Sunday, August 13, 2017


Part seven in a weekly(📖) devlog.

Standard development disclaimers apply. This is pre-pre-pre-Alpha content, everything is subject to change, features may not be present in the final version, there's a very strong chance none of this will ever be released, etc. etc.

There shouldn't be any plot spoilers in these posts, but there will be occasional discussions related to characters, locations, mechanics, and other aspects of my potential upcoming Shadowrun campaign (tentatively titled "CalFree in Chains"). You may wish to skip them if you'd like to be completely surprised.

I chuckled a couple of weeks ago when this screenshot popped up on my twitter feed:

I remember running into that dialogue during my original playthrough of Dead Man's Switch back in 2013 and doing a double-take, then a triple-take. I'd assumed that it would have been patched by now, but apparently it's still alive, well, and astounding new players four years later.

I've definitely written dialogue that's at least as bad, but I'm also pretty sure that there isn't currently anything like that in Antumbra Saga or The Caldecott Caper. The secret? Proofreading, editing, and revising!


After I've written all my dialogue in Google Docs, I sit down and read it all back. It would be much better if I had another person to proofread and edit for me. My favorite studios like BioWare and Failbetter Games have small armies of editors, which is a big part of why their writing is so good: a professional outside perspective will be much better at identifying weak points and pushing a writer to become better.

Still, I think I do a decent job at self-proofreading. It helps that, because the writing itself takes me so long, by the time I circle back around to re-read it several months have passed, so the words feel fairly fresh to me. If I re-read a sentence immediately after writing it, I'll probably glide over any rough spots, but if I've waited a while, I'll see the problems with it.

Typically, the corrections I make during this phase have to do with spelling, grammar, and punctuation, fairly simple mechanical stuff. There's built-in spellchecking in Google Docs and (now) in the Shadowrun conversation editor, but they won't catch problems like homophones or misspellings that are also words ("through" for "though" or "throw", for example).

I'll also do some general tightening up of the writing. Since about 95% of my text is dialogue, I don't have all that much passive voice to begin with, but if I run across any I'll see if I can fix it. In general, things get better when they get shorter, so I'll try and snip out unnecessary transitions, redundant information, or other space-filling words.

Finally, I look at formatting. My natural inclination is to write one paragraph for each character's line, but in practice it can "read" better if it's split into multiple paragraphs, adding more emphasis and a sense of the character's voice. For example:

The only problem is, a bunch of pacifist deckers aren’t particularly well-equipped to lead that fight. We’re retraining and getting better, but in the meantime, we rely on professionals. That means shadowrunners. That means you.

That can be restructured to:

The only problem is, a bunch of pacifist deckers aren’t particularly well-equipped to lead that fight. We’re retraining and getting better, but in the meantime, we rely on professionals.

That means shadowrunners.

That means you.

Same content, but I think the second version is a bit more dramatic when displayed in the dialogue window. (Even more dramatic if you use a separate node for each line, but I think that would be overkill here.)


I typically do this at the same time as proofreading, but it's really a distinct activity. Proofreading focuses on the narrow technical aspects of the writing. Editing is a higher-level activity, looking at the overall flow of a conversation and the information it imparts. Proofreading problems are easily spotted and have straightforward solutions. Editing issues are more subjective and harder to quantify. Some examples of editing issues I address are:

Character voice

When I first start work on a story, I'll have a general idea of who a person is and how they "speak". Even without voice acting, there are a lot of tools at my disposal to differentiate major characters: How much they swear, under what circumstances they swear, how big their vocabulary is, how likely they are to use contractions, unique verbal tics, and so on.

For example, in The Caldecott Caper, Sable is very excited! He speaks with many exclamation points! He has a splendid vocabulary, astounding the varied intrepid personages surrounding himself. Persi swears frequently and speaks in dialect.

As I write, I'll get to "know" each character's voice better; as a result, when I circle back around and re-read the early part of a story, I'll often come to find that the voice has "drifted", so I'll try and update it to be more consistent. The biggest example of this kind of editing was probably with Dorbi in Antumbra Saga: in my first version, she "sounded" too much like Hailey, so I pushed her further and further to her extremes (megalomania, cheerfully violent), and then needed to update her voice to match.


Does this conversation make sense in the context of what the player is doing? A long meditation on the philosophy behind an organization's cause may make sense if you're chilling in your base between runs, but not so much if you're currently under fire.

Is this conversation fun to read? I try to avoid "lore dumps" where you're just clicking through pages of exposition. If a conversation is veering too far in that direction, I might try to break it up into multiple dialogues separated over time, or just cut out stuff that isn't germane to the plot. I also try to make it somewhat interactive, allowing the player to react to and comment on the knowledge they're receiving.


Much like character voices, the plot can shift and drift as I write it. So I may discover that I was foreshadowing stuff that was later cut from the plot, or am referring to an event that no longer occurs; or conversely, that characters aren't acknowledging a major new development that was added later.

This all gets more complicated due to the non-linear nature of game design. Sometimes I'll need to make dialogue changes, but other times I might need to make sure I have appropriate state flags and checks set up so it makes sense for the player's current situation. That might mean having two different versions of a conversation depending on when it occurs and what previous decisions the player has made.

This is also a good time to double-check that I'm within bounds on the official lore. I'm not too obsessive about this, but I know that some players are taken out of the story if it seems like I'm contradicting Shadowrun canon, so I try and complete my due diligence to make sure that I'm conforming. If I'm not, there are several ways I can respond. I can just cut the dialogue, or update it to be correct, or even lampshade it to acknowledge the problem in-universe.


I tend to think of proofreading as operating at the level of letters and words, and editing as operating at the level of sentences and paragraphs. Revision is a level above that, of looking at entire chains of conversations and thematic aspects of the story. Revision can happen before, during, or after the editing process as I discover major issues and work to address them.

Sometimes this is due to a technical problem that needs to be solved, similar to what I described in Editing. If a character mentions an event that no longer occurs, it's a simple edit to remove that reference. But if there's an entire quest built around that event, all related conversations need to be changed.

Personally, my biggest revisions tend to be more about items that I'm aesthetically unhappy with, due to the tone or subject matter of something I've written. I mentioned before about the major revision I did to the submarine quest in The Caldecott Caper, where I completely threw out the initial squicky storyline and made a new one from scratch. The original version was technically correct, but not something I wanted in my game.

I've hit a couple of decent-sized revisions so far in CalFree in Chains. One had to do with the chronology of the game. I had initially planned to start the game several months after a major political event, and wrote the opening scenes with the assumption that the player character was already familiar with the event and the fallout from it. However, I felt like the opening was too slow and not very engaging, so I ended up deciding to move up the timeline so the major political event is happening simultaneously with that opening scene. This adds a much greater urgency to the scene, and stronger motivations for all the characters in it, but also required rewriting a lot of the early dialogue to reflect the updated setting.

I also had a semi-comical experience a few weeks ago. I was watching a totally unrelated movie, and thought, "Ugh, I hate it when [bad thing] happens to [type of character]. That happens in WAY too many movies, TV shows, and video games." Then, a few days later, it suddenly occurred to me: "Wait a minute! [Bad thing] can happen to up to FOUR [type of character]s in CalFree in Chains! Aaaaagh!" Now, some of those events are contingent on decisions the player can make, so I'm not as concerned about those, but one of them in particular struck me in hindsight as gratuitous. Worse, I realized, I was doing it to make the character "darker," which is exactly the kind of thing that I complain about all the time on this blog! It's kind of mind-blowing that I would just unthinkingly echo a trope like that.

The good news, though, is that I caught it well in advance of the game going out, which is plenty of time to address it. I'd initially thought that I would need to rewrite an entire arc, but now that I'm reading back through it, I think it should be fine so long as I change the end of that story to give a happier ending.

Soliciting Corrections

I don't mean to imply that my campaigns shipped with perfect dialogue. I'd guesstimate that, on each pass through a section of text, I fix about 70-80% of the problems with it. The remaining items are largely things that my eyes just glide right over. I'm pretty happy with how things are when I release it, but there are still improvements to make.

I've actively requested feedback from my engaged players, making it clear that I want to hear about typos and other problems. I get the bulk of corrections in the first couple of weeks after release; a few more might trickle in later, but it seems like I succeed in eliminating technical errors (typos and such) early on. It might be tempting to rely on those early players as quasi-proofreaders and cut out the proofreading process, but I think that would be counterproductive. When people read something that is mostly correct and engaging, they'll notice the imperfections and assume that I'll want to know about them. But if there are a lot of problems, then they're less likely to take the effort of documenting dozens of typos, and may be skeptical that I'd bother correcting them if they did.

Speaking personally, at least, that's been true of me as a player. On the (very rare!) occasions where I spot a typo in Fallen London, I always email the developers about it: I've read roughly one million perfect words in that game, and know that they care deeply about that stuff. But I never bothered to send typo reports for Dead Man's Switch or Dragonfall: there were enough that I couldn't easily keep them in my head while playing, and I didn't want to interrupt my gameplay to write a long missive. I think there's a critical mass of nitpicky errors beyond which people just throw up their hands and don't bother reporting them. Getting below that critical mass prior to release helps ensure I get much more helpful feedback afterwards.


So, yeah. That's what I'm working on now! (And writing all this up makes me feel a little self-conscious, since I never proofread or edit this blog. I probably should, but, meh.) I've finished my initial revisions of the "Main Dialogue" and "Banter" documents, and am close to done with the "Crew Dialogue" one. Still more to go, but it feels like I'm making progress.

Update: Oh, I should probably include an actual example. Here's the current version of the Sacramento dialogue I shared in a previous post.

Plumber: Mmmmf. Hey, can you pass the mustard?
> How’s the food here?
P: Not bad! Then again, everything tastes great when you haven’t eaten in twenty-four hours. It’s impossible to find decent hot sauce around here lately, though.
> Are you on break?
P: You know it! I’ve been running around Sactown nonstop ever since the Free State fell and the Protectorate rose. Infrastructure has been a MESS since the invasion. Bullets and bombs punch holes in things, pipes are things, and so a fuck-ton of water started spraying buildings throughout the city.
Kora: This area doesn’t look too bad.
P: Saito’s engineers have repaired the crucial infrastructure in the green zone, but they don’t give two shits about people living in the rest of the city.
Isas: That unfortunate is. A sovereign his citizens should care for.
P: It’s a shame. The old government in Sactown also neglected the poor, and a lot of folks hoped that Saito would transform this place from a stinking shithole into a place worth living.
Valiri: Yeah, good luck with that.
(Once you’ve asked about what she does:)
> What sort of work do you do?
P: Plumbing, mostly. I’m doing my best to help repair the damaged buildings, but honestly, so far I’ve been busy just fixing the wealthy mansions. I feel kind of bad, but they’re the only ones with the credits to pay.
Arelia: At least you’re helping, in whatever way.
P: Outside of the core and the wealthy enclaves, though, the sprawl is royally screwed. Who knows if and when Saito will actually repair the destruction he caused, and not just reward his megacorp backers in San Francisco.
(After flooding:)
> {{CC}}Quickness: $(story.Global-Skillcheck_Hard){{/CC}} {{GM}}Swipe her plumbing supplies.{{/GM}}
PC0: Hey, look over there!
P: Where?!
{{GM}}While her back is turned, you steal her kit and toss it to your waiting companions. With any luck, she won’t notice the theft until you’ve finished here.{{/GM}}
> Never mind. I thought I spotted Maria Mercurial, but it was just a mirage.
P: Pretty nice mirage.
> Don’t you think that cloud looks like Dunkelzahn?
P: Maaaaybe. No such luck, though. The Big D is long gone.
> I thought I spotted a sucker getting robbed.
P: There are are a lot of those in the world today. Maybe Saito will do something about them.
(All return to root)
(After flooding:)
> Can I borrow your plumbing supplies?
P: You can’t be serious. I need these to work, and I need to work to live!
> {{CC}}Etiquette: Gang{{/CC}} Between you and me, I’m going to stick one to the Man. Strike a blow for the little people.
P: Heh heh heh. Yeah. That sounds fun! Hopefully I’ll get to hear about your exploits later.
> {{CC}}¥30{{/CC}} I’ll make it worth your while.
P: Hm, I COULD buy a lot of soy dogs with that… All right, fine. Just return the tools here whenever you’ve finished whatever kinky activities you have planned.
> Goodbye.

It isn't great, but better than the rough draft! Some of the lines have been tightened up, I added a bit more content (allowing the PC multiple lies during the theft), filled out the details on skillchecks and bribes.

I'm coming closer to the time when I'll be migrating that 500-ish pages of Google Docs into tens of thousands of nodes in the Shadowrun Editor, and cursing my existence and questioning my life choices... but that'll be another day, and another blog post.

Finally, a couple of tangentially-related videos that I want to mention but that don't merit their own post:

The Shadowrun community has been mostly enthusiastic about the trailer for the upcoming Netflix movie "Bright", and I can see why. While it doesn't seem to have actual cyberpunk in it, it totally feels like a Shadowrun entity, nailing a lot of the atmosphere and dynamics of the setting.

I'm withholding judgment for now. The "fairy lives" joke at the start gets a big Thumbs Down from me, but the rest of the trailer looks awesome.

I doubt that CalFree in Chains will be done this year, but if it was, December 22nd would be a good deadline to try and hit. If the movie is a success, I imagine it will spark a lot of new interest in Shadowrun-esque properties, and I think The Caldecott Caper and CalFree in Chains might be particularly relevant to the thematic issues they seem to be exploring.

Speaking of thematic issues, one of my favorite bands, Garbage, recently released an awesome video for their new song "No Horses". I'm absolutely mesmerized by it. I've probably watched this a dozen times, and have come to believe that it's sort of a CliffsNotes version of CalFree in Chains.


  1. This is amazing, my friend q-q I've spent upwards of several hours fawning over your work to friends, and though I know you've moved on from the modding scene, I just wanted to thank you for producing such an incredible experience with Antumbra/Caldecott/CalFree. I've yet to play CalFree but I've just come off of Caldecott, and seeing your process and the product is awe-inspiring and just plain-inspiring in general. There's way too much for me to say and the rest fucking difficult to put to words so [insert superlative praise here]. You're a hero. I hope you've been told how enjoyable and incredible your work is, from the overall down to the smallest detail.


    1. Aw, thank you so much for the kind words & thoughtful comments! I especially appreciate that you've told friends about them, thank you for spreading the word! Will write more on your other comment. :-)