If I go far enough back, to when I was in middle and high school, I "played" with computers for fun. That mostly meant making goofy little BASIC programs to do random stuff, and it was rewarding to see code turn into something useful. Before I got to college, I became totally infatuated with broadcast work, radio and TV, and mostly let the computer interest subside or sit in the background, even as the machines started infiltrating all of the broadcast equipment. Obviously I went back to computers when it was clear the Internet wasn't going away (and felt like another form of media to me), and within four years out of college, I learned enough to make software my profession without the formal education.
That fun hobbyist approach led to the sites that have been with me for two decades, and I've always worked on these on my own time. It was always supposed to be a fun distraction, but also a natural way to keep up on new technology that I might not get to work with in my day job. This became really useful as I started to pivot away from heads-down coding jobs, because at many points in the last six years, I've written as little as no code as part of my job, and in others, was there largely to validate and guide the work of others. You keep street cred when you're still in it in some fashion.
For most of that time, I was able to use a combination of my forum project and the sites to stay at the very latest versions of Microsoft's dev stack, even when we might be behind a version or two in my work. But about four years ago, two things started to make it hard to keep up in a fun way. First, Microsoft basically rebooted their entire platform, which was good because they were open sourcing it all, but bad because it was a constant moving target. Every few months they'd change things enough that trying to get any non-trivial project into the current preferred state was a hassle. There were also a lot of things missing. Concurrently, front-end technologies were changing just as fast, and there were (and still are) several dominant frameworks that are competing for your attention. I think I've been involved in evaluating those at least three times in the last few years, and while I have a cursory understanding of them, I'm not fluent in any of them.
In my day jobs, the amount of context switching from business execution to technical overlord (I kid, it's not like that) gets to be a little exhausting at times. As much as I've tried to focus my career on the execution and leadership angle, company and/or team size has often required that I also be technical and write code. So combine that situation with all of the rapid change, and it shouldn't be surprising that trying to maintain my own stuff, the stuff that's supposed to be fun, isn't really very fun. I suppose maybe that's why your hobby should be different than your work.
But during my period of involuntary non-employment, I had some time to get back to it, and the platform has largely stabilized. In recent weeks, I was able to get the forums and CoasterBuzz up to the latest bits. This means I can probably do a real release of the forum app soon, which is really just a port from the old platform, and not modernized, but it's a step in the right direction. I can't even tell you how good that feels, because for the first time in years, it's fun. It stopped being fun because it was hard to move it forward, but now it's moving again, and that's exciting. Even the build automation behind it, which I rarely ever messed with for my hobby stuff, is working. That's pretty cool.
I've had some serious challenges lately, and I needed something to be fun and winning. The timing is excellent. There may be some truth and wisdom to varying your hobby from your work, but I think it's just different enough that both can be awesome, even if they are related.