The article seems to be blocked, and isn't showing up. That said, regarding improvisation - Improvisation is not in my GMing career, improvisation is my GMing career.

I can't read it from here.

I second Knaight's statement. Virtually everything I do is improvisation, usually around themes and outlines. With a few nearby maps fleshed out in advance.

I run large sandboxes as my GM motif and let the players gather story elements from their own interactions from the world. I often set things in motion in the world at the start of the story, but the players as a general rule are free to ignore it. (It however may very well come around to bite them in the but later.) I typically flesh outlines of the cities and regions players are in and have certain things waiting in the wings. I jot notes on when I have to make up things so I can cohere to them consistently in the future. I find the players often end up writing half the story for me this way.

They can surprise me, and in fact my players most often surprise me and break my sandbox by eventually blowing something up, literally. Which has caused me to introduce a luck mechanism for when my players do something so completely insanely suicidal that the only thing it makes sense for them to roll, to see if they survived, is luck.

That's a lot of good ideas. Sorry about the blog. i wonder if I released it wrong or something.  :\ There's an email for Ann.

It seems to be fixed now. Saw this go out the other day on the Grey Ghost Games twitter. Nice blog post.

Now that I have the article, I can discuss it. As such, I'll be giving a diametrically opposed view on two points. The first is that you should know your setting perfectly - I disagree entirely. You need to know it just well enough, and be able to improvise parts of it. What you want in a setting is gaps you can fill as you need, and a perfectly defined, perfectly known setting can be limiting. Knowing a setting well is only one method, in short - after all, one can improvise entire settings on the fly. This is particularly easy with low depth games in low depth settings, such as Shallow Graves. As for the second point, I don't think that the GM has a story. There is no "your story" as the GM, there is "our story" as a group, and by letting the players drive much of it improvisation becomes that much easier.

Knaight,

I see where you are coming from. I guess this falls under the category of who likes to run their games in what ways. I prefer to have the planning (hence the article. Pretty much the way I like to do it).

Having located and read your article and these forum postings, I got to pondering on "improvisation" and "planning" -- how do they overlap and how do they oppose each other?  My earliest experiences with improv were in a theater group.  As unprepared as some of the mayhem skits drawn from audience suggestions may have seemed, the actors had a premediated structure on at least the order of a Mad-Lib.  But like GM-player interaction, there was also a playfulness (perversity) in some actors (most probably) to see if they could introduce a (pausible) line of dialogue to throw a fellow actor out of character (crack up, shun into non-response).  Cooperation, coordination, and some common knowledge (often on-the-job learning) of the improv story-building techniques amongst the troupe made the improv skit a success, but the fun (for both players and audience) came from the flexibility and the surprise.  Scripted drama has its entertainment value (for the audience, mostly clearly) and its satisfaction for those who produce and perform it.  Perhaps its pathos and its humor are deeper and stronger for all its planning and fine-tuning, but its rigid and repetitive nature are contrary (I opine) to the primary goals of RPG.  RPGs are fundamentally improvisations -- almost entirely by the players (unaware of the GM's secret set of challenges and obstacles) and nearly inevitably by the best GM melding the players' actions into a story to be retold in legends at card tables, conventions, and blogs for years to come.

So, having a thoroughly planned (and memorized) setting and (expected, potential) story may be an approach to players' surprises  (where many of the planned contigencies never get actuate by the choices made by the players).  There are likely GMs who get much enjoyment (and probably multiple adventures) out of the Creator/Author role amongst the necessary GM skill set who would find contentment in the opportunity to prepare for as many surprises as they could conceive.  I think that is more than your article suggested -- you guarantee the GM will be surprised beyond any amount of preparation and offer suggestion for coping and recovering.  Executing a prepared contingecy plan is not improvising (although it might look like it to your players who thought they had thrown you a curve).  As you state, improvising is working with the unanticipated novel elements introduced in mid-stream.  The objective is to integrate them into the ongoing story ... and, most importantly, keep it ongoing.

The suggestion to "know your setting" is most strongly illustrated in the directions to avoid fumbling around in your source materials and rulebooks.  It is not so much that you need eidetic knowledge of your setting as you need to be the authority on your setting.  You are the person who weaves together the preparation and the runtime extensions of the setting and the players need to be ready to accept that.  Your confidence (and ability to tie in "new knowledge" introduced) as the spokesman for the setting helps the players fall into their roles as partners, not competitors, in the RPG process.  Here may be a big difference in GM preferences -- how big of a setting to "prepare" (organize, formalize) and how much to "grow" (listen, seed into imagination/explanation, take notes).  Both preparation and growth of a setting may tax a GM's memory (or filing ability) to produce a satisfactory recall at game time, but the degree to which the GM takes Dirty Harry's advice to know your own limits may help define the precision and accuracy with which the game uses the "known facts".   Perhaps a more reality based game (WW II simulation) might require extensive knowledge of the truth of that reality.  A GM operating in a more fantasy setiing (of his/her own outline) might be less challenged by the truth.  The sparsest set of facts are the most easily amended and extended -- overplanning may hinder improvisation.

The direction to "flesh out the story" is somewhat contradicted by your explanation that follows.  I think Knaight's point that many GMs and players consider it the group's story, not the GM's, is important.  Sometimes fleshing out means sewing another arm onto the Frankenstein monster the group is building.  The GM is the Architect on this project (including the revisions introduced after construction begins) and the principal Motivator (Foreman) when the players lack direction, but, to continue to stretch this analog, you may not be the Buyer of the final product.  Yes, if the players are stalled, the GM has a role to open up some pathways to explore (with perhaps an ulterior motive to get back to prepared ground), but recognizing the players' role is improvisation (and that is their challenge as players, not characters, to play these games) should urge the GM to let them lead when they have the momentum.  A mission based adventure should allow the players to self-assess whether they have lost the focus and aim without the heavy hand of the GM.  A wandering, let's-see-what-happens adventure does not really set the players up to expect a secret agenda from the GM (so it will be they surprised beyond their preparation).

And your final suggestion "Practice" is right on.  It is how the theater group "learned" improv.  The patterns emerge from repeated attempts to extemporize.   In theater improv, stereotypes as a foundation bent with an unusual quality (like John Wayne as an English butler) were often the first step in the few moments the troupe would huddle after the audience solicitation.  The wider the practiced forms differ, the easier it is not to appear too boilerplate and predictable.   The chance encounter tables often published in RPG manuals are examples of "formula" response to aid the GM to focus his/her imagination to a specific within a stereotype.  When the gaseous character attempts to slip pass the ordinary mechanical obstacle too easily for the conflict needed, the GM should consider introducing another challenge (not insurmountable) to the character from a random table.  Will the players cry foul?  Probably if the GM's add-on seems too specifically designed to stop only the highly improbable gaseous assailant that no vault designer would have anticipated.  But something that would coincidentally cause a problem (a live guard/creature inside thwarting the opportunity to solidify to operate the mechanism, hidden/complex interior unlocking system requiring specialized thief/locksmith skills) would be seen as an obstacle any vault cracker might have faced.

Vermonster,

Thank you VERY much for your detailed critique of my article. Such insight is very nice to get as I try to write things fudge related. I agree with all you said and also would like to point out a few things.

When I mentioned fleshing out the story, I guess I was referring to knowing everything necessary about the story so the GM can improvise parts that weren't originally in it. Like going in to caves that were not originally planned. You can then use that to your advantage when you know how your story is supposed to play out. But I can see how this also could fit under "Know your game setting" so I might just have been redundant here.

I really liked how you compared improvization for GMs close to improv in the theatre. Players and GMs are pretty much actors at a table instead of a stage. It helps me clarify in my mind what I was trying to say using that as an allegory. Thank you!

All of you, thanks for taking the time to read the article and giving me your feedback about it. It's greatly greatly greatly appreciated!

Knowing your setting is critically important to successful improvisation. For me, information doesn't help as much as visualization. If I can visualize the setting then I'll be able to interpret it intuitively, without having to memorize a bunch of facts about it. For example, if the characters are camping, I try to relate the environment to various places I have camped in real life. This allows me to instantly fill in all sorts of details - plants, animals, sights, smells, landscape, etc - without ever having to look at the setting book or memorize anything. The same technique applies to NPCs... identify them with a real person you know. A lot of people use movie stars. That doesn't work for me, I don't follow pop culture and don't know them well enough. Instead I use someone I know well, or someone I have read about extensively. IMO the best thing you can do to improve your improv GMing is to go out and experience - in real life - the sorts of things you want to use in the game. Then, relate these to what you're actually doing. They tell aspiring writers to "write what you know"... i say "GM what you know."