I have just been accepted as the Games/Roleplaying/Free Systems editor for the Open Directory Project. My question is do you think GNS categories could/would be helpful to people looking for games? Currently there is either no divison of games, or they are divided along genre lines. Specifically I think the Universal systems (of which the directory currently has 60+) could be broken down along GNS lines to help narrow searches.

What do people think about the above example and of GNS as an organizational tool as a whole?

-Shane

sounds useful...
especially the universals....
would save the downloading time...only to find a type I wasn't looking for.

Bob McNamee

The problem is that saying a game is G/N or S is shorthand for saying, "This game facilitates mostly G/N or S decisions during game play." There is also a vocal group (on rpg.net, at least) that would be turned off by seeing GNS distinction.

Worse, even if you were to decide well what systems support which mode best, you'll find that the split will be like 10% Gamist, 55% Simulationist, 5% Narrativist, and 30% too incoherent to peg.

What good is that?

Mike

Hi Shane

I'm afraid I'm going to have to agree with the rest who say that GNS would not be a helpful tool for you. It simply isn't that widely understood of accepted. Someday perhaps, but not now.

This means you're stuck with genre lines and then grouping them alphabetically. There are worse things, but not much.

You could try "realistic" "cinematic" and stuff like that, but I personally would advise against that.

Hey,

In theory ... fantastic. In practice, at this time ... ouch.

The main conceptual problem is the whole "facilitates" concept, which is apparently a serious stumbling block for some folks. I've repeatedly tried to convince people that no, I don't have a little queep-machine which classifies either games are people necessarily. However, I can just see a hell of a lot of people thinking that that's exactly what categories in your database are doing.

The real practical problem is incoherence. So many games are incoherent, and the reasons for this are so multifaceted, that I think tons of games would be thrown into the "whatever" category. And that wouldn't be terribly useful.

I will say that I'm flattered and that I wish my notions could be of more help.

Best,
Ron

Yeah, I was mostly just testing the feeling in the air. I don't think it would be great, but I am looking at like (due to my own finds as well) almost 80 games in the Universal section. I can live with putting games into genre distinctions (which I hate) but to just toss The Pool in with 75 other "universal" games seems so rude =)

Any other ideas for catagorizing "universal" systems?

-Shane

Buddha Nature wrote: Yeah, I was mostly just testing the feeling in the air. I don't think it would be great, but I am looking at like (due to my own finds as well) almost 80 games in the Universal section. I can live with putting games into genre distinctions (which I hate) but to just toss The Pool in with 75 other "universal" games seems so rude =)

Any other ideas for catagorizing "universal" systems?

-Shane

Buddha Nature wrote: I can live with putting games into genre distinctions (which I hate) but to just toss The Pool in with 75 other "universal" games seems so rude =)

Any other ideas for catagorizing "universal" systems?

would it be possible/useful to sort them by rules amount Heavy/Light etc and stances they promote? Actor,Author,Director

course thats probably just a slippery slope too....
hmm
Bob McNamee

Hi there,

Using stance classifications might be even more difficult and inaccurate than GNS ones ... the GNS classifications would be, I think, not terribly wrong so much as hard to use or interpret. However, stance is by definition a labile, ephemeral element of play.

It's easy to see where a game does promote a specific stance - overt enjoinders to Actor stance, mechanics that by definition require Author or Director stance. However, when a game text is silent on the matter, it's not that no stance is being promoted, just that which one is left up to the persons in play (or more likely, assumed to be some particular combination).

What I'm saying is that identifying a "true positive" (e.g. Extreme Vengeance requires Director stance to use certain mechanics, which it does) is easy. But that doesn't mean that any and all games can be classified like this, for each and every mechanic or combination of them.

Best,
Ron

How to split up Universal games? Of the games you have listed, what kind of split would put it roughly in half? I think that would have the most utility.

Zak,

I can't really be reading what I'm reading ...

"... what kind of split would put it roughly in half?"

Are you suggesting that the category-variables be chosen in light of what quantitative distribution they will deliver?

That's hair-raising. Shane, please consider that such a choice of variables would be among the worst offenses against rigor.

Best,
Ron

P.S. Anyone who wants to dispute this point with me, contact me off-forum.

I'm thinking of utility + a programmer's sensibilities (I admit that a programmer's solution can be counter-intuitive and bad in the real world), and only for Universal systems because it sounds like a problem area. If you do a Rules-Light/Rules-Heavy split, and you have 10 games in one and 60 in another, your users are going to have trouble going through Rules-Heavy. Unless you further split up Rules-Heavy into sub-categories.

I've been mulling this (again Universal only) over, from rules-light/heavy, to 1-page/full, to partial/complete. They're all too vague. Even if the split was A-K, L-Z, it would be a little easier to navigate.

What about Freeform & Other (what would be the name for non-Freeform)?

Lastly, is this same issue a problem with the Forge's Resource Library? I get frustrated going through the pages of Free RPGs to find something interesting. Then again, I may be impatient.

(And Ron, I agree that a distribution split would be bad, but I see it as a last-resort rather than a never-do)

There are many easily stated (though not often easily determined) information items for universal RGS that I would find useful in selecting games of further interest, but would not be appropriate for grouping into categories. These include things like original edition publication date, number of editions, most recent edition publication date, core rules page count, number of requisites (character stats required of all characters, whether primary or derived), and resolution system granularity. The latter I express as the maximum probability that the minimum possible situational modifier will alter the outcome. It ranges from .01 for some d100 base systems, to 1.0 for karma-based systems. (It's N/A for most freeform systems, but that itself is useful information.) A d20 system is 0.05, 3d6 is 0.13, The Pool is 0.17, WoD Storyteller is 0.16, some dice pool systems are 0.5 or higher.

So to me, more useful than categories, would be a tablular display that I could sort by any of the above criteria.

Zak wrote: Of the games you have listed, what kind of split would put it roughly in half? I think that would have the most utility.

Ron wrote: That's hair-raising... please consider that such a choice of variables would be among the worst offenses against rigor

Zak Arntson wrote: How to split up Universal games? Of the games you have listed, what kind of split would put it roughly in half? I think that would have the most utility.

I would like to be able to cut it down a bit. Once I am done this weekend I think there will be upwards of 80 games in there, and that just strikes me as too much to look though with much ease.

Don't worry Ron, I am looking for a solid way to divy things up, not just one that "works." I think there have been some good ideas so far, but to be honest nothing has struck me as overly useful other than GNS so far. I mean we could go with terms like "transparent" (which I think is still being argued over) and freeform (which I don't even know the meaning of), but do those really help much?

Maybe length would help, I mean it would be objective, but it seems kind of dull--I mean just because a game is 217pgs it doesn't mean it is anymore complete than a 3 pager.

I could fo with extent of depth and complexity, but those are more subjective terms and subjectivity is poo-pooed at ODP.

What about supported vs. unsupported?

I still think GNS is so far the best route (but I am still very tentative as to whether or not I would use it - people could figure out what they meant in the category descriptions if it would help).

-Shane

Le Joueur wrote:

Zak Arntson wrote: How to split up Universal games? Of the games you have listed, what kind of split would put it roughly in half? I think that would have the most utility.

How about splitting them off? Make 'Universal Games' a division unto itself?

Just thinking out loud....

Fang Langford

Perhaps we should construct a forum in wh9ich games are rated, and then we can have endless fun fighting over the ratings. Perhaps not the most elegant of proposals, but it would at least do some clarifying and give people something to look at in terms of wondering how GNS would be applied. Lastly, it provides a material realisation for the theory which has the virtue of making all our tacit assumptions explicit so that, indeed, we can fight about them.

Caveat: I'm still in favour of something like: Bue Planet G3, S 9, N 2; i.e. that any game supports each axis distinctly and relatively (in this sample max was 10).

Secondly, I'm generally in favour of some sort of classification system, some sort of comparative method. on rgfa years ago I did post a sort of "system algebra" so that you could write a whole dice resolution system in a single string. I've forgotten almost all of it, but trying again this is how I would write the WoD mechanic:

Pool[Att+Skill](d10)! >= Dif (2-10)

Pool of dice = Attribute + skill, all d10's, compared individually against difficulty of 1-10.

Obviously this sort of thing would need to generate explicit conventions and short-forms for writing fairly complex procedures, like the die pool above, or more complex systems which invoilve discards. Systems with lots of qualifiers end up with long strings.

well this is kinda like what I have been thinking about. I am now thinking that I should just categorize based on the dice mechanic (or if diceless, say so) like "trait + roll" or "dice pool" or "percentile". Though could also go by character creation - "point based" or "descriptor based".

Thoughts?

-Shane

Speaking as a former marketing bod... *hiss hiss*

Who's your audience? What categories are they likely going to search on?

I suppose I can see some gamers favoring percentiles systems over 3d6 for whatever insane reason, but I wouldn't like to start categorising systems like Deadlands or Earthdawn. Something like the Storyteller system is attribute+skill, but points are also expended from a pool. Character creation is point-based, but there are also descriptors, and occasionally rolls. The categories get murky.

If you don't know, then p'raps allow your audience to make their own categories, with a voting system.

Or use 'keywords' rather than strict genres. Blue Planet becomes 'cyberpunk, western, future, sci-fi' or Puppetland becomes 'descriptor, surreal, puppets' etc. Like the IMDb.

Joe.

Well at the top of the thread I mentioned that I "work" fo the Open Directory Project @ www.dmoz.org. _Tons_ of search engines (chiefly Google Directory) use it. One of the chief tennants is not to have "symlinks" - don't list something in the directory multiple times (ala putting Blue Planet under cyberpunk, western, future, and sci-fi).

Anyway the major problem I am having (as is mentioned earlier in the thread) is with the Universal systems. They have no "genres" so how can you subdivide 80+ of them so that it is easier to find what you want other than alphabetically looking through the list...

-Shane

Buddha Nature wrote: Anyway the major problem I am having (as is mentioned earlier in the thread) is with the Universal systems. They have no "genres" so how can you subdivide 80+ of them so that it is easier to find what you want other than alphabetically looking through the list...

My apologies, I hadn't paid enough attention to your first post. Sorry about that.

Well, categories *will* be a problem with that sort of classification system. Ack. Hmm.

A few years back, I researched 'rules light' systems and found things like Fudge. But I didn't find bump into games like Sorceror, which while possessed of _few_ rules, uses *every* rule. Thus, 'rules light' wasn't an especially helpful category.

You could hope there'll never be more than a few hundred 'Universal' systems, and as proposed earlier, use categories like 'rules light'. Perhaps you could do a little survey on here or RPGnet to see what sort of categories appeal to 'typical' gamers. RPGnet supports polls, IIRC.

Joe.

Shane,

One of the chief tennants is not to have "symlinks" - don't list something in the directory multiple times (ala putting Blue Planet under cyberpunk, western, future, and sci-fi).

The problem you're confronting is one of trying to put a database into a flat file format. With a database, you could categorize the games by their use of Karma/Drama/Fortune mechanics, rich/sketchy setting, rich/sketchy characters, class-based, designer's name, keywords, etc., and let users sort/filter/search by the categories that are important to them. The project is forcing you to choose one set of categories, the "best" set of categories for all users, and it's simply an impossible task. My recommendation is to pick something arbitrary, like alpha categories.

If you could do each game blurb as a separate page, you could implement systematic keywording in the blurbs and provide an overview of your keywords and instructions for how to use Google to search within the site. That would basically get you database functionality. You'd be treating each game as a separate record in a database.

Paul

Hi Shane,

I suggest you go with the dice mechanic idea that you presented above. "When in doubt, stick with alpha taxonomy."

Best,
Ron

Perhaps GNS classification would be useful as an addition to whatever else you've got. Group by genre or alpha taxonomy and whatever, and put things in the descriptions like "primarily Gamist, also a strong Narrative element" so people who understand the model (all three of them :) ) can use that as an additional guide to what they're looking for.