system into shape, and quickly ended when I got whatever info I needed. Only after getting the decision point system up I arranged another extensive playtest. There's been two of these post-decision point playtests thus far and the results have been very encouraging; both however used a place-holder conflict resolution mechanic, so I'll have to arrange a new playtest soon just to find out how my current conflict resolution works. I'll try to write a bit about the summer playtest at some point, but now I have to wrap this up and go back to figuring out Universalis for Essen.

A diversionary note about the resource system: there is a minority school of thought (read: one smart guy) claiming that the resource system of dream tokens and awareness tokens is unnecessarily elaborate in the game as it stands, considering the target audience. I kinda agree on a philosophical level in the sense that while the basic idea of how you get awareness tokens is pretty nice, the actual ways you use them are a very organic shamble; it's just a long list of things I find necessary or fun to be doing. Not very analytic, and probably the list will be revised heavily based on playtest. A number of streamlined variants have been discussed, including one where dream tokens are a common pool for the whole group (so all dreams end at the same time), the essential awareness functions are paid for with dream tokens (so using them actually shortens the dream) and the rest of the awareness functions are scrapped or made free.

Another diversionary note about the decision point system: I should give mad props here to a Finnish peep, Ari-Pekka Lappi. He's one of those keen rpg theorists you'd have called (and perhaps still could call) a Nordic theorist a year or two ago. Lately he's been playtesting and designing crazy "Forge-style" (for the lack of better word) systems with a very unfocused method, but lots of cool ideas. I got the whole idea of structured decision-making mechanics from one of his games. Took me a couple of weeks after playtesting his game to realize that it was exactly what Eleanor needed, but there it is now. I find it very gratifying that we are slowly starting to have what could be considered a kind of a designer community here in Finland, with people who I can actually discuss my game design with meaningfully. Been a lonely decade, but it's getting better!