Every post

A while back I set a rule for myself– no more short game jams, a weekend just isn’t enough for me to make something satisfying while still getting enough sleep and being a functioning member of the household.

… and then I spent much of today waffling over the Kenney Jam (Theme: “Scale”). I had an idea for a silly mountain climbing (or scaling) game, but doing it the way I wanted would have meant figuring out some Godot features I haven’t used before, and I think the tutorial I tried to follow was out of date with Godot 4.7. I briefly considered taking the Kenney Sokoban assets and making the obvious game, but struggled to find a way to integrate the theme. I still might do that as a learning project, though. There was also a thing with a UFO and a shrink ray.

I’m not doing this one. There are 15 hours left and I’d like to get some sleep tonight.

I’ve been thinking about the Cult of Done manifesto lately. I think maybe a better thing to do would be to try to get one of my many started, unfinished things to some version of “done” this month.

I’m gonna sleep on it.

I do kind of like this mouse-following UFO thing though…

At my employers suggestion, I studied for and passed the Claude Certified Architect - Foundations. It’s whetted my appetite to get more hands-on experience with agents, and I need a personal project to learn/practice on. I’ve experimented a bit with a QA agent for dctech.events, which fixes janky titles, duplication, and adds categories where they are missing. It’s been humming along for a few weeks. I’m going to look for further opportunities to use dctech.events as an agent testbed.

Currently listening: The Friendly Orange Glow The Untold Story of the PLATO System and the Dawn of Cyberculture by Brian Dear 📚

I’m not very far into this, but one detail that makes me smile: the early PLATO keyboards had an “aha” key. “Aha” was how you exited the help system, as in “aha, I found the information I needed”.

From a 1965 paper on PLATO:

If he had difficulty with a question, he could push the button labelled “help”. The “help” button took the student into a help sequence which pertained to the question. The logic in a help sequence was similar to the logic in the main sequence. The student was presented with additional explanatory material and “help material”. Each question in a help sequence had to be answered correctly before proceeding further through the help sequence.

After completing a help sequence the student automatically returned to the question he was trying to answer in the main sequence.

However, if the student wished to return to the main sequence from any point in the help sequence, he could push the button labelled “aha”. An additional request for help on the same question would return the student to his previous position in the help sequence.

March was not a productive month for me, gamedev-wise. I spent more time playing Civ VII, reading, watching Monarch: Legacy of Monsters and re-watching season 1 of Jessica Jones (each!), than in Godot or other creative tools.

I don’t mourn the time spent or not spent, but the loss of momentum burns. In April, I want to turn that around. I find that if I launch the game, I’ll quickly find the thing most annoying or unsatisfying at the moment, and naturally get to work on it.

I just need to rebuild the habit of doing that more days than not.

Someone reached out a few days ago, trying to sell me “dctechevents.com”– which once upon a time was the domain for my DC Tech Events site.

Have I grown to like dctech dot events?

Yes.

Am I nostalgic for the old domain?

Also yes. But not $500 nostalgic.

Thankfully, my friend S here didn’t actually own the domain. When they wrote that email, the domain was in PendingDelete status, which is final. Whoever had owned it, stopped paying the bill. Once a domain has reached PendingDelete, nothing can stop it from eventually hitting the open market. At that moment, S might as well have been trying to sell me the Brooklyn bridge.

If I had expressed interest, they probably would have tried to grab it once it hit the market, through a back-ordering service, sell it to me, and pocket the difference. That would be the most honest way this scheme could work. Would they have taken my money before having the domain in hand? What if the backorder failed? If they have my money, why even try?

I don’t have to find out, though. I didn’t respond to their emails, and I was able to back order the domain successfully. It’s all (once again) mine!

I have a theory about creation with AI.

(wait! please come back)

My gut says this can’t be original, but I haven’t seen it exactly expressed anywhere.

1. The best art (and I’ll extend this to products in general) reflects the taste and discernment of just one person, or at least very few people.
2. LLM’s represent the antithesis of that, the voice of the masses, and perhaps worse: the voice of the masses digested by a deflavorizing machine.
3. The skill in creating with LLM’s is then to remain in charge, to make sure YOUR taste and discernment are dominant. If this effort gets in the way of creating, then the LLM is harmful to your endeavor.
4. Code has the opposite dynamic! You generally want the most generic, boring, un-creative implementation of a thing, unless you’re trying to break new ground in computer science. The voice of the masses is an asset.

I think one way to navigate #3 (at least for digital projects) is tool building, using AI not to necessarily make the thing, but make to make peripheral tools that improve your life or make creating more productive or pleasant. For SpinDoc, I don’t plan on using AI for graphics, game design, level design, etc– but the level editor is a place where “taste and discernment” doesn’t matter so much (at least not yet), and letting the LLM do most of the work on that has been helpful to me.

So, instead of putting the SpinDoc level editor aside, I dug in and now the thing generally works. I even got online level sharing to work: it spits out a URL that anyone can use to play your level. That’s something I’ve been day-dreaming about for over a year.

Next, is making the game more presentable and fun, and I’ve got some ideas for new puzzle mechanics and a whole new game mode I’d like to try.

I am experimenting with a 3D view (only dots and wands have been implemented so far). No decisions on that yet.

According to itch.io analytics, on most days, at least one person visits Brakes Escape. I suspect it’s just one person, who has made it one of their favorite time wasters. I think that’s cool! None of my other games exhibit that sort of steady usage pattern. It’s a small impact, but… it’s impact.

Even without that data point, BE is one of my games that I think should be developed further. It definitely wants to be a mobile game– I just have to get rid of the “frame” (borrowed from it’s most direct inspiration, NS Tower) and make it a full (tall) screen experience. From there, the graphics are pretty raw, and there’s room for power-ups and enemies of some sort I think.

What else?

I think The Zummoning has potential, but the card system is currently pretty half-assed. I could see simplifying it (have cards directly indicate the creatures being created) or leaning in to a more complete and coherent magic system.

SpinDoc has sort of been my white whale– one of the earliest things I started working on after getting into Godot. I think there might even be some niche-commercial potential, since the original game isn’t on the market anymore. (there was an official iOS port, but that hasn’t been maintained and is no longer for sale)

Exactly a year ago, I declared that 2025 would be the year of 3D. Not so much– it became a year of worrying about my job and finding the next one, and games took a back seat for most of it.

My goal: 2026 is going to be a year where I finish something, outside of jams. Maybe I’ll take a jam entry and try to make it into a complete, polished thing, or one of my other experiments or projects (remember the roguelike? Spindoc?), or it’ll be something from my notebook, or maybe a brand new idea.

It probably won’t be my favorite notebook game idea: “Breakout Teaches Typing”.

Along the way I would like to become productive with tools like Crocotile 3D, PixelOver, and Blender.

recently on DC Tech Events:

• I’ve added a new page for virtual events
• improved the layout on mobile
• I took a little troubleshooting journey into figuring out why NovaLUG events weren’t working– and now DC Tech Events can handle iCal recurrence rules AND calendars can have a url_override, which in this case is because the NovaLUG iCal feed was sending the URL for the location (and not a correct one), in which case the main “meetings” page is the most useful place to send people.

Without saying the “v” word, with LLM coding assistance I recently experimented with a couple of more ambitious visions for this site and codebase: for example, expanding into a broader network of similar sites for different cities, and/or making DC Tech Events more of an active community hub and meetup.com alternative. Ultimately, I decided against both. I enjoy the simplicity of the current site, and I don’t want a web project that eats more of my time and attention. My hobby is making games, not websites.

I do intend to keep running it in it’s current state. It still helps me find events, either directly, or when I occasionally go looking for events and groups to add.

I was playing Glypha the other day, and wondered if an “endless Joust” game would be fun– instead of discrete levels, you keep proceeding upward. It turns out, Joust Revival is exactly that. I can’t get it to work with Wine, though, and if a game isn’t playable on Linux, does it really exist?

Either way, Revival is otherwise a re-creation of the classic arcade Joust game. I have something different in mind, but definitely with the DNA of Joust. Or Glypha. I’m pretty sure I played Glypha first. I think I’m going to give it a try in the upcoming GitHub Game Off.

2025 hasn’t been a productive year for making games. The job hunt and related Government uncertainty basically ate up the first seven months, and I haven’t really recovered the lost momentum. I am going to try to end strong with three simple jam games before the end of the year, the one described above, and two jams I’ve enjoyed participating with in the past: the 1-Button Jam and the 20 Second Game Jam. Embrace constraints!

Pinball Putt was my entry in the 2024 1 Button Jam, and twenty thousand milliseconds my entry in the 2023 20 Second Game Jam.

At work, I’ve been diving into AI/LLM stuff, and I’m still pretty much always using coding assistants when I work on DC Tech Events. I hadn’t really attempted to bring that into game development, though. Last week, I decided to try it on my current roguelike project.

This game began life as part of a 10-day game jam, and I was beginning to regret some of the architectural compromises I made under the time pressure. In particular, the lines of responsibility were pretty murky between the class that generated maps, and the one that displayed the map and tracked positions. I gave Github Copilot some direction on what functionality belonged where, and it did a credible job cleaning things up, which made it easier for me to move on. Neat.

This led to some excitement and further experimentation, which resulted in two MCP servers, which allow a coding assistant more visibility and control of a a running Godot project. This one runs outside of Godot, and connects to the debugging port and language server, while the other is a Godot Plugin, and provides an HTTP server that allows for an agent to manipulate the node tree, and even send input. It’s cool when it’s all working.

As a “big swing” experiment, I created an empty project with a single spritesheet, and asked copilot to make a match 3 game. The game itself turned out OK, but across many attempts it completely failed to identify which parts of the sheet should be used as game pieces. Score one for humanity!

I’ve started a lot of projects and finish basically none of them– I think the promising thing for me is that I can use coding agents to get past the annoying/tedious tasks that cause me to lose interest and move on, and finish more games. For example, at one point I started “CISO Clicker”, meant to be an idle/clicker game that hopefully imparts some wisdom about the asymmetric nature of cybersecurity. I never really got past generating some company names.

With a few prompts, I’ve got a start on exactly the sort of simulation I’ve been thinking about:

The deadline is fast approaching, and I haven’t made a ton of progress– but now there are mobs that actually chase you. There’s no combat or death yet. That’s next.

The graphics and colors here were meant to be placeholders, but it’s honestly growing on me. With the black shadows and bright light, I find it kind of pleasing in a brutalist way. Perhaps I’m rationalizing laziness, but I think I’ll lean into abstraction here. Colors and shapes, without any traditional genre references.

I’ve got my 11th bullet journal set up for September.

Even when I fall out of using it as a sort of daily productivity/mindfulness practice, I’ve benefited from simply having one notebook active at a time, with dated entries. Pre-BuJo I would jot things down wherever, including digital tools like OneNote, etc, and often those notes were never to be seen again.

I know, a few weeks ago I wrote “the internet doesn’t need more words about AI”.

Since then, I started working at a company that considers AI as part of it’s toolkit for helping customers, contributed some code to an internal project that uses the Google Gemini API, and my first client project?

Helping write LLM policy.

¯\_(ツ)_/¯

I’ve tried and failed a couple of times to write some sort of position statement in this space, that would capture a mix of hope and fear, while giving myself permission to keep learning and experimenting. I could probably have ChatGPT write it, but I’m not so far gone as to subject readers to machine-written prose.

I am giving myself that permission to learn and experiment, though. Program or Be Programmed seems newly relevant, and I think the closing words of LLMs Are Weird Computers captures the moment pretty well:

But I will say that if you don’t want to be caught with your pants down when your workplace does expect you to do more and different things with this technology, there’s no better time than now to start learning how to use it.

I’ve finished listening to All of the Marvels by Douglas Wolk.

From the premise, “man reads 27,000 comic books”, I expected an A.J. Jacobs-style story about how attempting a ridiculous project impacts the author’s life and relationships, with some tidbits about the comics thrown in. That would have been OK with me!

This is not that sort of book, though. Wolk spends each chapter pulling on a thread of the Marvel story, either for a character or team (Spiderman, Thor, The X-Men), or a theme (film, music, Presidents), highlighting specific issues that are notable. It’s one man’s highlight reel of the Marvel universe, and I enjoyed the journey.

If I had been reading this book on paper or device, I would have dog-eared a lot of pages, or taken notes. I listened to the audiobook, mostly while driving or walking, so I couldn’t note as much. I was still able to remember or jot down a few things. Here are the comics ‘All the Marvels’ has added to my reading list:

• Hickman’s Fantastic Four
• Priests Black Panther
• Simonson’s Thor
• Ryan North’s The Unbeatable Squirrel Girl
• Warren Ellis Thunderbolts
• Some amount of the Dark Reign saga

Today is my first day as an Ad Hoc employee. I’ve watched from the sidelines over the years as they’ve hired a number of my favorite people, both former CFPB colleagues and “internet friends” whose work and careers I’ve taken an interest in. I’m looking forward to being part of the team!

The actual client I’ll be working for is… a different civilian federal agency with a financial focus. You might be thinking “out of the frying pan and into the fire”, and aren’t completely wrong about that. My discussions with folks from Ad Hoc, the “prime” contractor subcontracting to Ad Hoc, and the client agency have convinced me that the work is necessary and should be funded for some years to come. I think it will be stable at least through the length of the contract, so long as I do good work.

Hello from Camp T. Brady Saunders. It’s the last day of summer camp. The kiddo mostly had a good time, but we’re both ready to go home.

Today is technically also the last day that I am a CFPB employee(last day actually working was 7/3). I wasn’t laid off this time, I’ve got a new job that starts Monday. I’m pretty excited about it, but I’ll save writing about that for next week.

I’ve been mucking around with my operating system setup for reasons too tedious to describe here. One hilarious bit of hijinx involved trying to use the Nobara repository exactly in the way it says not to…

It is NOT recommended to try to upgrade your Fedora installation to Nobara and support for doing so will not be provided.

Turns out, there are reasons.

But, the point of this post is actually a nice thing I discovered:

• When you install fedora, by default your /home/ directory is placed on a BTRFS subvolume
• If you ever need to install again, the Fedora installer recognizes the subvolume and can preserve it.

In the past, if you wanted to re-install a distro and preserve all of your files, you either needed the foresight to put /home' on it’s own partition or device, or have backups. Now, it’s no big deal.

I’m not that proud of The Invisible Alligator in its current form, but the Godot Wild Jam 82 deadline is here, and it’s nice to ship a game after almost 6 months of not doing much gamedev.

It could use more levels, sound and music, animation, more NPC variety, and more effort spent on graphics. The core gameplay isn’t that fun yet, but Benjamin calls it “satisfying”, so that’s something.

I started out thinking about it as a 2D version of Hitman Go, with animals. It ended up in a place that (if built out more) would feel a bit like Hoplite.

Maybe I’ll keep tinkering, towards a more polished post-jam version.

I know, the internet doesn’t need more words about AI, but not addressing my own usage here feels like an omission.

A good deal of the DC Tech Events code was written with Amazon Q. A few things led to this:

• Being on the job market, I felt like I needed get a handle on this stuff, to at least have opinions formed by experience and not just what I read on the internet.
• I managed to get my hands on $50 of AWS credit that could only be spent on Q.

So, I decided that DC Tech Events would be an experiment in working with an LLM coding assistant. I naturally tend to be a bit of an architecture astronaut. You could say Q exacerbated that, or at least didn’t temper that tendency at all. From another angle, it took me to the logical conclusion of my sketchiest ideas faster than I would have otherwise. To abuse the “astronaut” metaphor: Q got me to the moon (and the realization that life on the moon isn’t that pleasant) much sooner than I would have without it.

I had a CDK project deploying a defensible cloud architecture for the site, using S3, Cloudfront, Lambda, API Gateway, Cognito, and DynamoDB. The first “maybe this sucks” moment came when I started working on tweaking the HTML and CSS, and I didn’t have a good way to preview changes without doing the whole cdk deploy routine, which could take a couple of minutes.

That led to a container-centric refactor, that was able to run locally using docker compose. This is when I decided to share an early screenshot. It worked, but the complexity was starting to make me feel nauseous.

This prompt was my hail mary:

Reimagine this whole project as a static site generator. There is a directory called _groups, with a yaml file describing each group. There is a directory called _single_events for events that don’t come from groups(also yaml). All “suggestions” and the review process will all happen via Github pull requests, so there is no need to provide UI or API’s enabling that. There is no longer a need for API’s or login or databases. Restructure the project to accomplish this as simply as possible.

The aggregator should work in two phases: one fetches ical files, and updates a local copy of the file only if it has updated (and supports conditional HTTP get via etag or last modified date). The other converts downloaded iCals and single event YAML into new YAML files:

• upcoming.yaml : the remainder of the current month, and all events for the following month
• per-month files (like july.yaml)

The flask app should be reconfigured to pull from these YAML files instead of dynamoDB.

Remove the current GithHub actions. Instead, when a change is made to main, the aggregator should run, freeze.py should run, and the built site should be deployed via github page

I don’t recall whether it worked on the first try, and it certainly wasn’t the end of the road (I eventually abandoned the per-month organization, for example), but it did the thing. I was impressed enough to save that prompt because it felt like a noteworthy moment.

I’d liken the whole experience to: banging software into shape by criticizing it. I like criticizing stuff! (I came into blogging during the new media douchebag era, after all). In the future, I think I prefer working this way, over not.

If I personally continue using this (and similar tech), am I contributing to making the world worse? The energy and environmental cost might be overstated, but it isn’t nothing. Is it akin to the other compromises I might make in a day, like driving my gasoline-powered car, grilling over charcoal, or zoning out in the shower? Much worse? Much less? I don’t know yet.

That isn’t the only lens where things look bleak, either: it’s the same tools and infrastructure that make the whiz-bang coding assistants work that lets search engines spit out fact-shaped, information-like blurbs that are only correct by coincidence. It’s shitty that with the right prompts, you can replicate an artists work, or apply their style to new subject matter, especially if that artist is still alive and working. I wonder if content generated by models trained on other model-generated work will be the grey goo fate of the web.

The title of this post was meant to be an X-Files reference, but I wonder if cigarettes are in fact an apt metaphor: bad for you and the people around you, enjoyable (for some), and hard to quit.

The main DC Tech Events site is generated by a custom static generator, in three phases:

1. Loop through the groups directory and download a local copy of every calendar. Groups must have an iCal or RSS feed. For RSS feeds, each item is further scanned for an embedded JSON-LD Event description. I’ll extend that to support other metadata formats, like microformats (and RSS feeds that directly include event data) as I encounter them.
2. Combine that data with any future single events into a single structured file (YAML).
3. Generate the website, which is built from a Flask application with Frozen-Flask. This includes two semi-secret URL’s that are used for the newsletter: a simplified version of the calendar, in HTML and plaintext.

Once a day, and anytime a change is merged into the main branch, those three steps are run by Github Actions, and the resulting site is pushed to Github Pages.

The RSS + JSON-LD thing might seem arbitrary, but it exists for a mission critical reason: emojis! For meetup.com groups, the iCal feed garbles emojis. It took me some time to realize that my code wasn’t the problem. While digging into it, I discovered that meetup.com event pages include JSON-LD, which preserves emojis. I should probably update the iCal parser to also augment feeds with JSON-LD, but the current system works fine right now.

Anyone who wants to submit a new group or event could submit a Pull Request on the github repo, but most will probably use add.dctech.events. When an event or group is submitted that way, a pull request is created for them (example). The “add” site is an AWS Chalice app.

The newsletter system is also built with Chalice. When someone confirms their email address, it gets saved to a contact list in Amazon’s Simple Email Service(SES). A scheduled task runs every Monday morning that fetches the “secret” URL’s and sends every subscriber their copy of the newsletter. SES also handles generating a unique “unsubscribe” link for each subscriber.

(but, why would anyone unsubscribe?)

Some progress on the “beat game” remake (for the moment, called “Loopnik”). It’s much easier now that I have a better idea of what I’m doing and how to structure things. Aside from level advancement, this does everything the original does, with half the lines of code.

The “graphics” are kind of charmless and generic right now, but they’re placeholders. The donut is from Kenney.

My first game jam entry was beat game, submitted to the So Bad It’s Good Game Jam in July 2023. Until then, I’d worked on tutorials, toy projects, and had done some tinkering on a recreation of Spin Doctor.

The game ranked 4th overall and won the “uniqueness” category. I’m proud of that, but clearly the “so bad” theme worked in my favor. I periodically wonder what a good version of that game would look like, and I’m looking for a way to get back into game dev after a couple of months out of it.

So, that’s my summer project, I think. I’m eyeing a few upcoming game jams, but if I get some momentum on this, I’ll probably just keep plugging away on it.

Because you, as a person who reads blogs, are dear to me– you get to be among the first to know that the new DC Tech Events site is live.

I’m still tweaking, but I’m pleased with how it works right now– it’s a static site deployed on Github Pages. There’s a form for submitting events, but behind the scenes that just opens a pull request.

Next up: getting the newsletter piece up and running.

I think I’m just not going to write about what’s happening at my job again until there’s some definitive resolution. It’s too fluid, and that isn’t what I want the blog to really be about. I definitely appreciate everyone’s kind thoughts, though.

This week is about:

• preparing for the AWS Security Specialty exam, which I take on 4/30. I passed this three years ago, and it’s time to renew.
• getting ready for the kiddo’s first campout as a BSA Scout, (as opposed to a cub scout) this weekend.

Next month, I really want to get back to some game stuff. So far the year of 3D has been the year of nothing. Maybe I’ll do the Godot Wild Jam again.

I have made some progress on the mysterious non-game personal project. I hope to unveil that next month too.

update: things are on hold again.

Along with 1,400 of my closest friends, I received a “Reduction in Force” notice yesterday. At 6pm today, I’ll no longer have access to work systems, and in 60 days I’ll no longer be a CFPB employee.

I’m sure I’ll find more to say later, but I’m heartbroken that I don’t get to keep working with my amazing team, supporting an important mission for all Americans.

• I’m still employed (and even doing some work, given that FISMA-related work counts as “statutory duties”). The next milestone there seems to be Monday, when the judge will rule on what happens next.
• I haven’t been doing much game-related stuff, with the exception that the kiddo has challenged me to make another, more difficult level for Robo Rampage, featuring a new enemy. I’ve made a little progress on this.
• I’ve been working on a non-game personal project, which will be familiar to anyone who knew me 10-15 years ago…

Today, with my family, I watched the first episode of Human Footprint, which is about invasive species. It’s full of US government employees (I noticed USGS, BLM, and Army Corps of Engineers) doing important science and engineering: keeping asian carp out of the great lakes, studying pythons in the everglades, and managing the wild horse population in the west.

I had to wonder, after these last few weeks… do these people still have jobs? What will happen without them?

The kiddo and I worked through this Godot Reinforcement Learning tutorial, and (after some troubleshooting to get onnx to install) it worked! We trained a model to be able to play a circular pong game.

I don’t necessarily understand most of what I did there, for example why I would pick “Stable Baselines” vs " Clean RL", but it definitely seems like something worth experimenting with some more. It might be fun to use to use RL to improve the the opponent in The Zummoning.

It’s a shame that in order to use a model in-game, you need to use the .NET version of Godot, which (apparently) can’t yet export to the web.

I’ve finished “Robo Rampage”, and thus I’m done with the Complete Godot 3D course.

Towards the end, I kind of wanted to just get it over with, so I didn’t put much effort into decorating the level (in fact, most of the decoration you do see was added by Benjamin). Still, I like that it’s got enemies that chase you, some interesting lighting effects, and two different guns that recoil and spawn little sparky-effects where the bullets impact.

What’s next? I want to get out of the tutorial headspace and work on my own projects for a while. I did start this Blender tutorial a few weeks ago, and plan to pick it up again. In the meantime, revel in this sweet-ass trash can I made all by myself!

This week, work is still in a holding pattern. I applied to a good number of jobs, talked to people, and un-mothballed one of my earliest Godot projects (a digital version the Balanced Attack Model game). My current thinking with that game: spend some time making the best version of that I can, then reaching out to the creator of the BAM deck to see if there’s some way we can collaborate and/or if he’d be OK if I make it available online.

Today, I’m working on the Amazon Bedrock Workshop, the first “homework” from the Become a Solution Architect program.

Starting Monday, and for the next four weeks I’m attending CISA’s “skilling academy” for System Security Analysis full-time.

So far it’s been four days since I and most of my colleagues were ordered not to work. I started the week getting over a cold, so didn’t do much of note Monday or Tuesday. Yesterday, I registered to take the AWS Security exam in April (my certification expires in May) and took a practice exam.

Will this last days? weeks? months? I have no idea.

Benjamin’s recent game jam entry is Rolling is Spinning. It gets pretty tricky at the end!

I did the coding, following his concept and directions. He also did the art/animation for the ball that represents the player.

The theme of the jam is “spinning”, hence the name, to remind the judges of that rolling should count as spinning.

• I’ve been helping my son on an entry for the Boss Rush Jam 2025– the concepts are his, most of the art is royalty free stuff from a humble bundle, and I do most of the coding and Godot-wrangling.
• I’m on the last project of Compete Godot 3D, which is a first person shooter. I’ve got a world to run around in, an enemy that follows me if I get too close, and a gun that can shoot then enemy and recoils when fired, and has a muzzle flash. There’s more to do, but it’s a satisfying project.

Every three or four years I get the urge to build a PC. Most of the time, the urge passes and I move on to other ideas/projects/what have you.

This time, I built the dang thing!

Component-wise, the most interesting part is probably the GPU– I was able to get my hands on an Intel Arc B580 (Battlemage!) and it’s spent the last two weeks sitting in my office, unloved. Last night, I trekked down to Microcenter to buy the rest of the computer and, after some drama (it took me a while to figure out where all the screws were, how to attach the CPU cooler, and that the motherboard wanted the 8 pin OR 8+4 power connection, but not BOTH), I have a nicely specced little PC next to my desk.

By little, I mean about the size of a mini-fridge. A “mid tower” is still pretty huge.

Battlemage support in Linux still has a ways to go. I decided to start with Fedora Rawhide, so I can get any improvements soon after they land in the kernel.

So far, I’ve been able to play some games that were unusable or pretty rough on my laptop (a circa 2016 Lenovo Carbon X1), and Godot is humming along nicely. The Abandoned Spaceship Demo was really jerky on the laptop, but runs on the new box like it’s no big deal.

Somewhere along the line, I told myself a story: Before making 3D games, I needed to learn 3D, as in get a handle on 3D graphics as a general topic, and perhaps develop some expertise in something like Blender.

From talking to people at gamedev meetups, it seems like maybe that story is nonsense?

To test that theory, I’ve been working on a Godot 3D tutorial and finding that things generally makes sense. Coming from 2D, I was surprised that:

• Up is positive, and down is negative (on the 2D canvas, it’s the other way around– like you’re always in the bottom right quadrant of the standard coordinate system, the origin is in the top left corner)
• I’m measuring things in meters instead of pixels.

Aside from that, the additional axis, and a new set of nodes to use, most of what I’ve learned about Godot still applies. So, in 2025, I’m hoping to make some 3D games.

And, maybe learn Blender.

I was on the fence a bit, but I allowed my yearly micro.blog subscription renew today.

There are a thousand ways to put a site or blog up online, but I like the apps, and how micro bridges to BlueSky and the fediverse. I also enjoy the bookshelf feature, which seems like a quirky one-off (I mean, why not movies, TV, and games?) – which somehow makes it more lovely.

Some progress on “SpinDoc”:

• “rings” (aka, dots you can only use once)
• enemy wands that can move to appropriately colored dots

By the end of this month, I’d like to have the project in a state that at least resembles a game: a couple of levels, a few missing Spin Doctor features (in particular, doors), some sound and animation. In November, I’ll focus on the GitHub Game Off. When I get back to SpinDoc after that, it’ll be more level design, some original mechanics I’ve been thinking off, and polish.

Finished reading: How to Read Superhero Comics and why by Geoff Klock 📚

This was more academic than I was expecting (full of words like ‘misprision’ and ‘kenosis’), but Klock makes a convincing case that the books he writes about exist in conversation with each other, the comics of past eras, and criticism like Seduction of the Innocent.

I’m not sure if that interplay is supposed to answer the ‘why’ question by itself, but it does have me somewhat intrigued– at least enough to check out some books I’ve missed (like Marvels, Astro City, Kingdom Come, America’s Best Comics, and Grant Morrison’s JLA).

The ratings are in, and Quack Circuits… ranked near the bottom in most categories (there were 156 submissions).

Getting 21st in “originality” suits me just fine though– that (or whatever similar category is available, like “innovation”) is usually where I do best, and doing well there is the kind of thing that makes me think I have something to contribute to the game dev world.

The comments were generally positive, which isn’t unusual– game jam participants are a friendly bunch.

For 8 of the 9 days of the jam, I was working on a fairly straightforward puzzle game. In the final 24 hours, I pivoted to making the puzzles a mini-game in a larger story of a duck trying to escape… something (in my mind, it was an animal testing lab, but that never made it to the final product). Would it have ranked higher if I spent that last day refining the puzzle game? Maybe? I can’t know.

One thing I did for the first time, was get deployment nailed down first. Before I wrote a line of game code, I had the base project all set up with Github Actions and successfully deploying the base project to itch.io. I’ll definitely keep doing that for future projects.

This morning I was thinking back on some videos I’ve watched (like this) about abandoned NFT games, and it occurred to me that such a virtual world might be a compelling setting for a new game.

A world of half-built and/or I’ll-conceived villas and islands, bugs (exploitable?) that will never be fixed, bad algorithmic art, a massively deflated in-game currency. Everything you see was put there by VC, whales, crypto true believers, speculators, scammers and dupes.

Any random Molly White blog post could be a plot line.

When I was writing this post, I almost committed to sitting out game jams for a while, in order to focus on a small number of longer-term projects. Honestly, I could use some momentum right now, and I feel the most momentum after finishing a game jam.

So, I’m signed up for Godot Wild Jam 72.

This time around, I’m going to try to make a really simple game, and focus on the visuals and providing a quality, polished experience. In other words, I won’t try to design a whole-ass magic system.

(though thanks to a recent humble bundle, I’m looking forward to diving in to Game Magic: A Designer’s Guide to Magic Systems in Theory and Practice.)

Two years ago, while my son was attending an event at the library, I spent some time flipping through a book about the Unity game engine (maybe this one?). It wasn’t intimidating, and looked fun. I came away thinking: “huh, I could make games”.

Then I got home and couldn’t make Unity run on my laptop. There’s a happy ending, though: I found Godot, which ran just fine.

I followed a few tutorials to make a flappy bird clone, a breakout clone, a simple platformer, and made progress on a much larger platformer, but after a few months I kind of drifted to other pursuits.

One year ago, I decided to dive back in an embrace game development as a serious hobby. I finished the platformer tutorial, and started entering game jams. Between July 2023 and June 2024, I made six games. There were no unqualified successes, but I usually did OK in one or more categories, like “innovation”, “mechanics”, or “fun”. What I’ve proved to myself is that I’m pretty good at coming up with ideas for interesting game mechanics, and have enough Godot-fu to turn those ideas into actual software that someone can play, and even have fun playing.

Honestly, this makes me feel a little like a wizard.

BUT a good game is more than just a collection of playable mechanics, and a successful game takes way more than just interesting ideas. It takes quality, style, a well-designed progression/difficulty curve, marketing (and a market!), and good timing.

I think my next step is to take a couple of the ideas I’ve worked on (either as jam entries, or other projects) and try to make good, complete, games out of them. I’m not sure about trying to market anything, but I’d at least like to have one or more games that are, in theory at least, marketable.

I think three of the six game jam entries represent concepts that could be developed further (Beat Game, Zummoning, and Brakes Escape), but I should probably narrow it down to one or two.

I’d also like to keep working on a Spin Doctor-inspired puzzle game, and a cybersecurity-themed game I’m still fleshing out (working title: “CISO Clicker”). I’ll try to use #screenshotsaturday and local meetups as my drumbeat: in a successful week or month, I’ll have some progress to share.

And I guess, I’ll plan to post a new retrospective/update in July 2025!

• I still haven’t entirely kicked the Marvel Snap habit, but I feel like I’ve made peace with it. It’s a thing I enjoy, so I shouldn’t beat myself up about it. I think I’m paraphrasing Ted Lasso here.
• When on the phone or tablet, and not Snap-ing, I’ve been playing Slice and Dice
• I recently spent about an hour or two digging in to Book of Hours. I don’t love it yet, but after reading the two Alexis Kennedy books, I want to dive back in and try to pay more attention to the actual writing. Likewise, I want to revisit Cultist Simulator (which I at least got to the point with where I could seemingly live indefinitely without accomplishing much) and Sunless Sea (which I only launched once, but haven’t really played).
• I spent a little time today finally playing Dome Keeper. I’m only one run in, but I’ll be coming back. So far: good game, great atmosphere.

Currently listening: Whole Earth by John Markoff 📚

A year or two back, I listened to Tom Wolfe’s “The Electric Kool-Aid Acid Test”, and soon after, for unrelated reasons, started reading some stuff about the early years of personal computing (including Markoff’s “What the Dormouse Said”).

Stewart Brand was the common thread running through all of it.

I spent some time last night and this morning learning about Orchestrator, a visual scripting plugin for Godot. I was mostly taking a look to see if my son might want to try it out, as an alternative to “regular” coding. He knows Scratch pretty well, but hasn’t taken to written coding yet. I thought Orchestrator might offer him a way to make stuff happen in Godot, using a style of programming he’s used to.

At first, I was amused and appalled that in an example, this short line of gdscript:

rotation += angular_speed * delta

… equated to a six-node graph. After completing the step by step tutorial, I appreciated it more.

I’m missing the words for describing this coherently, but, to me, it encourages a different way of thinking about the functions you write. Instead of the step-wise, imperative march from input (or not) to return value (or not), you kind of need to start at the end, with the impact you want to cause, and then work out the information and logic needed to inform that action.

I haven’t spent much time studying (let alone doing) functional programming, but… is that what functional programming is like?

I’m not sure I’ll reach for Orchestrator in my own projects, and I’m not sure the kiddo will either. I did show him the results I was able to get it, and if he wants to learn more I’ll certainly work with him. I was wrong to assume that it’s anything like Scratch, and I don’t think it really improves on gdscript in terms of reducing the cognitive work involved in programming. Orchestrator doesn’t simplify Godot.

I used the Pong Wars deployment as an excuse to try out GitHub actions. It’s nice! Almost too nice, like the fact that it’s free makes me suspicious.

I ended up struggling with what I thought would be the easy part– GitHub pages. Back when I was a young man, all you needed to do was push your site files to the gh-pages branch. That still works, after you do some extra configuration. It was only the next day that I figured out what the problem is.

Godot-CI worked pretty well, though I needed to build my own version of the container image, because I’m not using a supported Godot version. I’m using a pre-release build of Godot 4.3, because I wanted to take advantage of the improvements to web exports.

I’ve published my Godot re-creation of Pong Wars by Koen van Gilst on GitHub (and embedded it below!) It’s not exactly a game– more of a “visual toy”. I did add some fun extras:

• left click a block to instantly flip it’s color
• right click anywhere to flip ALL of the blocks.

Honestly, I had convinced myself that I was going to rank higher in Godot Wild Jam #68, at least in the originality category. Being in the top third for originality is OK, though.

Much of the written feedback (even on the much rougher Ludum Dare version) supports the idea that this is worth developing further. I’m thinking of digging in and trying to make the best version of “Zummoning” that I can, and see where I am at the end of summer.

I’ve made many improvements in the new version of “The Zummoning” submitted to Godot Wild Jam. Unfortunately, the name isn’t one of them. Oh well.

• Every card combination now actually does something
• There is help text that explains what the cards do, while you’re playing
• It’s clearer what is happening during the combat phase of the turn.
• More art, some music, and sound effects!

It’s still far from a professional production. But, I think it’s fun?

One thought I had while relaxing after submission: the game seems to come down to the decision every turn to play a strong creature on a single location, or weak(er) creatures on two locations. Maybe the card system is sort of extraneous?

I’m cleaning up “The Zummoning” for submission to Godot Wild Jam this weekend (The GWJ folks usually disallow games submitted to other jams, but make exceptions sometimes when schedules completely overlap, which was the case with Ludum Dare this month).

Tonight, I hope to improve the card-combining systems and remove some of the annoying gaps.

If I get more time before the deadline (Sunday night), I hope to add some UI improvements (primarily, make it less mysterious what the cards do, alone and in combination). I’d also like to make it clearer what’s actually happening during the combat phase.

As the screenshot shows, I’ve at least added a background image and improved the layout, since the LD submissions. Not obvious from the screenshot, the color of the spooky forest background gradually shifts between a handful of hues. It’s stupid simple, but I find the effect really pleasing. The (gdscript) code is just:

func next_color():
	var colors = [
		Color.DARK_RED,
	 	Color.DARK_GREEN, 
		Color.DARK_BLUE, 
		Color.DARK_ORANGE, 
		Color.DARK_OLIVE_GREEN, 
		Color.DARK_MAGENTA]
	var tween = get_tree().create_tween()
	tween.tween_property($SpookyForest,"modulate", colors.pick_random(),15)

The next_color function is connected to a timer that fires every 15 seconds.

This feels a bit like cheating: but immediately after my last post on the topic, I built a paper prototype of the Spoon Theory game using index cards. I’m going to count that as hitting my goal to have a prototype by the end of the month.

For April and May, I want to return to Brakes Escape– adding polish, more variability in scenery and obstacles, and perhaps some power ups. By the end of June, I want to have it in the Google Play Store.

I still like my spoon-theory game concept (as amended), though there’s a lingering skepticism over whether the thing I’ve described will be fun and actually a challenge to solve– but that’s what testing and iteration are for.

I’ve slept on it for a week. I now have 7 more days until my arbitrary, self-imposed deadline to build the prototype. Let’s go. Start. Now.

(Is blogging this itself procrastination? Maybe.)

Me, a couple of hours ago:

Is it possible to take what I’ve described and actually make it fun and challenging? Or, will the right path (or a right path) always be obvious at a glance?

Now that I’ve thought about it some more, I think part of the answer here is: the player shouldn’t be the only entity moving across the board. There should be NPC’s representing social interactions that the player may want to encounter (that add spoons) and others that the player will want to avoid (that take away spoons). They should move in predictable patterns, and the player’s challenge becomes:

• observe and understand the NPC’s movement patterns
• figure that into their plan for traversing the board, along with destinations to reach, routes to take, and turns available.

That sounds more like a puzzle.

Found this list of ideas I had posted to Twitter over some number of years. It’s not all gold, but I still like a few of them:

• a first-in, first out laundry hamper
• Shazam for the noises a malfunctioning appliance or car might make.
• URL shortner that implies value judgement. Think like.ly, hate.ly, whatev.er
• a site for stolen recipes, called TastyLeaks

I vaguely recall trying to build this:

• site that lets you take any book, and turn it into a chapter-a-week lesson plan. discussion and vote-based rev. q’s and homework

This is basically haveibeenpwned.com:

• site where you can sign up, submit your email address(s), get notified when it shows up in a dump from @LulzSec & friends

Here’s my initial concept for the spoon theory game:

Each level/puzzle is a grid, with a starting location and one or more “must do” spaces, zero or more bonus spaces, and other spaces that either add or remove “spoons” (which can be thought of as stamina or energy). The player’s challenge is to design a route that connects the starting location to (at least) each “must do” location in order to complete the level, without running out of spoons or turns.

Levels can vary in size, number of must do or bonus spaces, and the number of turns available to complete the puzzle. I currently plan to offer unlimited undo.

Example tiles:

• must do: appointments, social obligations, school or work
• bonus: things that might represent trade-offs, like a party can be both fun and overwhelming. May cost or earn spoons.
• other spaces: these might represent normal day-to-day tasks and chores that cost spoons, or preferred activities that add spoons, like rest, play, listening to music.

Perhaps you earn a star rating on each level (ala Angry Birds), 1 star for passing, two for passing and getting some (not all) bonuses, and three for passing and getting all bonuses. You could then revisit levels to increase your stars.

prior art

Spoon theory lends itself pretty naturally to games– stamina/energy/“spoons” are a limited resource you need to manage, and I’m not the first to attempt to translate it into a game mechanic. You might even see the reverse– video games as a metaphor for how some people need to think about and manage their energy.

Most of the games I’ve seen are choose-your-own-adventure-style interactive stories, where you play through the day of a character, and make decisions that cost or earn spoons. If you search itch.io for “spoon theory”, you’ll get a number of those, with a few variations. If you search on Steam, you’ll find Robin, which fits that description as well.

Some other things I found:

• Here’s one modeled as a chat conversation
• This one, where you must rapidly tap the space bar to generate spoons (it’s really hard!)
• A board game and a dice-and-paper simulation.

I guess my idea isn’t not a choose-your-own-adventure, it’s just that all of the the choices and their consequences are revealed at once, and the challenge is in planning your path through the day, managing your spoons and time.

open questions

Is it possible to take what I’ve described and actually make it fun and challenging? Or, will the right path (or a right path) always be obvious at a glance?

I’m not sure I can know without building it. And if that is the case, then having a working prototype will allow me to troubleshoot the gameplay. I hope to have that prototype by the end of the month.

One of my previous complaints about the game was that the water was kind of… blah. I added this shader and now I like it!...

Finished reading: The Bullet Journal Method by Ryder Carroll 📚

I skimmed or skipped some of the more philosophical material (like a chapter on “radiance”), but I have some good ideas about how to reinvigorate my bujo practice:

• get back to the habit of reviewing the month-so-far every morning, or at least weekdays.
• be more liberal about creating “collections” outside of the daily log flow, for particular projects, topics, or goals
• just write more! let the journal be more of a log of ideas, thoughts, frustrations, and victories, and not just a todo list

One thing I’ve learned about myself after participating in a couple of game jams: I like having deadlines.

So, I’m going to try setting a deadline for a personal project that I’ve been meaning to make progress on: a game about spoon theory. I want to have something playable (not necessarily done) by the end of March.

My aim here is:

• a fun game
• that leaves the player with increased empathy for/ understanding of folks who need to consciously manage their energy/stamina

… with a bonus goal of the game being somewhat relaxing and not too taxing. The game should be a good stim. It should restore spoons, rather than take them away.

I’ve done a round of post-jam tweaking on Brakes Escape. The game no longer simply stops generating new platforms (and walls!) once you hit a certain level, sound effects have been added, and some bugs have been fixed. To make things a little more difficult, I’ve introduced “solid” platforms that you can’t jump through, and made the rising water gradually increase in speed.

There are still some things I’m unsatisfied with:

• the death animation is pretty janky
• the water is portrayed as just a translucent blue sheet that rises. I’d like it to look more like water.
• Some sound effects are still missing, and there’s no music
• I think it needs more platform types (like conveyor belts) and some additional hazards (enemies?) to really be interesting and fun.

So, what’s next? I think I want to adapt the game for mobile, while working on the above items.

I do plan to update the itch.io web version at least once more, to generate a single-threaded version once Godot 4.3 is released (soon?).

Some lessons learned from my recent game jam experience:

• Don’t wait until the very end of the jam to download the export templates bundle (see screenshot).
• The last few hours of the jam are NOT the right time to learn shaders
• The last few hours of the jam are NOT the right time to learn parallax backgrounds

Godot features I attempted to re-create (poorly)

• Path2D (i was trying to move platforms back and forth by tracking points and managing velocity, before I figured out that Path2D could do this for me)
• One-way collisions (I had used an Area2D and code to manage what collision layers a platform is, to create the effect where you could jump through a platform from below, while it acts as if it’s solid when you land on it. Turned out this was built in, and a matter of checking one goddamned checkbox.

Godot features and nodes that I had never used before (besides Path2D/PathFollow2D and one-way collisions)

• ParallaxBackground and ParallaxLayer (though I messed this up somehow, my background that is supposed to repeat stops repeating at some point)
• The ability to place limits on camera movement, which I’d previously done myself via code.

The only add-on I used was Godot State Charts, which continues to be the bees knees.

Previously, I posted two ideas for Godot Wild Jam 66. I ended up spending a few days on the first one and realized I didn’t like where it was going and I didn’t feel good about getting it into shape for the end of the jam. So, I switched gears to the second one (a variation on an idea I’d already been tinkering with).

It’s going better, but I’m not sure it’ll be shippable by the deadline (4pm tomorrow). Still, it’s nice to have some momentum. I’m really happy with how the player character moves, but the game definitely needs more.. game. We’ll see how for I get it tomorrow.

Current thinking for the jam, either:

• starting over from scratch on the NS-Tower-inspired game I’ve tinkered with on and off (see a GIF of it here), and adding something to escape from, like a rising water level.
• a horizontal - scrolling space shooter where you can only move in short bursts, which I’m calling “we only have hyperdrive”

I’m signed up for Godot Wild Jam #66, with a theme of “Escape”– which is pretty straightforward, as far as game jam themes go. I do think it’s quite cool that GWJ has accessibility as a rating category, and I’m going to try to see how much of the accessibility guidelines I can meet.

The jam goes for nine days. My current thinking on timing is: have something basically working by early next week, then use the rest of the time to polish the graphics, UI, and audio.

My son and I are still making tweaks to the roguelike that resulted from following @selinadev@indiepocalypse.social’s tutorial: a new spell, visual effects, weapons, and new enemies. I enjoy the collaboration, so we’ll probably keep it going for a while.

The most notable new feature is a minimap! It still needs work, but I’m happy that I was able to get it working. This is my first time writing Godot code that generates a graphic from scratch. Every pixel in the minimap represents one tile in the current dungeon. Right now, it draws the entire graphic (walks the entire array of tiles) on every turn, which seems excessive and dumb. On the other hand, it’s performing OK, so “fixing” that might be a premature optimization.

In theory, the next Godot Wild Jam starts next Friday, I was thinking I might give that a try.

I did the thing! I finished all 13 parts of Selinadev’s Godot 4 Roguelike tutorial.

There were a good number of things that were new to me: procedural map generation, AStar navigation, custom resources, and using those custom resources to compose entities. I’m happy with the knowledge gained by working through this.

The end product is… not my cup of tea, really.

On roguelikes generally: When I was a kid, one of the computers available to me had nethack installed. I poked around a bit, but it never really clicked for me (and the same computer had SimCity! ). One turn-based roguelike I do play these days is Hoplite. I think what I like about Hoplite is that it plays almost like a board or puzzle game, the challenge is in reasoning about how to get through the level safely.

On this particular game: after adding armor, weapons, and the power-ups available when you gain a level, combat is super easy, and the challenge curve stops when you get 8 levels deep. I appreciate the 1-bit art style, but it kills me a bit that a game that includes fireballs and lighting attack’s doesn’t make those things visually interesting events– baddies take damage, maybe die, and a message is logged. This isn’t really the author’s fault, since she adapted a tutorial that builds a classic terminal/console game. It’s just a shame this leaves behind a lot of the cool stuff Godot can do.

My son/project manager designed some new enemies and other graphics to be incorporate in the game, and I want to honor that effort by integrating those assets and ideas. I’ll probably spend another week doing that and maybe looking for other ways to increase the usability and fun factor.

I received a (legit) notice of eligibility for this Verizon settlement in my junk email folder.

I know there are best practices and a class of professionals focused on email deliverability. I wonder if the opposite exists? A service for companies who are legally obligated to send out notices of some sort, but would prefer they not actually be received.

(to be clear, I’m not claiming that’s what’s happening here)

I’ve finished part 11 of roguelike tutorial. Now, you can delve deeper into the dungeon, and the player gains experience points and levels. I’m excited about the next part:

Our character can now go deeper into the dungeon. However, right now all the floors are identical in their challenge. The RNG will mean that they will encounter different mixtures of monsters and items, but it still lacks progression. So next time we will look at a way to increase both the difficulty and the reward the deeper the player goes

I’m up to Part 10 of @selinadev@indiepocalypse.social’s Godot 4 Roguelike tutorial. What’s new?

• Potions to heal with
• Scrolls of lightning, fireball, and confusion
• saving and loading
• That main menu screen where you can either start a new game or continue where you left off

My son suggested adding some instructions on the main menu screen. He’s basically my project manager at this point.

Via CRASH THE ARCADE, I learned of Matt Mullenweg’s birthday wish:

I want you to blog.

Publish a post. About anything! It can be long or short, a photo or a video, maybe a quote or a link to something you found interesting. Don’t sweat it. Just blog. Share something you created, or amplify something you enjoyed. It doesn’t take much. The act of publishing will be a gift for you and me.

I thought it might be interesting to use this as an excuse to collect a sort of history or timeline of my blogging, and memorialize the various iterations of karchner.com and related web things I’ve built and torn down over a quarter century.

High School (1996ish-1998, Milton Hershey School)

I definitely had “home pages” on services like GeoCities and Angelfire, but I don’t remember much. I think there were Ayn Rand quotes involved.

This might be hard to believe, but in 11th grade I was an officer in our chapter of Future Farmers of America, and had built a homepage for the chapter.

I’m fuzzy on the order of events here, but somehow one of the MHS college/career counselors hooked me up with an internship interview with a guy at Hershey Foods, who was impressed that my FFA site used frames and hired me! The school loaned me a Powerbook, and I was paid a hourly rate to convert recipes from Word documents into HTML for Hershey and other brands it owned (at that time, this included a lot of pasta brands, like Ronzoni).

A person I worked with at Hershey connected me with a summer internship at a web design firm in Mt Gretna, PA, called Kindle Design Group. Kindle Design Group owned “kindle.com” before Amazon. I wonder if they made some money from that?

I think I continued doing Hershey stuff during the year after that. Between high-school and college I spent a summer working at a harness-racing magazine called “Times: In Harness”. I can’t find much on the web about T:IH right now, but here’s an excellent TV commercial.

College (1998-2000, RIT )

I had a homepage on an RIT server. The things I remember are:

• I had been learning Macromedia Dreamweaver, which among other cool DHTML tricks made it easy to build drag-and-drop interfaces. I made a site logo that you could drag around, for no particular benefit. I called it “the gratuitously draggable logo”.
• It did have a space for periodic, single-line, updates from the life-of-Ross. I probably started out editing it by hand, but at some point I started using Blogger, which means I had entered my RIT network account password onto a third party, internet-start-up website. In 2024 that sounds absolutely bonkers, but it didn’t seem that way in the 90’s.

This was peak “dotcom” era, and when I wasn’t learning from RIT’s excellent teachers, I spent a good amount of my time working on terrible internet ideas, as was the fashion of the time. Some of the terrible ideas were:

• metamovie.com. A website about websites about movies
• xmlonthemac.com. I don’t know, I guess it seemed like mac users had special XML needs or something. I’m sorry.

In this era I was reading Scripting News a lot, and I spent time tinkering with some of the stuff that Dave Winer and Userland were putting out, like editthispage.com and weblogs.com. I definitely maintained something of a personal blog on one of these services, and a Nelly Furtado fan-site.

I think this is also when I got my hands on the ‘karchner.com’ domain. Someone reached out who had bought it accidentally (while building a site for Kartchner Caverns) and I was able to buy it from them.

What I wasn’t doing so much, was studying and going to class. I left RIT after two years without a degree.

Back in PA (2000-2005)

I moved in with my mother, who had an apartment in Harrisburg at the time. Work-wise, I got involved with a web hosting startup that didn’t really go anywhere, worked at Blockbuster Video, spent a few years at Earthlink before going to Knowledge Planet(which later became part of Mzinga).

I think it was in the Earthlink era that I started a blog on karchner.com, using Movable Type, but I also recall experimenting with Radio Userland. The oldest version of karchner.com I can find on the wayback machine was using Radio Userland, so perhaps that was first?

This was the time period when journalists thought blogging was interesting/weird/scary, and I actually managed to get written about:

• More employers reading workers’ blogs in Central Penn Business Journal
• There was a Sara Bozich article in “The Patriot News” that profiled local bloggers. I’ve lost and found that article several times over the years, and presently can’t find it.
• Part of a Harrisburg Blog Meetup (I was an organizer) was recorded for the NPR show Weekend America.

I was really excited about RSS in those days, and the possibilities around not just using an RSS reader to read a bunch of sites in one place (though that was cool!), but as a way to programmatically mix content together in interesting ways. The first of these was “harrisburgblogs.com”, which was a simple feed-of-feeds: all of the Central PA bloggers I knew about, in one place. People seemed to like it.

OK, what if there was a “Harrisburg Blogs” for everyone? I moved on to creating something new, called “Localfeeds”, which did the same trick for cities around the world. Instead of using a hand-selected list of feeds, I combined two ideas:

• The ability to tag web pages with geographic coordinates, provided by geourl.orgs “ICBM” metatag
• RSS feed discovery tags

With those two details, I could process a big list of sites, and sites that had both of those things were added to the database. The result looked like this.

Localfeeds got some attention in the wider world, and was written about in an issue of Esther Dyson’s Release 1.0, blogs, and podcasts, and got me invited to be on a panel at the first Bloggercon.

DC (2006 and on)

After moving to the DC area, I got involved with the DC Blog Meetup for a while (and may have been organizer for a bit). At some time before or after the move, I also moved my blog from karchner.com to rossnotes.com, and switched from whatever I was using previously (either Movable Type or PyBlosxom) to Wordpress.

The most notable thing in this era is that I started tracking different technology events that sounded interesting. There’s a lot going on, and nobody was listing it all in one place. So, “DC Tech Events” became an ongoing series on my blog, and then it became a newsletter, and a dedicated website.

There was a mini-scandal when another site started re-publishing it without credit.

I credit dctechevents with opening a lot of doors for me, including my previous and even my current job. At some point during this journey, I decided the software I was writing to manage dctechevents (I called it “eventgrinder”) was unique and something more people should be able to use, and started down the path of turning it into a business. At least, I created an LLC, opened a business checking account, and worked on eventgrinder like a serious side-hustle for a while. After a while this led to opening more dctechevents like sites for other cities(like Localfeeds), but that’s where it kind of lost steam. I made myself a publisher, dependent on this software I was writing, but actually hated using.

By 2013 (or before)I had given up on eventgrinder, dctechevents, and the network of similar tech events sites.

After this, I cycled between not blogging for long periods, and starting over with some new platform or tool. I recall Wordpress (self-hosted), Wordpress.com, and GitHub pages being used at certain times. I think Twitter played a role here: it was easier to post and that’s where the people were.

I think around 2018 I became aware of micro.blog and the indieweb movement. I definitely kicked the tires on micro.blog, but also I was infatuated with “Serverless”, and the idea that there was an optimum expression of the indieweb’s standards and protocols that could be built out of AWS serverless tools. This also led to it’s own cycle of neglect and occasional rebuilding-from-scratch, and ultimately went nowhere.

I hit a moment where I decided to just commit to not doing it. I had discovered game development, and building new web things wasn’t interesting anymore, at least as a hobby. It’s hard to find a problem more solved than blogging (except maybe text editing), and I wasn’t even sure I wanted a blog.

Until, a couple of months ago, I decided I did want to blog again, and re-approached micro.blog without the not-invented-here attitude and decided to try again.

In the next year, I’d like to:

• Watch less YouTube, watch more Udemy, more O’Reilly Learning, more books, more blog posts and independent websites.
• Whip the not-yet-titled “jumping game” into something you can download (buy?) on (at least) the Google Play Store.
• Create a new game, that teaches some cybersecurity concepts, and get it out in the world (even as a beta or preview) in time for Cybersecurity Awareness Month (October). This is not courseware. My reference point is SimCity. SimCity doesn’t teach you how to work in local government, but you walk away with insights into the sort of decisions and trade-offs that go into running a city. I want to make something that puts fun first, but where success means thinking like a CISO (or someone similarly in charge of cybersecurity for an organization).

I just finished part 7 of the roguelike tutorial.

New stuff:

• Baddies!
• … that chase you
• … and can also die
• (but you can die too)
• A user interface that let’s you make sense of what’s happening.

I like how the author sets up the UI, I’ve learned some neat tricks there. I was a little fuzzy on the vim-like modal input system before this latest part, it clicked when adding the ability to scroll back through the message log.

Most of the graphics are (per the tutorial) from the Kenney 1-bit pack, but my son had an idea for a new graphic for the wall tile. I like it!

I usually don’t consider myself very “handy”, but I’m proud of the job I did today switching which side the refrigerator door opens on. One of the screws holding the bottom bracket on broke of, leaving most of itself in the hole. I tried a few methods of getting the partial screw out, but that was a doomed effort.

So, now our refrigerator opens on the right instead of the left. Annoying, but better than any of the alternatives.

I’ve submitted twenty thousand milliseconds to the 20 Second Game Jam.

I’m not even necessarily sure it qualifies as a game. Maybe it’s more like a mutant skill challenge?

I wanted to do more with sound and graphics and juice and didn’t quite budget my time well enough to get there. Still, it’s nice to ship a thing.

About a month ago, I migrated my mastodon account over to this blog (via micro.blog’s built-in support for ActivityPub). How’s that going?

Mostly good. I like that I’ve got a full-fledged personal website here, that is fully customizable and could theoretically run anywhere. Behind the scenes, it’s just hugo.

I appreciate how micro.blog is trying to differentiate itself from traditional social media, creating a space that doesn’t reward popularity or sensationalism. There are no trending topics, follower counts, likes, or retweets to chase– just people writing and the things they write.

Of course, much of this is “good like eating your vegetables” and not “good like eating a cheeseburger”. All those little dopamine hits that make things like Twitter, Facebook, and Mastodon addictive and un-calm are also what make them fun.

I think discovery is worse here. The emoji thing is cute but I don’t think it works well in practice. For example, m.b wants 🖋️ to signify “pens and ink”, but sometimes it doesn’t. It’d also be nice to not have a hard-coded set of topics. Really, I just want hashtags, and I want to be able to follow hashtags. Like Mastodon. I can see how “trending topics” is a feature to avoid, but it’s perfectly possible to have hashtags without elevating trends.

One thing I have found useful is the book covers page. If you see someone posting about a book you like or that sounds interesting to you, that’s probably a good person to follow.

I’ve likened Godot to Hypercard before– they aren’t directly comparable, but my experience using Godot brought back a sense of possibility and fun that I felt when using Hypercard back in the 90’s.

I should probably be pretty excited to learn about Decker, which has modern touches, but also seems to lean pretty hard on nostalgia and Hypercard/Macpaint aesthetics. I’ll carve out some more time to kick it around some more, but honestly it leaves me a bit cold.

One thing I looked for, and couldn’t find, was some equivalent of “background” or template cards, which seem pretty core to some Hypercard uses, like address books or proto-Filemaker databases. OTOH, maybe you can accomplish that sort of thing with contraptions, which seem like a nice way to build widgets-of-widgets.

I spent at least 30 minutes trying to figure out why the score in my game wasn’t updating past the first point. It would go from 0 to 1, but then stayed at 1. I had a setter on the score variable, to update the score label shown to the user whenever the score changed.

It turned out, my setter wasn’t actually setting, so to score was always zero, and the incoming new score was always 1.

The missing line in the “set” code below is SCORE=new_score

Finished reading: Deadly Class Vol 12: A Fond Farewell, Part Two by Rick Remender 📚

Some of the late-middle volumes were a bit of a drag for me, the characters and their exploits didn’t seem that interesting after leaving school. In this final arc (volume 11 and 12), I enjoyed seeing how these people turned out as adults, and the ending was pretty satisfying.

I’ve been spending some time trying to clean up search results for my name (yeah, kind of douchy, sorry), and one thing that’s been vexing me is my old Twitter name. I deleted my ‘rossk’ account a while back, but I guess 15 years of google juice means my name will forever be linked with twitter.com/rossk, even though someone else owns it now.

All of that is to say, I’ve created a new twitter account, which I hope becomes more prominent in search results for “Ross Karchner”, and which links back to this site.

One week turned into three weeks, but I feel like I’ve worked that particular digital addiction out of my system (and diving in to Snap has actually helped crystalize an idea I’ve been thinking about, so it wasn’t entirely a waste). OK, back to making stuff.

My current idea for the 20 second game jam is really simple and pretty wonky: a trainer for being able to differentiate times shorter than a second (say, 500ms vs 700ms). Back in 2015 I went to some Site Reliability Engineering (SRE) training where this was mentioned as a useful skill for SRE’s. I didn’t entirely give up, but at that moment my dream of ever being an SRE seemed a little more remote.

There are some optional themes for the jam, but I think time itself is a sort of implied theme, and pondering that brought the SRE training back to mind. Thankfully, this isn’t a judged competition, so nobody else has to necessarily get it. Since the idea itself is so simple, I do want to lean in to trying to make it as fun and engaging as I can.

I’ve also started futzing with a micro.blog client written in Godot. We’ll see if that goes anywhere.

The results are in, and… it’s nothing to brag about. I do like that innovation is my strongest category again.

I don’t know exactly how many entrees I’m ranked against. The stats say that there were 1597 entries in the “jam” competition, and 66% of all entries (“jam”, and a separate competition, “compo”) met the ratings threshold for getting ranked, so maybe I’m ranked against 1067 or so “jam” entries? That puts me in the top half for most categories, and not far from it on the Overall. In a competition that includes professionals, hobbyists, and teams, I feel OK about that.

Here is some weekend progress on Don’t Get Crushed (post-jam version, to be released after the review window closes):

• Fixed the janky vibrations
• improved some art
• The level number now included a decimal component, making it clearer how much progress you’ve made within a level

You can also see the previously reported progress: visible scores, the ability to restart using the up arrow, and the crusher following the player more closely (and no longer leaving a safe area on the bottom of the screen)

Not much visible progress since I lasted posted this game, but I have implemented both the spring platforms and the “slam” mechanic (which can be combined for bigger jumps. I’ve also re-built all of the player movement around Godot State Charts.

So far, I've addressed a few items from the feedback list:

• There is now a visible score
• Players can restart the game after dying using the up arrow
• The crusher now follows the player more closely, eliminating the "safe zone" at the bottom of the screen.

Don't Get Crushed was my submission to Ludum Dare 54, my second-ever attempt to enter a game jam. The game didn't end up much like what I was imagining when I started working (blame the 72-hour time limit, and my own limitations for that). The graphics aren't great, and it's got a few glitches-- but I think it's fun.

Before entering Ludum Dare, I probably would have explained the appeal of game jams as a way to test your skills against other game developers. That's still an attractive aspect of it, but my thinking has changed quite a bit in the past week.

I'm new to making games, but I've made stuff before: writing, software, images, even a bit of music. Most of the time? Nobody cares.

Because of the way LD is set up, if you put in the time to review other people's games, your game will be prioritized for other people to review. You'll get actual feedback!

I made this goofy, glitchy thing, and people are playing it. 33 People have left numeric ratings (I won't know the results of this until the rating period closes, in 8 days)! 21 People left written feedback! One person even drew a picture.

It's kind of exhilarating. If I continue to enter game jams, it'll be because it's fun as hell to make stuff and get actual feedback.

What's next?

The feedback on the game has been quite nice, but here's some of the criticism it's gotten (all valid):

• Levels are too long
• No way to know your progress within a level
• The graphics are too "basic"
• It's kind of sloppy that while the game starts and plays all with the keyboard, you need to use the mouse to click "try again" after getting crushed
• No real sense of urgency...
• ... and without that sense of urgency, the game can be pretty easy if you play carefully.
• no score to see.

One thing I hate, that nobody has mentioned: the shaking effect that sometimes appears in the crushers or the obstacle blocks isn't something I programmed or intended, it's just some side-effect of Godot's physics, I guess. I'd like to at least have more control over that.

What I'd like to try to have a "post jam" version of the game, available the day ratings are released, that addresses most of those points.

After that, there are other game ideas I'd like to move on to, but might revisit this one sometime.

It's 2023 and I'm starting a blog. Again.

Why?

I'll write more later, but in the past I've enjoyed have a blog as a sort of public notebook, for posting things I'm learning about or just want to share. I miss that, and want to start doing that again.

One thing I want to make sure I do, is break the (self) expectation that blog posts need to be any particular length, so I'll end this right here.