In the grand tradition of IFComp, I have written a 'postmortem'; a short essay in which I reflect on the process of making the game, as well as what I would do better next time! What follows is an abridged and updated version of the post I made on the intfiction forum on November 21st 2018, with spoilers removed:
Beginnings
According the the timestamp on the Inform project folder, I began writing Alias ‘The Magpie’ on 3rd December 2007, almost 11 years ago. For the first few years, it was one of those projects that I would pick up, tinker with for a few weeks, and put down again. For a long while, it seemed as though it would never get off the starting blocks, but gradually, like a cartoon snowball, it grew and gathered momentum.
In August 2017 I suddenly lost my mum, and this put life into perspective for me. I realised that what I really loved doing was writing and creating IF in particular. I threw myself into the game, and over the next year poured hundreds of hours into getting it finished. After listening to the sound of the Spring Thing deadline whoosh past, I set my sights on IF Comp.
Inspiration
The inspiration for the ‘Magpie’ was Sir Charles Litton, aka the ‘Phantom’, from the Pink Panther films. The original idea was that this would be a game about a gentleman thief, but one who had somehow lost or mislaid his thieving equipment, and would be forced to improvise. In the game’s early stages, the ‘Magpie’ was more a physical artiste than a confidence trickster, and the puzzles were to have been very Wallace and Gromit; walking up walls with sink plungers, and that sort of thing. This version bubbled around in my head for a while before the concept settled down to more or less its present form, and I actually began work on the code.
An early influence on the game was the Infocom game Moonmist, but for all the wrong reasons. Though a fine game, Moonmist was one of the first I’d played which felt as though it was on rails. The authors seemed not to have considered that the player might choose to take any action besides the ones that advanced the story in the way they expected. My instinct when playing this sort of game is always to test the boundaries. Hence, when I was invited to dress for dinner, I came downstairs completely naked. When offered a chair at the dining table, I sat on the floor. If I was hoping to provoke a reaction from the NPCs, I was disappointed. The story rumbled on exactly as it would have had I done what was expected of me. When I began work on Alias ‘The Magpie’, I decided that the player should be allowed free reign to do whatever they wanted, and the non-player characters should react accordingly. This is, of course, entirely unrealistic, but what emerged from this attempt was the player character. Sir Rodney Playfair, alias the ‘Magpie’, would be a man who could behave outrageously and explain it away with an airy wave and a glib line. He would have the ability to talk his way out of any situation, and enough charm to get away with murder, figuratively speaking. The 'Magpie' ended up winning the 2018 XYZZY award for best player character.
Setting
Alias ‘The Magpie’ is set at an unspecified time in the twentieth century. I deliberately avoided making references that would ground it in a particular decade, but one or two might have slipped through. Most reviewers have noticed the influence the Pink Panther films, and if I had to put a date on it, I would have to say the early 1960s. Specifically, I wanted it to have a flavour of zany early '60s film comedies, but perhaps one that was set in an earlier decade, such as the '30s. This was the brief that I gave to my cover artist, Mads Weidner, and I think he nailed it.
Structure
So why did Alias ‘The Magpie’ take so long to write? My intention was to create something that played like a traditional, free-roaming parser game, but which also had a complex, 'Wodehousian' plot with disguises, assumed identities, and slapstick comedy. I approached this by basing the game around a number of set-piece scenes, gated by puzzles. In order to maintain a sense of player agency, I was keen that the majority of these puzzles could be completed in any order. The challenge was ensuring that the story made sense no matter what order they played them in. The disguises added another layer of complication, meaning that almost every rule had to have multiple exceptions built in. When it came time for beta-testing, my testers would do things in an order that had never even occurred to me, and I’d have to go back and account for all of those possibilities too.
Originally, the game was to have had multiple endings, including several in which the ‘Magpie’ got caught and sent to jail, but when I came to write them, they just felt wrong. I soon realised that the ‘Magpie’ could never fail. No matter how ‘sticky’ the situation got, there would always be a way for the player to get out of trouble, and thus the game could never be made unwinnable.
Puzzles
In designing the puzzles, I took a lot of inspiration from P. G. Wodehouse. The problems in Wodehouse’s novels are fundamentally about the relationships between people - typically sundered hearts or unwanted engagements - and yet the solutions almost invariably involve doing something physical; pushing someone into a pond, stealing a cow creamer, sneaking into someone’s bedroom and puncturing their hot water bottle. These sorts of actions are a perfect fit for parser-based IF. In Magpie I sort of turned this on its head. The ‘Magpie’ is no do-gooder, he’s out to steal things, and yet as an indirect result of his actions, almost everybody in the story gets exactly what they deserve.
Testing
Final testing took around 9 months. It helped that I was aiming for Spring Thing 2018, which meant that I went hell-for-leather in the latter part of 2017 and got the game into a playable state by December of that year. My beta testers were fantastic. They pried into every nook and cranny of the game, levered it apart with crowbars, and tried everything in every possible order. Feedback came in the form of transcripts, lists of suggestions and even spreadsheets, with every error listed as part of a table. Considering that three of my testers were working on their own IF Comp entries at the time, I wonder how on earth they found the time to be so thorough.
Reflections
Alias 'The Magpie' was a difficult game to make and a massive undertaking, and I would certainly think twice about making another parser game of this size and complexity. However, I'm not done with Sir Rodney Playfair, alias the 'Magpie' - there are more stories to be told about him. Interactive fiction is evolving all the time, and I'm currently looking into TinyQBN, a toolkit for creating quality-based narratives in Twine 2, as a possible approach to writing a sequel. All I have at present is a setting a few characters, and a handful of scattered notes, but hopefully it won't be another 11 years before the 'Magpie' returns...
