Projects

1. Struggles

I’ve been putting off writing this blog. For one, I wasn’t sure what I want to say this month. The world outside feels like it has been more of the same crap. More insane stuff happening. More terrible job market. More layoffs in the news. I think part of the problem is that I’m on social media again, and even though I try to curate a lot of that crap out it still creeps in. Because everyone is talking about it all.

I’m also more willing to keep an eye on the job market situation, because I quit my job. Yes, I’m aware that doing so was a dumb decision. It wasn’t paying the bills after factoring in things like fuel to get to/from work, and I didn’t like the job. Devoting my time and attention to a task I dislike for 8 hours a day in exchange for not enough money to pay the bills is absurd. Doubly so when I have a reasonably long list of valuable skills and abilities that I enjoy using.

So, I quit. Now I need to find the next thing. In the middle of a historically bad job market. While I wait for U.S. software jobs to exist again. Sigh.

It’s been kinda hard to sit down and have something to actually say to the world in light of both the micro and macro world on fire-ness. But say something I must. Because if I don’t, then I’ll stop writing. And I promised myself that I’d write something at least once a month. This is an extension of the ‘one hour’ method, and an example of how sometimes you just have to board the struggle bus and power through something you’re not actually totally enthused about doing.

2. The Mongoose-uline urge to shut up and build

For a long time I’ve noticed that a lot of people seem to get all their dopamine from talking about doing things. They don’t actually do them. They just say they are doing them, or will do them. I suspect that this sort of person is driving a lot of the AI noise. AI tells them that they’re absolutely right, their ideas are revolutionary, and helps them shovel out things that look somewhat work shaped without increased effort.

I’ve never been that person though. I’ve always been the kind of person that wants to quietly go do things and THEN talk about them. It’s something I struggle with as a streamer too. Fortunately part of the goal of becoming a streamer is/was to push myself out of my comfort zone. So this month we’re going to talk about what sort of things I’ve been up to with all this extra “free” time.

Streaming more, obviously. This week being the exception, because as I write this I’m listening to my heater kick on repeatedly. Because I’m currently surrounded by an extremely slippery layer cake of snow and ice, after several nights of near record low temperatures.

But I’ve got a couple other projects in the works. One of them I’m not ready to announce yet. I’m still deciding where I want to go with it, and if I want to stream it. Probably is the answer to the latter. It would be good to devote extra hours to progressing that project, and maybe it will help find more viewers. The ones who don’t know what a MUD is.

Let’s talk about the other stuff I’ve been up to though!

3. The self-hosting that never was

A couple of years ago, when I still had a remote job that paid the bills, I started thinking about self-hosting things again. In hindsight, streaming was a mistake. Not me streaming, I mean streaming music/movies/etc as the primary method for consuming that content.

I’ve been a Spotify user for a long time now. But lately I’ve noticed a distinct dip in the quality of their client. Turning shuffle on from the Linux client was broken for a while. Adding songs to the favorites list was a little bit unreliable as well. I recently went through my liked songs and found a ton of duplicate songs. Not two versions of the song, but the exact same song twice. I have no evidence to point at to blame vibe coding, but it is a little sus.

There’s also the removal of content, or addition of more paywalls. Even before all of this I noticed songs were being removed from Spotify here and there. I also learned the hard way when I had Covid last year that I have a limited number of hours of audiobook listening in Spotify before they ask for more money. Last time I had a free month of Amazon Prime I tried to watch something on Prime Video and found they wanted me to pay extra for them. I don’t recall what show it was, but I browsed around a little bit looking for something else to watch, and it seemed like nothing I actually might want to sample was available without paying an additional fee. Hello, enshittification, my old friend.

All of this to say, I started assembling a plan to go back to the old ways. I was fairly early on in the process when I got the layoff notice though. So I shelved that project in order to avoid adding extra expenses.

But, like with all projects, I learned a lot about the current state of the ecosystem. Take heed, dear reader, do the stuff you are interested in. YOLO that homelab, build that app! Even if you don’t complete them you can almost always apply that experience to something else later on.

4. Embrace AI or get out

Last August was a big month for GitHub. First the CEO, Thomas Dohmke, made headlines for telling us all to either use AI or find a new career. Fuck you, Tom. Then he quit to be a founder of something, allegedly. I remember a flurry of additional headlines about who was supposed to be running things over there, but what I took away from all the noise was this: GitHub is going to become an AI focused platform. No, thanks.

There was also an announcement recently about pricing changes to GitHub actions. I don’t believe this will impact me as the only thing I actually use Actions for is deploying this blog, but I didn’t look too closely at the changes. I did see that they inspired a few announcements of companies moving off of GitHub. All of this served to remind me that there are indeed good alternatives these days! I particularly enjoy the fact that people are taking ole Tommy’s advice and getting out.

I was initially going to self-host my code on Gitea, since it was going to be just my stuff, and only for me, I wasn’t really worried about public access. A local network accessible instance and a backup would have served just fine. Now though, I find myself still wanting to allow access to view some of my stuff. Mostly for self-promotional reasons, but I also already have one tutorial course with a public repo that people can pull if they want. If I added more I’d also want them to be easily shareable.

So, recently I’ve been getting things set up to have a publicly viewable ForgeJo instance. I’ve also enjoyed the fact that there are a ton of new-ish but still reasonably mature tools written in Go to help with this task.

It’s not quite ready to share publicly, but as soon as I have things moved over I’ll update the links in all the places. Getting the t’s crossed and i’s dotted has consumed a lot of my non-stream time attention recently. Despite the fact that we have a lot of great new tools, there’s still a lot of gaps for this use case. For example: getting a server to send alerts to your email, for things like backup failures, is still kinda a pain in the ass without setting up an additional dedicated monitoring server.

I’ve also been really frustrated at how poorly documented ForgeJo actions are. This could fully be a skill issue, but they feel pretty half-baked to me. To the point where I may actually miss Jenkins.

Eventually I’ll want to migrate this blog off of GitHub pages as well, so I’ve laid the groundwork for that. It’s just a slightly lower priority at this time, because there’s other things that need to happen first/faster.

5. Income

I’ve also been working on the aforementioned secret product. I wrote a POC for part of it, and I think I want to make it a real product. I need to do the legwork to get things set up so that I can take that POC further and share it. Betting on myself is scary, but feels like the smartest play right now. Since I wrote the first draft of this blog Amazon has announced yet another round of layoffs. While I don’t particularly want to work in FAANG it does indicate that the job market is still on fire (in the bad sense).

I need to solve the income problem sooner rather than later, and I’m not optimistic about the short term job market. We haven’t yet hit the tipping point where companies are going to have to admit that AI slop and outsourcing to the cheapest possible body shop they can find have had negative consequences.

I’m still bullish on the future needing people that can actually write code to clean up the mess, but the job market today still sucks. Things probably have to get a little worse for a little longer. It would be nice if by the time they do I don’t need to care about the job market.

The past year of struggling has really adjusted my attitude on a number of things. Having a body of work to show prospective employers and observing the reactions has been educational as well. Even if this doesn’t work out as a primary income source it will add to my portfolio. Maybe that will help put me in front of the right people to have a real conversation instead of the clowns posting ghost jobs on all the job boards.