Reflections- Wrong and Right

Snowset is big (relatively), but it's also simple. It is also, unfortunately, rushed. Allow me to explain in the form of a series of lists.
Some stats:
Snowset is:
-68  passages (not counting the guide, and no I'm not adding another.)
-Around 7,000 words (Discounting code stuff, but it's mostly just what you see anyway)
-A total of 6 different types of Twine functions (all of which I have used in other games)
In total.
Snowset took:
-About 14 hours straight of work
-A period of time ranging from 2:00 AM to 4:00 PM (For what it's worth, I'm an insomniac anyway)
-A series of increasingly questionable decisions
To create.

So what went wrong? What went right? We'll start with the former.
What went wrong:
-Scheduling. Very much so. The jam I submitted this to ended up running for more than a month, but I only got 99% of my work in during the last day. I blame real life for this.
-Scope. Things got somewhat out of hand when it came to both the length of scenes, and the ways in which they interconnect, and more importantly don't. Most of the scenes in the game are unique, meaning that I had to write multiple variants of essentially the same idea so that things flow together properly. I'll probably try to avoid doing this in the future, as it increases my workload tremendously, but can likely be avoided with  a better-planned story-chart.
-Premise. It may not be clear from the final product, but the original Snowset was meant to be significantly darker than it actually ended up being. While I'm actually fairly happy with how it turned out,  letting things slip from their original concept probably isn't a good habit to get into.
-Scope, on the other end of the scale. There's a number of characters I ended up cutting, though most of them I'm not too upset about because they either weren't a great concept to begin with, didn't mesh well with the final version of the game,  or both. There is one character I wish I'd had time to implement, though-  A once-hostile infiltrator by the code-name of Trick (which he would point out is basically taunting unaware enemies), who would only give you his real name (How about Nick? 'Cause it rhymes?) if you spend your final hour with him. I also had to cut an ending point that branched out into several single passages. In hindsight, though, this is probably something I should have actively cut anyway, since it broke up the pacing of the ending for no good reason... Not to mention that the game already has multiple endings to begin with!
So yeah, a lot of stuff could've been better. But I also think I got some stuff right.
What went right:
-The game overall, considering how rushed it was. I'm sure it still has typos and maybe some bugs, but it's largely complete and finished more or less as intended. While I did cut some content for expediency, I didn't end up publishing anything in an unfinished state. In short, despite everything, it is in fact done.
-Scope, again from another angle. Despite the crunch, this is still my biggest game yet, and I feel that it's suffered relatively little in terms of quality due to the rush. While getting everything to work together was a massive hassle, I did manage to mostly pull it off in the end, and in a way that generally doesn't require the player to backtrack to see everything of interest.
-The guide. Sort of. The actual implementation of the guide was very rushed (Think "10 minutes till deadline and I'm just getting started"), but I think the concept behind it is good for a game of this scale. Aside from design, a fairly explicit guide should help players get to what they want to see without floundering about, or (possibly worse), not being able able to tell if there's something they missed at all, especially since this game (at least currently) doesn't have a completion-tracker other than unlocking the guide after the first run. Speaking of which, I think that was a pretty good decision too- Having the guide available from the start would make "cheating" very tempting, but locking it behind a run-completion forces players to look at all "stages" of the game before they even have a chance to spoil it. Of course, reaching an ending is trivially easy, so it also doesn't prevent players who really need it from being able to access it to begin with.
-Technical simplicity. This game is a lot less programming-heavy than my previous ones. While I didn't completely shun Twine's "advanced" features where they filled a nice well, most of the game is just linked passages, as opposed to some of my previous games which made heavy use of paragraphs or links that only show up upon doing X thing or having Y variable active. (Fragments is a prime example of that, cramming in so many features that I ended up having to write three different versions of almost every passage,  and also ended up making most of the content nearly invisible to most players.) The most complicated variables used in Snowset boil down to "Who have you talked to?" This did help immensely with development time, simply because I was able to spend most of my time actually writing rather than working out bugs. (There were a few, but most of them went away without much issue.)
Well, that's the good stuff. I think I'm actually proudest of the guide feature, which is a little weird because it was kind of added at the last minute. Regardless, it's a fairly easy-to-implement concept, so everything else aside, I think I've gained a potentially valuable tool from this process.

And that's all my thoughts on this thing. Appropriately, it's also my longest devlog by far. Now if you'll excuse me, I still haven't slept in  more than 24 hours (I only started work on the game after already being awake most of the day),  so I'm going to go pass out.

G'night!

-Bon