Eidolon postmortem

(Mild spoilers throughout.)

First of all, I’d like to thank everyone who took the time to read my long-winded, unwieldy, save-featureless monstrosity of a game-thing. Triple-thanks to those who wrote reviews. It’s encouraging that players were sufficiently compelled by the story to overlook, or at least persevere in spite of, some of the more cumbersome design defects—and not only that, but inspired to rate the game well enough to grant me a spot on the top ten. I’m a bit stunned, frankly, to have placed as well as I did. I thought the reception so far had been pretty lukewarm, and understandably so: I knew the writing style would alienate some readers, and I knew I had not been wholly successful in my attempts to implement exploration and puzzle elements in Twine.

A close look at the ratings does suggest that Eidolon is a bit divisive: scores are spread pretty evenly across the 5-9 range. So, while reassuringly few people seemed to absolutely loathe the game, there were definitely varying degrees of enthusiasm. I suspect that no amount of revision is going to make this thing please everybody. A lot of reviewers found the beginning unnecessarily long and boring, but had fun exploring the dream-castle; others (like my mom) enjoyed the steady atmospheric build-up in the first half but gave up on the puzzles. I think it’s natural that only a small number of players will enjoy both parts equally; still, I do want to preserve that hybrid quality because I think it’s one of the characteristics that makes the experience unique, having those big bubbles of exploration embedded in a more traditional framework.

The reviews have been extremely useful from a formatting/design perspective. A few things I’ve gleaned:

NOTE: The following list is of pretty marginal interest. Feel free to skip it.

  1. The game is way too long to not have a save feature. The ability to bookmark your page would be the best option, I think, since a constant save button in the corner might ruin the minimalist aesthetic I have going on.

  2. All the clicking gets a little laborious, especially in those segments where the text appears sentence by sentence. Originally I hadn’t intended to have so many of those passages, but I ended up using them as a cheap way to regulate the pace and add emphasis; they can be substantially reduced. Each individual passage is limited to a relatively small amount of text because I found, inexplicably, that if I had too many words a scroll bar would appear while the text was fading in and out, even though I never wrote a page so long that scrolling was necessary. I couldn’t figure out what was causing the problem and ended up just abbreviating all of the passages until the scroll bar went away because I found its constant disappearing and reappearing disruptive.

  3. The way the text fades in and out every time you click a link confused some players. Inexperience with Twine was the culprit here. I wanted to use the <> macro to add text to each passage without the whole thing fading out and back in again, but because I was also trying to center the text in the middle of the screen, the first lines would always jump up when I added lines below them, which looked really awkward. This isn’t really an issue I can describe without going into boring detail, sorry. I’m still not sure how to fix it, having not really tinkered with Twine much since learning the basics. I pretty much settled on the look for Eidolon back in March when I started writing and stuck with it.

  4. Another text-related trouble, which I lack the Twine expertise to fix: I need to figure out how to fit the text to different screen sizes. I hoped to avoid downward scrolling, but it’s preferable to forcing some readers to scroll sideways all the time.

  5. Last but certainly not least, general consensus indicates that the puzzles don’t quite work. (The lack of a save feature compounds this problem, as there’s not much you can do if you get stuck.) I have not, alas, resolved how to address this issue yet. Making the puzzles easier and generally more intuitive would help, I think, since the story needs to keep flowing or frustration will infringe on the atmosphere. Players also seemed to want more explicit directives. Having the whole castle area open to exploration from the beginning might have been a poor decision, since the puzzles can only be completed in a prescribed order anyway. I may need to start from scratch and rework this whole section. Basically, I am open to suggestion on the puzzles.


With that out of the way, let’s talk about writing. Eidolon grew out of an aborted short story, a less overtly fantastical sketch about a girl who has a vision of a hole in the sky which seems to embody the mysterious sense of loss she feels, and at least one reviewer has suggested it might work better as a static piece. I do think the fiction elements are more refined than the interactive elements at present, if only because I have been writing static fiction a lot longer than I have been writing interactive fiction. But the story is more or less wed to its format now: I don’t think I could transplant it to a new medium unscathed. At the very least the investigative aspect would be lost; I like that the player must uncover clues and piece them together to gain a deeper understanding of the protagonist’s internal conflicts, since, at the outset, the protagonist herself does not deeply understand her internal conflicts. Interactive fiction allows me to simulate her journey of discovery in a way that static fiction does not.

The use of investigation as a storytelling technique means that a significant portion of the story is implicit and can be ignored, as the majority of reviewers seem to have done! But that’s alright. I tried very hard to craft a piece of fiction that would reward revisiting; I wanted players to be caught up in the plot the first time around and make new discoveries on subsequent play-throughs. I only hope I haven’t made my hidden meanings so obscure that no one will be inspired to go looking for them. I know the game’s inconvenient design and verbosity do not exactly encourage replays. But anyway, hopefully I do not sound too pretentious when I say there are a number of subtextual layers to the story that have not really been examined in detail yet. (Seriously guys, my game’s soooo deep.)

I should clarify that I am totally not the sole possessor of the “correct” interpretation of Eidolon. Symbols are only fun for me insofar as they are fluid. A lot of the images in the text I chose because their meanings seemed especially malleable, vulnerable to the influence of each player’s matrix of personal associations. When I talk about “hidden meanings” and “subtextual layers,” I am really just saying I haven’t seen a coherent interpretation put forth that expresses the bulk of the ideas I had floating around in my head when I wrote the thing. Which is not to say that reviewers haven’t continually impressed me with their insights, just that so far those insights have been fairly fragmented and nonspecific. No one has attempted to answer the question of the shadow girl’s identity, for example, beyond some vague speculation that she might be the protagonist’s mirror image come to life. I actually began writing under the assumption that she was the ghost of the protagonist’s twin sister, who died when they were very young. There are clues strewn all over pointing to this particular interpretation of the character, though as the story progressed, she came to represent many things.

Reaction to the primary NPC was…mixed. Reviewers described her variously as “a construct more than a person” and “an asshole.” Neither assessment is inaccurate, though I feel they both miss the point a bit. Her capriciousness is intentional; you might say she is emblematic of the sort of erratic and passive-aggressive relationship the protagonist has come to fear. I didn’t want the player to ever trust or feel safe around her, and I think making her more likable would diminish the impact of the ending. The ending does complicate things and forced me to think a great deal about where to draw the line with the shadow girl’s cruelty. I didn’t want to make her behavior so outrageously abusive that the protagonist’s decision to help her in the end seemed unhealthy, like a manifestation of Stockholm syndrome, but I also didn’t want to shy away from the realities of physical violence or severely damaged self-esteem. In the series of conversations you have in the throne room I tried to make the shadow girl gradually more vulnerable and the protagonist increasingly assertive so that when you rescue her you appear to do so out of empathy rather than subservience.

The somewhat highfalutin’ writing style was inspired by the likes of Angela Carter, A.S. Byatt, Karen Blixen, and Joanna Newsom, except I am not as good at writing as any of them. In my initial conception the game was to have taken place over several days and nights, with the daytime segments being only minimally interactive and the nighttime segments more exploratory; I was going to further reinforce this duality by having the daytime prose be very flat and devoid of description while at night the language would flower into a more poetic and elaborate mode. The plot was eventually compressed into a single night and the writing accordingly went full-on lyrical. Description and metaphor don’t come very naturally to me, but the atmosphere I wanted to convey seemed to demand an elevated style with lots of sensory detail. Plus, I wanted to give players some incentive to click all of those goddamn links, so I figured the descriptions had to be pretty good. It was a sort of challenge to myself as a writer, maintaining a style outside of my comfort zone over the course of 30,000 words. There are probably a few metaphorical flourishes I could safely excise, but I don’t intend to radically alter the tone of the text I’ve written so far.

So, looking to the future. I received an intriguing suggestion to expand the denouement in the forest into a third act with more exploration. I kind of like the idea, although I think for pacing reasons I would want to keep it briefer and more contained than the previous two exploration segments. If I do develop that part of the story further, I may try to make some of the buried themes more explicit there. Otherwise (excluding the puzzles, which I do want to extensively revise) I haven’t thought of any major structural changes I’d like to make for the final version. Nor do I have a timeline, yet, for releasing that final version. The laptop on which the source file is stored is languishing in the shop awaiting repairs at the moment. Tentatively, I will try to get the revision process started by the end of the month, and shoot for a final release early next year.

Entering this competition has been a hugely valuable experience for me and I definitely intend to participate again. Congratulations to everyone who entered, and thanks for reading this long-winded, unwieldy monstrosity of a postmortem.

Wow! Thanks for sharing this. My guess was that you’d be in the mix for the top ten, so good job making it and then some. I’d be interested in an expansion/sequel, as well.

Given your laptop problems (I hope it’s OK,) my techy side wonders if there is a definitive way to reverse-engineer twine code from HTML, and if so, if there’s a program.

There was one about. If not, there’s always cut and paste.
Let me know if you need any help.

I really enjoyed the puzzles, myself, so I don’t have much to suggest except that if we could solve them in any order – have the girl demand we bring her three things at the start, that might help it seem more flexible, but would probably add a lot of coding complexity. But I like big open exploration areas.

One thing I noticed reading reviews (and playing through multiple times) was that the intro sequence had a lot of flexibility – I’m a bit curious how that worked. The effect was very cool IMHO – I did entirely different things before the shift in different play-throughs. But some people’s reviews indicated getting stuck a lot before the shift happened; I wonder if a little more nudging could be put in in places where people hung up.

As for a save button – people naturally filter out things they expect and that remain constant – a save/reset frame by the side will pretty much instantly fade to the background and go unnoticed unless needed.

Seconded on the “do the puzzles in any order” thing. That should help alleviate the “some stuff randomly doesn’t actually do anything until the right time arrives” problem.

I’m not 100% behind the puzzles otherwise as is, but I’m not certain what tweaks I’d do yet. If you could make a version that has a save game feature so I could actually test things out in the middle section I might be more helpful.

I’m really glad this worked for you. I was kind of hoping some people would miss the downstairs area at first and discover it later. I made that whole section optional so that people playing a second time could skip it and get on with the story if they so chose. The actual coding involved is pretty simple: Once you have used the bathroom, the whole house opens up for exploration and the hole appears in the sky if you look out the window; once you’ve seen the hole in the sky, going back to your bedroom triggers the continuation of the story. You can pretty much do whatever you want between using the bathroom and returning to your bedroom: explore the downstairs region to any extent or not at all.

I think some people ended up wandering around downstairs for a long time trying to trigger something to happen and didn’t realize they needed to go back to the bedroom. I do have a line (when you’re in the living room, I think) suggesting that the player go back to the bedroom, and I don’t know if I want to be more insistent than that. I tried to achieve two things with the downstairs area: I wanted to drop lots of little hints about the protagonist’s family situation (nearly everything you can click on down there is a clue of some kind), and to build up a really solid foundation of reality before moving into the supernatural. So when you make toast, it’s no more earth-shattering than making toast (typically) is in real life. There is no deep significance or consequence to the act. Same thing when you do your homework: your answers don’t actually have any bearing on how things play out. I just wanted to have little extraneous details in there to suggest the outlines of a life extending beyond the borders of the story. I think I was afraid of losing people when things started getting really abstract and weird, and wanted to have something a little more concrete and universal for players to cling to as a kind of life raft during the surreal stuff later on. Whether or not I succeeded in that, I don’t know.

What was I talking about again? Right, I think I was trying to justify the aimlessness of the exploration in your house. But yes, I might need to insert stronger and more frequent hints to head back to the bedroom so people don’t get frustrated down there waiting for something to happen.

I do like the idea of allowing the player to tackle the puzzles in any order. But I know some reviewers (like Emily Short) have already expressed that they felt like the game was not giving them enough direction, and I wonder if having the girl tell you, “Find me three things,” or something similarly vague would only exacerbate that problem. Is there a way to be a little more specific about the items but still allow the player to gather them in any order? Hmm.

As for the “some stuff randomly doesn’t actually do anything until the right time arrives” issue, can you think of instances where this was particularly glaring and not remotely justified? I can only think of one place where I actually had a good reason: you can’t take the flower in the hallway because you have to wait for it to bloom. (The flowers are not static: they age gradually in reverse.)

I have a lot going on at the moment, but as soon as I can get a version with a save feature ready, I’ll let you know. Also, if anyone would like to test future versions, maybe shoot me a PM or an email. I’ll add a credits page and make sure your name gets on there.

It sounds like the data is safe; it’s just a motherboard issue. Unfortunately I bought the laptop while I was living in a foreign country and since then have returned to the US, and the repair shop can’t find the right motherboard to make the replacement. I think I may buy a new laptop on or around Black Friday and have the data transferred. If you find a program like the one you describe, though, let me know, since I’m sure something like that would come in handy some day.