Log

2533U

Yakshaving is a perfect term. Yakshaving is what I’ve been doing for days now, maybe a couple of weeks. I know I’m yakshaving when I suddenly find myself very annoyed that I didn’t get to finish a particularly meaningless thing that I can’t even remember why I MUST finish, and I realize I’ve been in this mode for a while, not thinking about anything else. Talking to people is a chore, I have to force myself to understand and think through other things. It’s not a good feeling. It feels obsessive, it feels like something other than me is in control and it gets angry when it doesn’t get its way temporarily.

This is how it happens: I might have an afternoon off, or a slow day at work, some form of unexpected free time. My brain goes “ooo, free time, we should do something, nothing’s occupying our focus right now!” and it goes into overdrive to find a thing to do. At this point anything that floats into my field of vision can become the subject of my next hyperfocus. This current time started when I read a bit of news about Microsoft bringing GitHub closer into the organization. I was in a bad mood (because my job feels insecure, but that’s another story), so when I read this, and thought about how GitHub might eventually go away like many other things the big four have acquired, my brain latched on with glee. Perfect, something nebulous to be afraid of but have no influence over, but with a specific, practical side!

The first phase was about setting up my own git hosting. It is perfectly logical: I like independent, open-source stuff, I think, philosophically, that the utopia of tech is an open-source world where everybody has their own little website and all our software tools are controlled by us, running smoothly on the family server sitting in everybody’s basement. So, let’s set up a git server and move all my dead projects and 2 living repos over there from GitHub. But wait, where will I run this server? It can’t be on a VPS, cause that means maintenance costs and that’s not something I want to start spending $10 each month on. “Luckily” I have a media server running in my mom’s house that runs a few VMs for backup and stuff.

Second phase: server setup. How much memory can I allocate, is my hypervisor overprovisioned, jeez these VMs don’t have regular backups of the system image, let’s add a disk so the new VM has storage, on and on and on. After a few days of this: burnout. I am knee deep in learning Ansible (a tool designed to automate server deployment and management) because I found a project that uses it to set up a few self hosted things, including Gitea, a git server, and frankly, I just can’t remember the original goal anymore.

Third phase: restart, build an app, for some reason. I need to stop this madness. I like technology because it can be fun and cute. I like cute little servers that I can play around with. I delete the VM I spent a week setting up, out of frustration, probably. Take a deep breath, find a better, more fun, healthy way to play around with a VM. (Do I remember when this has become about having a VM and looking for something to do with it? No, I don’t.) Oh look, a project called “Smallweb” that’s a standalone little webserver, serving folders on subdomains, with a cute design and emoji on the docs page. It says it’s designed for fun and tinkering! Well, it’s JavaScript based, but I know a bit of JavaScript. Fast forward three days: I’m learning server side JS, in an obscure web framework called Hono, I’m banging my head against JSX (templating, don’t ask), I’m writing notes, I’m furiously trying to get ChatGPT to ingest the docs to all these tools even its massive training data hasn’t heard of, and I’m feeling angry, frustrated and short tempered again.

Meanwhile I started on new ADHD meds (just made my hyperfocus impervious to distractions), set up rules and rituals to finish the day (I ignore them and stay up until 2am doing something I don’t remember why I started), my girlfriend’s getting more and more worried, I’m constantly slightly dizzy from looking at screens for 18 hours each day and I swear even the dog seems surprised when I fully pay attention to him.

Yakshaving is best described by Seth Godin’s blog post on the term. I remember hearing this first with trying to change a lightbulb, but his version works too. Basically, I start with wanting to wash the car. But the garden hose is broken, so I need to buy a new one. But the store parking is paid, so I should borrow my neighbors monthly pass. But I can’t until I return the pillow I borrowed from him last month. But I can’t, cause the stuffing fell out and it’s a yakfur pillow. And before I know it, I’m at the zoo, shaving a yak, just so I can wash the car.

You start with something small and eminently sensible, but that leads to a cascading series of problems and steps and requirements and in the end the whole thing comes crashing down when you’re seventeen steps removed from your original goal, in the zoo, shaving a yak. You can’t really see the way back from here to the thing that really mattered that led to this, but simultaneously you’re completely convinced that this is crucial and the whole project is at risk unless you shave this yak.

I hate this yak. This yak is the only thing in my life that matters. Fuck this yak.

Well, anyways.

2530U

There is a certain hacker-y, diy aesthetic in websites and software that I tend to like. Stuff like this usually feels like it’s plaintext, even when it’s not actually. The design is simple and feels like a basic text file but the UX is good and modern, just not the “modern” that tends to mean gimmicky and bandwagony.

I wonder how I could go about making a living working with this type of design and software, but for “normal” people, not just developer and sysadmin and geeky types. I love geeky types, but I’m not sufficiently geeky to feel really at home in that world. I can speak the language, but it’s not my native tounge.

I do think that the remedy for the highly polished, psychologically tuned exploitation machines that are modern phones and computers and apps is this type of diy, hacker ethos. The design, the aesthetic is a part of that, and I think an important part. Design can manifest the philosophy behind software, so something that looks plaintext, does work in a simple, robust way with plaintext files and folders (ala Unix/Posix) but has a well thought out, considered UX for people who aren’t willing to learn the weird logic of computers to use them is the way to go.
I hope at some point I’ll find a way to start working with this type of stuff, it feels more authentic to me rather than the modern polish and enterprise complexity.

This is the future of personal computing, if we make it. I want to make it.

Well, anyways.

2530S

I might need to learn or get better at some things. That’s true in general always, of course, but I am it that late nigth modd once again where I think about going indie, or starting a business. And if I want to eventually start a business, I need to both focus on my existing strengths and build up some new skills to complement them.
My job that I just started in recently seems to be collapsing soon, so I’m thinking about what business to start and how, as I usually do in turbulent times.

The question is what to focus on that gives me the most valuable skills when it comes to starting something new and hopefully profitable. I read through the website of 37signals, the makers of, among other things, Basecamp again. I’ve known about them for a long time, and have often found inspiration in their designs and products. Reading their website and ethos now makes me want to get back into the software/SaaS world. I am nostalgic for the time in the mid 2010s that I worked at an SaaS company and used to feel like this type of business was the futre. Those were the times of Nomadlist, Basecamp, Slack and the like. Software was eating the business world, the big players restricted themselves to building OSs and office suites, the field was open and VC money was everywhere. In those days I dreamt about starting a software business, an indie studio or something similar, building a niche, well designed product and building a digital nomad life funded by it.
On nights like tonight I feel that old hopefulness that anything is possible and well designed, well built, simple apps hold the key to the future, or at least my future.

My thinking has changed since then of course, as has the industry. Web apps are nothing new, we’ve learned that not everyone always has a fast, stable internet connection and the internet is about advertising and endless feeds now. A good idea is nothing without good, stable, long lasting implementation. Data is fragile, the field is pretty saturated and there is much less money to be had for free after the post-Covid correction of the software development industry.

There are still people thinking like we used to though. I just recently heard about a friend of a friend in Hungary that is building an AI-driven app for municipal governments to analyze and synchronize local statutes. When I heard about that, I felt that old excitement of a good idea and a solo developer building it from his bedroom.
I have learned about the importance of not just building, but also selling stuff since then.

All this is to say that I’m sitting here, late at night, and I’m thinking about starting a Ruby on Rails course again. From my past research it seems like the best option for an indie developer for building web app based businesses, an all-in-one tool to cover every need, with good performance, a solid, stable dev environment. Plus 37signals not only uses it but created it, so it feeds my daydreams of building something like they did, on a smaller scale.

Well, anyways.