I'm asking GM's for their own Actual Play examples of good scenario design, for a project I'm doing.  My own contribution is at the bottom.  Project details follow:

I've written several threads about my dearly beloved Alternity science-fiction campaign, which ran for three years, and was a blast.  Alternity is an OOP game with a pretty loyal fanbase.  As a farewell gesture to a game I loved but eventually out-grew, I've been meaning to write up my Actual Campaign as an unofficial setting, and post it to the "official fan site" so all the other folks can see how it's done, and hopefully get some mileage out of the fun ideas my players and I had.

Now, with this campaign setting dealie, it's pretty easy to include all the standard shite that goes into such things.  A few pages on the setting details, a lot more pages about Look & Feel, and (probably too many) pages of rule variants.  I will also probably include some material on Conflict Resolution, since it makes the game much more functional, and play more exciting (IMO).

But the real kicker, I think, is to advise folks on properly designing an engaging "adventure/session/story/fun time."  Like, here are these players sitting around the table with you.  You've agreed to play characters in this whacked-out "transhuman" science-fiction setting.  What Do You Do Next? 

Like, I think it's really important for RPG's to take the GM by the hand and say, "Do X, Y and Z, and presto, it will work."  Alternity does not have this, and I had to kit-bash my own thing from wisdom here at the Forge and on Vincent's site.  So, I'm trying to gather up people's advice, based on What Works In Actual Play, and distill it for people who may never have seen this before.

=====
My Own Stuff from Actual Play--

1.  Everything's got Three Acts.  In the first Act, you present the problem to the players.  By the end of the First Act, they should be intrigued and invested in solving the problem.  In the Second Act, you hammer them with the problem, and ratchet the stakes way up.  In the Third Act, frame something happening that is going to propel this to a conclusion one way or the other--probably, against what the players are aiming for.

2.  Get a cool, futuristic image thing.  If it involves deadpan comedy, insane action sequences, or general weirdness, that's good.  Build one of these into each Act.  (This should be replaced with whatever is genre-appropriate.)

3.  Take some sort of issue, something where there's at least 2-3 legitimate points of view.  Figure out a way to dramatize this issue in a way that you can make it pretty high stakes. 

3A.  Best issues are those implied by the characters' backstories, or through the players' own actions.  "What, you don't like the Archvillain's ideology?  Well, here's a situation where the villain might actually be justified."  "What, your guy has issues involving motherhood?  Well, let's harp on that a little bit."

4.  Come up with a bunch of characters who have taken sides on this issue.  Because it's still up in the air, they're trying to get help.  Enter the player characters.  Locals try to exploit the player characters into resolving the issue in their favor.

5.  Generally speaking, these locals are pretty desperate folks, and will play dirty if opposed.  Also, it will probably turn out that the situation isn't entirely as it first appears.

If all of this sounds like warmed-over Sorcerer or Dogs, that's because it's basically where I got it from.  But if people have additional styles of prep, I'm very interested in hearing about it.

Well, I think you've basically got all my scenario-prep ideas.  I suppose I can throw in the heretical one:  I find that scenarios often work well when I convey a very strong sense of "This is what I expect to happen, but you can change that."  Like, if I say "This will be the story about how these two factions in town finally lost it completely, and everyone died ... unless you make it some other story, of course."

Which is, of course, basically DitV's "What would happen if the Dogs didn't arrive?" cranked up to an explicit technique. But hey, if you're going to steal, steal from the best.

Hi James,

I'd like you to make your example more specific - tell exactly what happened in one instance, when you used these principles. Describe the people and the game, referring to older threads if you'd like, and really dissect out how your principles functioned in application.

Yes, the whole "So Symeera the ex-marine was trapped against the bulkhead with the slime-aliens attacking. Alice was pretty uncomfortable and I wasn't sure what to do, because I didn't want to kill her character (Alice is kind of hot). Bob says, "Bye Symeera" to tease her, making it worse. Then Alice hit Bob on the shoulder, picked up her dice, and rolled. She hit! Bob turned right around from trying to piss her off and cheered. Without my rules mod, the same roll wouldn't have hit." That kind of thing.

Best,
Ron

Hi Ron, I'll try to oblige, but I suspect my answer will be fuzzy.

First, let me clarify my objective in this thread.  Situation is an important part of Exploration in the Big Model, yet only a few RPG's contain explicit advice on how to create juicy Situations.  This means that the GM's for many, many games have to play this by ear, and I'd like to make this easier for my eventual audience by recommending stuff that reliably works.  Thus, I'm hoping folks on the Forge have some handy AP experience.

(And as a contribution to design in general, my advice is: include specific advice about generating good Situations in your game.)

=====
My own experience, based on the Alternity game:

"Blackmail in Bettina Cartel."  Early in the campaign, back when I didn't know what I was doing, really; an example of the Bad Way to GM, perhaps.  I'd had this image of organ tissue transplants being harvested from fetuses, and sold on the biotech black market.  The fetuses, of course, were taken from shanghai'd women who had been memetically reprogrammed into becoming sex workers.  There was a Femme Fatale involved, who could turn invisible, breathed fire, and had a monomolecular cat-o'-nine-tails.  Thus: weird, creepy imagery and a Villain to beat up.  That part of the adventure worked fine: players were suitably creeped out, and had a hard time dealing with the cyber'd up S&M assassin.

The adventure as a whole, however, was a dud, and one that I was pretty disappointed in.  I threw the players into a lame situation: "Your employer wants the McGuffin Device from Mr. X--who's been murdered, and the device is missing!  Guess you'll have to find out what happened to it."  This was completely imposed on the characters.  The McGuffin Device was important each of them individually, but nobody knew what it was, or what its relevance was.

Also: this was the second adventure of the campaign, and I wanted to introduce some characters who I knew would be important later on.  The McGuffin Device was the thing that linked all the players together (though they didn't realize it until the end of the campaign, practically) and it was their first exposure to the Conspiracy, etc.  In effect, though the outcome of the adventure wasn't railroaded, exactly, I had a definite self-created plotline for what this adventure was meant to achieve in the overrall campaign arc, and the players couldn't meaningfully affect that one way or the other.  It was like the opening moves of a chess game.

From a Situation prep standpoint, this was pretty badly done.  "Um, here's this goal that doesn't mean much to you as a player.  Solve it.  And by the way, as the GM I'm not too interested in the outcome."

"You Can't Go Home Again."  Much later in the campaign, after some exposure to the Forge and Vincent's site.  One player had invested a lot playing up that his character was sort of a Luddite policeman, exiled from a backwards space station still using 20th Century tech.  Another player had invested heavily into solving social problems non-violently.  Groovy: social question is, "What's more important: preserving a culture, or helping its inhabitants even if it means abolishing its tenets?"

Situation was created like so - Representing the cultural preservation side, a Police Chief who was heavy-handed, probably corrupt, and who was willing to discreetly kill his political enemies.  This Police Chief was a bad guy, who had helped cover up the PC's exile, and had later conspired to kill his sister.  Representing the side of "Progress" was a Gangster--the same PC's contact and occasional rival.  The Gangster wanted to open up the Luddite Society to foreign investment & modern technology, but this would ruin the society's uniqueness and charm.  And to drag in the other PC, the Gangster proposed that the simplest, easiest way to do this would be to assassinate the Police Chief in cold blood, rather than risking many deaths in a widespread revolt.

Confronted with this situation, the players had to figure out how to remove the Police Chief, while making sure the Gangster's unsavory crowd didn't end up taking over the Society completely.  This ended up being much more fun for me to watch.