This page was translated from the original by AI.
As 2025 draws to a close, looking back, I realize I made quite a few major decisions this year. Some of them look good in hindsight, some are still playing out, and some won't show their results for a while. Today I want to talk about the key choices of this year, and what each one cost me.
Buying a house in Japan, selling the one back home
The biggest decision of 2025 was buying a house in Japan. It wasn't just a financial decision—it was more about psychologically admitting: I plan to stay here long-term. After buying, I decided to sell my property back in China. That was actually a tough call, because I took a significant loss on the sale. I ran the numbers several times, and every time I wondered whether I should wait a bit longer. But in the end, I sold it. Why? If I kept it, the price would keep dropping, and I'd still be paying the mortgage every month—a considerable burden for me. After selling, I felt lighter. No more juggling two places, no more worrying about the "what ifs." That feeling is pretty good.
RESULT
Chose: putting down roots in Japan, focusing on the present
Gave up: the so-called "safety net" and "flexibility"
Gained: a sense of psychological stability
Tech stack: from Rails to Phoenix and back to Rails
I'd already admired both Rails and Phoenix before, and had built some projects with each framework. Since the UI I write on my own tends to come out pretty ugly, early this year I started trying Rails + Inertia.js + React + shadcn/ui for new projects. It felt great at first, but once I got into forms, it became a hassle. So much boilerplate, way more complicated than Rails' built-in form helpers (though to be fair, it might just be a skill issue on my part). Around the same time, Phoenix released a new version (1.9?) and introduced its official MCP, so I gave that a try. It was fun to write, but LiveView didn't click with me the way Hotwire does. So eventually I circled back to vanilla Rails + Hotwire. For UI, I'm currently using DaisyUI—not exactly my taste, but there aren't many convenient options right now. Good enough for the time being.
This process taught me something: tools aren't inherently good or bad, only fitting or not. For me right now, Rails' simplicity and familiarity matter more than Phoenix's "better architecture." Maybe that'll change someday, but for now, this is where I stand.
RESULT
Chose: familiarity and productivity first
Gave up: chasing novelty and "looking cooler"
Gained: faster development speed, less mental overhead
Career: promoted to Engineering Manager
I got promoted to Engineering Manager in 2025.
Honestly, my feelings about this promotion are complicated. When I was a Team Leader a few years back, the feedback I got was that my management and reporting skills weren't up to par. That feedback was fair—I really just wanted to write code back then, and I half-assed the management side.
This time is different. I've started to understand what management actually is. I bought Become an Effective Software Engineering Manager and I'm reading it, and I've been writing team documents, thinking about how to do this role well.
But I'm still not sure I'll be good at it. And I'm not sure leaving pure technical work is the right call. I've just decided to give it a shot.
Current state: sometimes excited, sometimes anxious, and most of the time both at once.
RESULT
Chose: accepting the challenge, stepping out of my comfort zone
Gave up: the "security" of a pure technical role
Still learning: what makes a good Engineering Manager
Writing: from biweekly to monthly, plus daily journaling
Mid-year, I started publishing a blog post every two weeks. After a few months, I found it exhausting. Around that time, I switched to monthly posts, prioritizing quality.
That adjustment worked well. The pressure dropped, but the quality of the posts actually went up (I think). Each piece gets enough time to think through and polish.
Another change: I started keeping a journal. Every day I write something by hand—not for publishing, just for recording. Sometimes it's just "tired today," sometimes I write a lot. Blogging and journaling are two different kinds of writing. Blogging is public; it needs structure and logic. Journaling is private; you write whatever comes to mind. I need both.
RESULT
Chose: a sustainable pace
Gave up: the "sense of achievement" from frequent publishing
Gained: better writing quality, more honest records
Life: travel, games, enjoying the moment
I went to quite a few places this year: Matsumoto Castle, Kusatsu Onsen, Oarai Aquarium, Niigata Prefecture, the Ghibli Museum, and Mount Tsukuba. Next year, I'm planning to visit the Nintendo Museum.
I also played a lot of games. I managed to snag a Switch 2 (got lucky with the preorder), and played Donkey Kong Country Returns HD, Hollow Knight: Silksong, Hades II, and Xenoblade Chronicles X.
None of these things are "productive," and they don't improve my skills, but they make life interesting. I no longer feel like I need an excuse to relax.
RESULT
Chose: enjoying life
Didn't give up: work and learning—just not letting them take over everything
Gained: a more balanced state
Plans for 2026
I didn't set any goals at the start of 2025, and it turned out I made a lot of big decisions anyway. So for 2026, I'm not going to set overly specific goals either, but I do have a few directions:
Work:
Figure out what kind of Engineering Manager I want to be
Keep my technical skills sharp, don't drift away from hands-on work
Writing:
Keep the monthly blog going
Make it to year two of journaling
Read more, not just technical books
Life:
Visit the Nintendo Museum
Keep exploring Japan
Play games without guilt
Mindset:
Accept uncertainty
Stop demanding the "optimal solution" for everything
Showing up consistently matters more than occasional perfection
Summary
The biggest takeaway from this year: when making a choice, you don't need to be sure it's "right"—you just need to commit to making it right.
Some choices look good in hindsight, some I don't know yet. But at least I'm making choices, not endlessly weighing options and putting things off.
Continuing into 2026.
Loading comments…