So, it’s the start of development week 13, and I’m looking back on week 12, trying to figure out how it all went.
At first, I wasn’t exactly super-satisfied with it, but I’ve changed my mind.
A lot of stuff happened during the week, which took time away from actual development, but apparently I got a fair bit done in the little time that I did have.
The narrative gained a little over 7,700 words of weight during the week. Much less than I might have hoped for, but more than I’d expected before I’d done the arithmetic. It’s a little over 195,000 words now.
Twenty-five more narratives to write to hit my next milestone.
Under the hood, more got done. Let’s talk boring code.
The situational-formality model basically broke down outright during my biggest scene. It’s a difficult scene to write, given that every one of the dozen protagonist characters are in the same room at the same time, with major plot-points going on.
The situational-formality model (which was my second revision of that system) basically keeps track of how formal a given character believes a situation is at any given time. This is important for certain things like word-choice, particularly when referring to or addressing other characters.
The first model was quite crude, and while the second version was more sophisticated, it broke down when groups of characters had disparate ideas about situational formality. The new version (revision 3) is much more adaptable and robust, but that doesn’t mean there won’t be a version 4, in the future.
During the course of writing this week, it turned out that I needed another support system in the software, to do with relationships.
I’ve already got ways of quantitatively defining relationships between characters, which in itself is quite useful. The problem was that the system didn’t have a lot of scope for qualitatively defining them, and that was going to be needed later when the narrator wants to cast characters for more story-arcs with more complex requirements. While the new system is not much more complex to use than unilateral-flagging, it allows me to handle straightforward things like friendships, senpai/kohai relationships, and such, as well as more intricate things like frenemies, love-hate relationships, and all manner of complicated interpersonal relationships without having to force character-relationships into any sort of normative models.
The other half of that (complex story-arc preconditions, and expanded script-conditionals) will jump out at me later, but the support-system is there.
Lastly, a discussion with @AuroCyanide wound up giving me the idea of building a telemetry system for my engine. That resulted in new build of the Alpha demo (now at Alpha-2 – those of you who have Alpha-1 should go back to the same page to download the new version. The notes have been updated as well), which can send me error reports, as well as providing some direct user-feedback in the Alpha-demo, by adding a couple of (admittedly not-pretty) like/dislike buttons to the UI.
Users who are trying the Alpha-demo out can use those to register direct feedback on a passage. Is it working for them, is it broken, did I end up mangling the grammar, or is it missing an Oxford-comma? Either button lets me know whether the user feels positively or negatively about the passage (enough to press one of the feedback buttons) as well as a providing me with a copy of the text they are actually seeing at that moment.
While I was putting it in, I also used it to provide a little instrumentation on save/load times – because I’m slightly concerned that they might be a little high on lower-end hardware. We’ll see, once the numbers are in.
So, actually a much more productive week than I’d actually thought I was having.
Now, there’s some general life-weirdness happening this week, so I’m not sure how week-13 is going to treat me, but if I got that much done last week with so little useful development time, I’m not terribly worried that I’m going to fall too far behind.