Cross-posted from Overanalysing AIF
Fully half of the entries in this year's Minicomp were made with Twine, confirming that CYOA is now a feature of the AIF landscape. That alone would mean that the Minicomp rules have to be revised, since they were written with only traditional text parser systems in mind. However, as the deadline approached this year, there was also discussion on aifarchive around relaxing the size restrictions. It came to naught, as it was obviously too late to change anything for this year's Minicomp, and will now probably be forgotten about until the deadline for next year's Minicomp is fast approaching. Unless, that is, we start having a robust discussion about the rules now, which is what this post is intended to stimulate.
Let's begin by really playing devil's advocate and asking if the Minicomp should be continued at all. That's not a facetious question, as the Minicomp doesn't provide many concrete benefits to the community. For example, if you compare the total number of games released in the years before and after the Minicomp started, it's pretty clear that the Minicomp has not led to an increase in the overall number of games. It's also clear that the Minicomp isn't doing much to nurture new authors, as only one person who's made their debut in the Minicomp has gone on to release a full-length game. By way of contrast, look at the reception Palmer's Emily demo got. Would it have received the same amount of attention (which led to Emily: Sister Attraction being the most anticipated game of 2014) if it had been entered in the Minicomp and been forced to share the spotlight with several other games? I'm inclined to doubt it. The same goes for Karrek's Study Date, which could have been squeezed into the Minicomp restrictions with a little effort. The counter-argument is that the Minicomp's status as an 'event' has the power to motivate people to try to create games, and there is the possibility that its removal would have a demoralising effect, leading to even fewer games.
If you agree that the Minicomp should continue, the question becomes what should its purpose be? Should it be primarily a competition, with honour and glory to the victor? Is it supposed to encourage new authors? Is it primarily for the enjoyment of players? I get the impression that the current Minicomp is trying to be all of those things simultaneously, which might explain why it's not particularly successful at any of them. Either way, the biggest problem that needs to be addressed is probably the severe drop in participation after 2009 (38 entries in 2006-9, compared with 19 entries for 2010-3). The 2014 Minicomp bucked that trend somewhat, but I'm not sure how comparable it is with past years due to the presence of Twine games for the first time, and because the numbers were padded by the inclusion of incomplete games. Interestingly, the decline more or less coincides with the last time the Minicomp rules were modified (to increase the maximum number of rooms and characters, and to allow discussion during the voting period). It's hard to believe that increased discussion could discourage people from making games, but it's a lot easier to imagine that increasing the maximum size of games has made it more difficult for authors (especially new authors) to complete their entries before the deadline. There have certainly been a number of games in recent years that have had to forgo testing in order to meet the deadline. So there's an argument to be made against the restrictions being relaxed even further.
During the pre-Minicomp discussion, both BBBen and Purple Dragon said that the Minicomp needed some restrictions to stop people entering games of the scale of PAC. That seems a little disingenuous to me since no one was suggesting that there should be no restrictions at all, and even raising or removing the room limit (which was suggested) wouldn't have that effect. More pertinently, they also stated that the intention of the restrictions was to encourage small, self-contained games. If that's the case, then you'd have to say that the current rules haven't been very successful. The prototypical Minicomp game (setup -> obstacle -> reward) still gets made, but they're frequently overshadowed by games that have subverted the rules in ways that allow more content. That's true even of the very first Minicomp, where the winning entry included a combat system and limited graphics. The fact that it's experienced authors who are better able to subvert the rules in this way (and thus more likely to end up dominating any discussion) may also discourage new authors from entering.
The problem with imposing size limits is that there isn't a simple way to measure the size of AIF games. Back in the early days of IF, size was equated with the number of rooms (for example, Level 9 boasted that Snowball had 7,000 rooms, even though 6,800 of them were a colour-coded maze with minimal description). However, in a character-oriented genre like AIF, the number of rooms is a bit meaningless. It's even worse for CYOA games, especially since Once Upon A Dream (written by the person organising the Minicomp at the time) established the precedent that CYOA sequences don't count as rooms. The actual size of AIF games is more closely connected with the number of characters, but it's possible to create a large game even with only two sexually interactive characters and a handful of rooms (as GoblinBoy proved with Second Guest and In Darkness).
For those reasons, I think it makes sense to think of game size in terms of scenes. If you accept that, then the most straightforward way to limit the Minicomp to smaller games would be to restrict entries to one sex scene (in one location with two participants), period. For the avoidance of doubt, anything involving physical contact (including masturbation), even if it's non-interactive, would count as a sex scene. While that probably sounds draconian, it's a standard that the vast majority of past entries would meet. Unfortunately, the minority of games that would have failed that test includes many of the most popular entries, meaning that players would lose out (assuming that those games wouldn't have been made otherwise, which is debatable). It's also true that authors would probably find ways to subvert even this restriction. For example by making the 'sex scene' as large as possible (The Snowstorm from this year's Minicomp is an example of how that could be done), or by having a greater proportion of non-sexual content (although I suspect that audiences would react negatively if the imbalance was too great). However, it would still be more of a limit than the current rules, meaning that the playing field would be more level if that's a consideration. It would also be simpler than the current rules, which is desirable given that nearly every Minicomp I can remember has been preceded by people asking what is and isn't allowed. Finally, it's a restriction that could be applied to CYOA games with equal fairness.
But what if your aim is to nurture new authors? This is not something that the Minicomp has done well in the past, which I suspect is due to the limited amount of feedback that new authors get in the Minicomp. Back when there was no discussion during the voting period they would be lucky to get any. But even though discussion is now allowed, it generally focuses on one or two games (which are more likely to have been written by experienced authors), while the weaker games in the competition (which are more likely to have been written by new authors) get ignored. Since feedback is the only reward that authors receive, it stands to reason that less feedback leads to less desire to write another game. It also can't be very encouraging for new authors who have spent months working on their games to be forced out of the spotlight by a game that a more experienced author whipped up in less than a week. The only way to stop that from happening would be to limit the Minicomp to 'new' authors (i.e. excluding anyone who has released a non-Minicomp game or, less restrictively, excluding anyone who has won the Minicomp in the past). After all, if you're capable of making a complete game in a week then presumably you're capable of doing so at any time and don't need to prove yourself in a Minicomp. Unfortunately, that would have the undoubted effect of reducing the number of entries even further. It could also make the Minicomp less attractive to new authors by turning it into a tallest dwarf contest.
A lot of the issues around 'fairness' could be eliminated simply by removing the voting element of the Minicomp, and making it more of a 'game jam' kind of event (perhaps with a different theme every year). However, I suspect that would lead to the Minicomp losing some of its lustre as an event. Saying that everyone's a winner just for taking part is all very well, but it's not the best motivation. It also wouldn't have much effect on the actual prize that authors receive (feedback).
If you are going to have voting, should it be broken up into different categories? Back when no discussion was allowed, it would be useful for authors to know where their games were weakest and/or strongest since they probably wouldn't get any other feedback. These days, authors theoretically get feedback during the voting period, which should give them (again, theoretically) more detailed and useful information. Having multiple categories can also make it difficult to calculate the overall 'winner', as we saw with this year's Minicomp. Another issue to consider is that the growing popularity of CYOA systems means that ranking games is no longer an apples to apples comparison. What makes a Twine game technically more impressive than a TADS game (or vice versa), for example? Having either a single vote or ranking the games in order of preference would simplify matters greatly. Making voting a simpler process could also have the effect of increasing the number of votes, which I would see as a plus since voting is a gateway to participation in the larger community (or at least it was for me).
When I started writing this post I was in favour of relaxing the Minicomp restrictions somewhat. But, as sometimes happens, the act of writing about it has forced me to consider my reasoning more closely, with the result that I've completely reversed my position and now think that the rules should be more restrictive. Admittedly, part of why I've reached that conclusion is the conviction that authors are better off releasing games outside the Minicomp, but limiting games to a single sex scene should have the effect of making them more focused and more in line with what was originally intended. It would also result in smaller games, which should theoretically result in greater participation since more authors would complete their games before the deadline.
Anyhow, my opinions probably shouldn't be given a great deal of weight, since it's highly unlikely that I will ever run or enter the Minicomp. What I hope that this post will do is stimulate discussion. There's a tendency on the part of the community to forget about the Minicomp for eleven months of the year, with the result that any changes are either made unilaterally by the organizer or, more often, not made at all. What should happen (if the community were more engaged) is that any changes should be hashed out well in advance, so that authors can make their plans accordingly.
Over to you.