fuzzy notepad

[blog] informative title

Hello here is what’s going on.



In all seriousness, it’s pretty nice to work for a place that maybe people have heard of. For four years i’ve had to mumble something about taxes and Florida and quietly realize i barely understood what i was doing, either. No longer!

Enjoying myself so far; it definitely has the feel of a company of engineers who build stuff just because they like building stuff. The difference is night and day.


I’m no longer capitalizing “i” as though it were a proper noun. Right, okay.


I’m kind of liking this whole full-disclosure thing with my dumb glitches. In the past week i have noticed:

  • I still try way, way, way too hard to find elegant solutions to problems. I said this before in some form, but i had engineering in mind. This week, i needed to get an Android device for Mel so she could use Square to peddle her wares at a con this weekend. I put considerable effort into figuring out how to re-juggle wireless service, or excuse buying a tablet because it’d be cheaper than a family plan, or whatever the goddamn fuck.

    In the end i just strolled into Best Buy, bought a new phone i wanted, and gave her my old phone. Which was my original plan.

    Embracing failure don’t cut it. Fail fast.

  • I seriously need to stop making ad hoc todo lists and dump stuff into bug trackers.

    I spent the first two weeks of the month out in a tiny hotel room in San Fransisco. And it was great. I worked on things i’ve been avoiding for ages, i started a project that’s only been a daydream for years.

    Then i returned home, and immediately noticed that everything seemed like a drag again. It took me a week, but i realized why.

    I have dozens of things i want to do. Some of them involve writing code. Many of them do not: replacing the router, hanging my whiteboard again, doing laundry, getting this crap off my desk, whatever endless list of errands.

    When i was out of town, none of these things were relevant; i couldn’t really do anything about them, so they weren’t on my mind at all. The moment i got home, i suddenly had a mental laundry list of Things To Do Right Now again. Because that’s where those things lurk: in RAM, in volatile memory, where i have to keep thinking about them for fear i’ll forget. I jot them down in todo files (todo, todo1, todo-new…), but then i have to keep glancing at the list to avoid forgetting about it, and i just see an impenetrable wall of stuff i don’t want to do.

    I don’t know what to do about that, specifically, but it helps tremendously to not let development contribute. I keep accumulating such lists of dozens of bugfixes or small features i intend to get to Real Soon Now, because they each individually seem like they should only take a few minutes, and it’s “a shame” to fill up the bug tracker with… bugs. (???) Naturally i don’t get around to them for whatever reason, so they become persistent baggage; things i have to juggle in my head all the time or keep jotted down on loose paper spread across my desk. It’s paralyzing, and it’s exhausting. Also, it’s dumb and i keep re-learning this lesson.

I’m toying with the idea of flipping my sleep schedule so i work at the end of the day; it might help some with the burnout and distractions, since nothing really happens at 3am. I don’t tend to deal well with weird sleep patterns, though. I may try it next week and see how it goes.


Speaking of, here are some things.

veekun has, er, not moved too much. A lot of stuff i want to do is held up while i fuck around with a JSON API. I’m happy to say that i’ve managed to un-overengineer what i wanted to do here, and i have a little test suite, and it’s gradually creeping closer to passing.

floof is, at last, ported to Pyramid. This doesn’t mean anything to anyone, which makes the work not worth it at all. Maybe now i’ll do something user-visible. After having pored over every last goddamn line of code in this thing, i really hope i’ve learned something about not overengineering in the beginning and focusing on getting a usable product first. (Doubtful.)

I agonized over AI for raidne for ages and finally implemented something, which involves passing Action objects around. You can now be attacked by a newt until you die. Death is indicated with a (deliberate) Python stack trace.

I started sanpera, an attempt at building a not-PIL imaging library for Python, atop Cython and GraphicsMagick. Fun times so far; i haven’t touched C in ages and it’s fun to do so in a somewhat padded environment. The “library” is just a bad port of the example GraphicsMagick program right now and i haven’t much touched it since i got home, but i’d like to actually make something of this.

Also, i threw some stuff up on github. I don’t think i like GitHub Issues enough to jump ship right now, but it’s a nice git interface, so i might as well mirror to it manually or something.

I’d like to make enough progress on something to deserve a dedicated blog post, but durrrr. Getting better.

Christ i have a flight at 7:30 in the morning goodnight.

If you like when I write words, you can fund future wordsmithing (and other endeavors) by throwing a couple bucks at my Patreon!

(illus. by Rumwik)