Puppetmaster FAQ


  1. Meta
  2. Puzzle
  3. Trout
  4. Spec
  5. PM


Q   When did you start working on this project?
A   We specced out some back-story for the world of AI in the fall of 2000 for a project that never happened.  Work on this game started with a Puppetmaster Cabal on Jan 4 – 7 2001, with the first sites going live on March 8.  We hope you’re impressed by that, because it just about killed us.
Q   How much input did Kubrick have on the Game?  What about Steven Spielberg?

Unfortunately, none of us have ever had the pleasure of meeting Mr. Kubrick.  We did our best to create an experience that he would be proud of.  Both Steven Spielberg and Kathleen Kennedy were enormous supporters of the game.  They moved heaven and earth to get us the materials needed, character information, art samples, and insight into the story.

Without the tremendous support, encouragement, and more importantly trust, of DreamWorks and Warner Bros. none of this would have been possible.

Q   Did you anticipate the formation of a "collective detective"?

We were counting on it.  The premise from Day One was that the entire Internet should be considered as a single player; that we could put an ad in a newspaper in Osaka in the morning and have some kid in Iowa using that information by supper time.

That said, the most visible player groups (you know who you are) did exceed our expectations in their civility (most of the time) and esprit (nearly always).  We feel enormously lucky, and privileged, to have connected with such an extraordinary group of people.

Q   Were you guys on the boards and in the chatroom?  How often did you post there, either to give us a hint or throw us off the scent?

We did read the boards and snoop in the chatrooms.  Point of pride, though:  we have never at any time posted anything story or puzzle related in either venue.  Everything you got, you earned.

Q   How did you feel about players "brute-forcing" the game (e.g. WHOIS lookups, rational hatter, RUR-14, Catskill Patient Files)

It was very important to us not to establish any rules.  One of the things we were very excited about was the blurring between fiction and reality.  This was best achieved by putting material out there and reacting (as best we could) to whatever method you decided to use for a solution.  We were, after all, encouraging you to be resourceful.  So, sometimes you were, um, REALLY resourceful.  In fact, we were so amazed by the distributed client RUR-attack (in spite of it bringing the server to its knees) that we submitted it to the Guinness Book of Records (still pending, as of last report).

Q   What were your back-up plans for when we didn't find or solve a puzzle?

Our initial rule of thumb was, subtle clues on any subject on the first week, more direct clues the second week, and explicit hard info on the third.  But one of the reasons we had to watch the boards was to get a sense of how things were playing.  Sometimes a puzzle stayed unbroken because players were on the wrong track, but sometimes it was just not a strong design on our end.  In either case, we wanted to be able to respond rapidly to where the game was going.

Q   Was there a deliberate bias against Macs?  Netscape?

Not at all.  The frantic pace of this project made it really hard to test everything that we should have.  Often, we’d finish writing pieces only minutes before they had to go live!  We pretty quickly became our web developer’s worst nightmare, but they did the best they could with our impossible schedule, and programmed and tested first on the most common platform (IE).   We learned from our mistakes as we went, but never did anything to intentionally promote one platform or browser over another.


Q   Were you impressed by the speed at which players solved puzzles?  Did we ever solve something so fast it created a problem for you?

YES!!  You guys were frightening!  The Declaration of Independence puzzle was one we thought would take a full day, for instance.  More problematic was solving the Reconstruct puzzle before even getting to the page that asked for the solution (and consequently deciding you hadn't solved it yet.)

As far as genuine trouble, you guys got the original planned Audrey Green/Paco entrance weeks ahead of schedule; as it happened, we hadn't even had time to build the password functionality in yet, so we were saved by the patented Just In Time content delivery schedule.  Ahem.  But we all agreed that (as an empty gesture of fairness) the final password solution would still depend on an Audrey Green birthday reference in honor of the player (ollie on AICN, we think?) who first got what was supposed to be the right answer.

Most incredible of all, to us, was the speed with which certain players put together the Donutech conspiracy angle, and guessed at the role the TP web was to play, by patiently plowing through reams of unrelated material in microscopic detail and somehow separating the little bit of wheat from all that chaff.

After one week, we realized we had to completely recalibrate how we delivered puzzles and information, and try hard not to put out even the appearance of a puzzle before a solution was firmly in place.

As much as anything, it wasn't just the speed with which players solved puzzles that was overwhelming, (though that was a beautiful and awesome thing).  But what we truly hadn't expected was the rapacious appetite for more content of every kind, and the level of scrutiny that would be applied to every chance comment or whimsical graphic.

The up side of this enormous, beady-eyed, voracious player-monster was that less than a week after the A.I. trailer hit the Web, CM and The Trail were our definitive continuity source.  "Read the Trail!" and "Search the board!" were heard around here pretty much as often as they were out there in player land…

Q   Were there puzzles the players were surprisingly slow to solve?

Getting the info out of the picture of Origen.  We clued it, clued it again, then went into week three desperate to get that part of the story out.  We got spoiled by the expectation that nothing would go unsolved or unnoticed for a week, and sometimes let the difficulty knob spin a bit too far.

The flip side of this question is that of puzzles we never had time to test.  The business puzzle and the clay puzzle were unfortunate results of wild experimentation and intense sleep deprivation.  And what the hell kind of last name is "Wheti" anyway?

(Thanks for bearing with us!)

Q   Were there puzzles suggested by player spec?

Sean (story guy) kept the closest watch on the boards; Elan (puzzle guy) hardly ever had time to look.  Consequently, players drove the story more than the puzzle-building.  (Of course, we slowly learned what kind of puzzles players were best at, seemed to like, etc., by watching them get solved, but that's a little different.)

Q   Okay, then, what story elements did the players significantly impact?

Tons.  Players spotting typos were responsible for two entire characters (Jeanine's crabby AI and the elusive Jason Fertors).  Players spotting a re-used stock photo forced us to write The Step-Self thread.

Players also voted with their interests.  The Red King was a random GK flunky as outlined and wasn’t supposed to be mentioned past the first week.  But the web developers threw in a cool sound file on the Shado Paj, the players reacted, and a star was born! (The down side being of course, that we had to rewrite the story each time RK had a part.)

We had a variety of ideas about how the Loki problem might get solved, none of which involved a dream database, but once it was there, what a resource!

Our original plan for the game was quite different from the final product, which was written on the fly (in between marathon bug fixing-sessions and orange juice binging). That wonderfully dynamic interplay is entirely due to the players.

Q   Puzzles where the players were amusingly off-track…?

Guys, it was a photograph of real dominoes.  So don't blame the art dept. for not making all the pips perfectly circular, blame the manufacturer…

Q   What was the hardest puzzle to create?

Not counting Eliza, you mean.  Probably Dripping Tap Enigma and the Beautiful Land (Elan modeling clay straight through the night before the last update Tuesday, Dan driving back into work at 3 AM to bring a digital camera to help make the measurements exact, Sean writing alternate Laia mail in case the whole beat had to be changed, and the puzzle finished at 8:20 am PST …)

The circular aspect of the story was also quite a challenge.  Finding ways to encapsulate multiple story threads, a puzzle solving game, a meta-game, communication channels, and a movie in a way that would all inter-relate and reveal multiple levels of depth was a massive effort.

Q   What did we miss?

Damn little.  There are still a couple of pretty things lying out there, but they tend to be at the meta-level.  You might try taking a look at the Ghaepetto registration addresses, for example.


Q   So what was the deal with the Chatbot?

Chatbot was created by Warner completely independent of our game.  Later, Warner generously offered to crack it open and help us out.  We used it to point people toward the game, but felt that putting real hard clues in it at this late date would probably lead to bloodshed.  


Q   So how did the world end, anyway?


We know from the film that at some point in the future, soon enough that New York is still recognizable, the world will suffer a sudden freezing event.  It will not be a slow glaciation (the skyscrapers are frozen upright, not bulldozed into gravel).  By the time of lockdown, we also know that while there will be many, many robots left for archaeologists of the future, David will be the first one they locate with intact memories of the human era.

Why the world ends is a mystery.  Simple prion-like patterning?  An instability or positive feedback as the TP try to use the sphere as a nervous system? A deliberate act, either of aggression or despair?  A horrendous accident or mistake on the part of an erratic semi-conscious TP, flailing about like a 3 month old still trying to figure out the difference between "me" and "everything else"?

Without offering a definitive explanation, it seems clear that the TP will flash-freeze:  set for maximum reflectivity, and trigger the chemical "cold-pak" function, dropping temperatures significantly (though probably not freezing more than a thin layer of ocean).  It's possible this freeze will propagate at essentially the speed of sound, taking many hours to sweep around the planet; but many effects propagate much faster (just as changes in current travel many times faster than the electrons in a wire) so the effect may be nearly instantaneous.

Here are results from a basic model, taking into account the change in the earth’s reflectivity (“albedo”) and greenhouse gas levels.  The albedo is the critical effect, and at first is due to TP. Later snows raise the overall albedo still more. All temperatures are given in Fahrenheit degrees. While the change may seem a drastic one, it is notable that some scientists believe that the world has, in fact, spent significant periods as such a "snowball earth".


Average Temp (overall)

Temperature Range




Now (2001 AD)


-90 to 120


Mild        (0.5)

Movie (inferred)


-75 to 120



Right after crash

(within 2 weeks)


-150 to 55.


Low        (0.3)

After snow stops

(two years)


-180 to –10


Low        (0.25)

This final position is where things stay for the next 2000 years.


Q   After you finished working on the game each night, what did you most look forward to at home/after work?

What do you mean, "after"?  There isn't much to do between  2 & 6 am

Q   When did you laugh out loud at us?

There was a common mis-perception that we hovered around in ghoulish glee laughing at your misery.  This is mostly untrue.  Not totally untrue, but mostly.  A more common response might be something like "Omigod if they don't get Nancy's phone code soon I'm going to strangle myself with my own tongue!"  We are entertainers, not evaluators:  we wanted you to have fun, not get assigned a puzzle-solving grade.

Your triumphs were our triumphs.  (Of course, your failures could also be our triumphs.  This is why being a PM is such a cool job.)

Now:  to answer the original question, we laughed when you said something funny on purpose (e.g. the priceless revision of the Shatter movie). 

And OK, when someone TROUTED Laia for not labeling her post, that was funny, and when the chatroom discussion hovered on light bulbs for a day, that was hysterical, and the concept that something like troutmakers exists, still makes our sides hurt.

"Lowering global work productivity one person at a time."  We just tear up with pride.

Q   Who ARE you guys?

We’re a pretty small group with eclectic fields of expertise.  Some work for Microsoft, others just signed a contract without knowing what they were getting into.  Ultimately, there was a very small group of us that ever knew all of what was going on.  Everyone else learned how their piece tied into the story only after it went live.  It would be unfair to pay special attention to anyone here, and we don’t want this document to grow too long.  Check out the credits page for the complete list!

Q   If you could do one thing differently, what would it have been?

More time!  We spent so many of our days just minutes ahead of you guys that there was rarely time to just sit and watch (or play test, or any kind of test).  There was so much content that we wanted to include that just didn’t make it because of time constraints. An extra year of pre-production would have done wonders to the game and our social lives.

If you have any questions not covered by the FAQ, or just want to hang out and chat, we will have an electronic get-together at 9 pm EST/ 6 pm PST on Tuesday, July 31. (http://www.zone.msn.com/zzzz/auditorium.asp)