Developing characters in play

10 April, 2009 at 9:33 pm (game design, game mastering) (, )

I’m a big fan of starting with fairly sketchy characters (and world) and letting them develop in actual play. Sometimes it works well, sometimes less so. There is one phenomenon that causes a sort of disconnect. The phenomenon is that of pulling out skills or equipment out of thin air or non-defined backgrounds, and doing this exactly when having that skill would be convenient.

To clarify, the problem is not that this is abusive or makes challenges too easy. If merely having a skill is sufficient to make a challenge null, it was poorly designed to start with. Abusive use is likewise a nonproblem.

The problem is that it strains credibility it feels abusive, even when it is not. As a friend said, a set of lockpicks will not simply materialise out of nothing. Still, this sort of thing seemingly happens all the time in whatever media one prefers to consume.

So, a potential solution I’ll be experimenting with: Equipment and skills and so one must be foreshadowed in a different scene before they can be brought to play. This may be the character telling about some related events, or buying equipment (naturally), or player narrating that the character is oiling the daggers he usually keeps hidden.

As a bonus, if there’s a gun in the first scene, someone might very well use it later. Chekov is not assured to be happy, but some movement towards that direction is achieved, I reckon.

Permalink 2 Comments

Lost in translation

28 February, 2008 at 11:20 pm (definition, rpg theory) (, , , , , )

This post is about language. Tangentially related to roleplaying, more so in the end. I don’t have any training in the area, but meddling is fun, so…

There will be three examples used: Written English (you have probably heard of this one before), propositional logic (which can handle such phrases as “p and q if and only if q and p”, “if p, then q”, and usually has the operators not, and, or, if…then, if and only if) and a small indie RPG Risus by S. John Ross.

Defining a language

Language consists of four components: There is  a set of symbols (in case of spoken language, this would have something to do with the kinds of noise one can emit), a set of rules which define a grammar, a set of meanings and a set of relations that combine some meanings with some grammatically correct expressions.

English: Alphabet and other symbols used when writing constitute the set of symbols. Grammar tells that expressions such as “agonagasgjsegls” or  “write .ChA!oS” are not legal. The set of meanings is extremely large. The set of relations is even larger; it tells that such sentences as “Car is red.” can be used as a communication tool.

Propositional logic is an example of a formal language; notably, it is complete in that all grammatical expressions have a well-defined meaning. Set of symbols varies a bit, but may consist of between 2 and arbitrary number of operators and usually a countably infinite set of variables, like {c1, c2, …}. Parentheses are allowed (often). Propositional logic is severely limited, meaning-wise. A given variable can have two values (true or false). A given proposition (e.g. “c1 <=> (c2 or c1)”) may be a tautology (always true), a contradiction (never true) or neither (sometimes true, sometimes not). The relationship between propositions and truth is not terribly complicated, but also not worth writing down here.

Risus is an example of a supplemental language in that it builds upon a given natural language (English, for example) and needs one; it can have no meaning without a base language. Risus uses the symbology of natural languages. It adds new grammatically correct expressions, for example: Blogger (3) has well-defined meaning which involves rolling three dice in given situations.

Actually, I cheat

If I really wanted a good definition, it would have to be recursive so that, for example, words have meaning, words combined into sentences have meaning, and sentences that consitute a story have even more meaning, and so forth. It gets complicated and is not worth it.


The scope of language is all the meanings it could have. Natural languages have large scope; formal logic usually does not.

Supplemental (with regards to a given purpose) language does not have sufficient scope to communicate whatever  the purpose would require. Roleplaying games are generally supplemental to natural languages. Computer games are usually do have sufficient scope to be played; MMORPGs may not.


Translation happens when one is given a (grammatically correct) list of symbols and should retain the meaning while changing the symbols, the grammar, or the relations.

Translations are problematic. There are several reason for this. First and most obvious is the difference in scopes languages can have. The canonical (but wrong) example is the number of words that inuits have for snow of different kinds. If it were true, it would certainly either be impossible or require an exceedingly long list of symbols to make a difference between them. Correct example are not hard to come up with: Propositional logic is an obvious example. It can’t make a difference between, say, “all humans are mortal” and “one human is immortal”, while natural languages and more advanced logical systems (such as predicate logic) can.

Another, though less obvious, problem occurs when symbols are changed. IIRC, Japanese have problems with “l” or “j” or some other letter that is common enough in where I live. If I asked, say, Nakano to write this blog in katakana or hiragana, he would have problems, even if only the symbols were changing. This problem gets really bad in languages where a given symbol means a given word (there is a fancy term for such languages).

I prefer to read books that have been translated through as few languages as possible. Also, reading a book in two languages or changing the language in the middle of a series can be surreal (I did exactly that with the Song of Ice and Fire).

Implications to roleplaying

Some people prefer to create characters in play, others at start, and this is not a true dichotomy. DIP and DAS the styles were called in the r.g.f.a. days, according to what I have heard. Assuming a player who is of the DAS type, that player first builds a character in her head and then tries to translate it to whatever system is being played. Conversions of whatever between games are similar. Some games make this very hard (Burning Wheel is notorios at this with the lifepaths; anything with random chargen likewise). Even the games that try allowing all options, like JAGS or Silvervine, make certain options difficult to impossible. They usually break when trying to create extremely powerful, small, large, or weak beings. Wushu and the like may be an exception in that they don’t make things impossible; rather, Wushu doesn’t tell enough about a character. (There are games that don’t have this “problem”, at least theoretically.)

An exercise for the reader. Two, actually.

Take two game systems that have sufficient detail and are sufficiently different. Try to convert a character from one to the other and see what happens. Try to recreate a fictional character in both and see what happens. Both exercises are pretty illuminating when it comes to what the game rules are about.

Permalink Leave a Comment