#223 — End of 2022 Special

Dec 25, 2022

Merry Christmas! The end of the year is upon us once more!


This year I changed jobs. After after 4 years at GDS, I moved to GoCardless.

Over 2021 I started to feel like I needed a change. I’d been doing the same sort of thing for a while and, even if I changed teams, I’d still be doing something fairly similar. I wanted to work on larger and more complex systems, and I didn’t want to have my career advancement contingent upon doing more line management.

So I started looking around. I sent a total of 8 applications, of which 2 never responded, 5 rejected me before an interview, and 1 rejected me after an interview.

Then after all that work of tracking down and applying to interesting-sounding companies (I don’t just want a job, I want an interesting job), a GoCardless recruiter reached out to me over linkedin and that’s how I ended up at the new place.

I’ve been there for 9 months now, working on a team which owns a lot of the core logic and data which powers our payment processing. Right off the bat I was thrust into complex scalability work, so we could handle millions more payments a day than we currently processed, and it’s been a lot of fun.

It also pays more, which is nice.


As interesting as my job may be, it is still an unwanted imposition upon my life, an evil necessary to survive in our modern society. Were I financially independent, I would drop it in a heartbeat.

So, finances.

This year I didn’t really make any changes to my practices. I changed some account names up to make budgeting easier, but that’s it. I’m pretty happy with how I model things now, and don’t see myself making any big changes next year either.

I’m closer to financial independence (FIRE) than I was at the start of the year, but there’s still a long way to go.

Firstly, I’m now saving / investing over 50% of my net income. This is, really, the bare minimum to achieve FIRE, as it’ll let me FIRE in about 16 years. The next milestone I’m aiming for is 66%, which would let me FIRE in about 10 years; I’ll need to increase my income further to have a chance of hitting that, though.

Secondly, my “runway”—how long it would take to run out of cash if I quit my job this second, sold all my investments, and didn’t change my spending habits one bit—is now 3 years. That’s a long time to not need to work! But I shouldn’t relax now and actually quit my job for an extended break, that’d just eat up my savings and delay the goal of permanent retirement.

Knowing I’ll be working for 16 more years, at least, isn’t very encouraging. Likely longer, if inflation remains so high and wages don’t keep up. That’s half a lifetime away! But it’s better than working an entire lifetime.


This year I read 62 books, which is nothing to sniff at—it’s, on average, just over one book a week—but looking at what I actually read, it’s only that high because I read a lot of light novels. If we look at non-light-novel prose fiction specifically, which used to be the bulk of my reading, I read a mere 17 books this year! There were 15 weeks in which I didn’t finish any books at all!

Despite having so many books, quite often I felt as if I had nothing to read. No more big series I’d been meaning to get into. Just lots of individual books, some of which I’m really not that keen to read, and which have been sitting around unopened on my shelves for years. I should probably get rid of them. I’ve also not bought much new fiction this year (if you don’t count the light novels), but that’s partly because there’s not been much I’ve wanted to read.

I think I need to discover some new authors. Time to go hit up an actual physical book shop and see what’s recent.

All that said, here’s what I read this year:

My top three books read in 2022 are:


It’s been a year of changes, but few radical changes, just iterative improvements.

In October I moved these weeknotes from memo.barrucadu.co.uk to this domain. I wrote about why I did that at the time, so go read that.

I’ve continued tweaking how I run my computers. I…

I also started using some new software and stopped using some old software:

The biggest change is that I dropped Pi-hole when my Raspberry Pi stopped working, and wrote my own recursive DNS resolver with adblocking, which has been powering my LAN DNS since mid March. It was a cool project which taught me that DNS is, in fact, fairly simple.


This was a good year for gaming.

Computer Games

I mostly play roleplaying games these days, when I pick up a computer game I tend to get really into it for a few weeks, then burn myself out and never touch it again. But in those few weeks it’s really fun! This year I’ve particularly enjoyed:

  • Dyson Sphere Program

    I just like setting up production lines. But, at least when I played the game, there wasn’t really an end goal beyond “build a dyson sphere”. Which I did. And the moment I achieved that difficult and fun goal, the game lost all meaning and I couldn’t even muster up the motivation to start anew with a different approach, since I’d just be working towards something I’d already achieved once before.

    I have that problem with these sorts of games a lot.

  • Cult of the Lamb

    I liked the cutesy / culty mix of this game, and the roguelite gameplay was fun and engaging. I got up to the final boss battle, but found it just a bit too bullet-hell for me so I’ve kind of stalled there.

  • Crusader Kings 3

    Not a new game, but one I only really figured out to competently play this year after reading Brett Devereaux’s series on what sort of history the game models, which helped me to understand some of the mistakes I’d been making, and got me itching to play again.

  • Victoria 3

    Another Paradox grand strategy game, though of a different bent to CK3. I’ve not played much of this yet, only the tutorial, but that was fun and I think I just need to give it a proper go.

  • Dwarf Fortress

    I like colony-sim games. I’ve put a lot of hours into Banished and Patron, and I’m no stranger to Dwarf Fortress. It’s the ultimate. The ideal colony-sim. But it’s just so daunting!

    Until now, that is. The Steam release is out, with a GUI (!) and a general reorganisation of things to make the game possible to pick up without inhaling the entire wiki up-front. Finally, I’ve managed to do more than just the quickstart gudie, and had a fairly successful fortress of around a hundred dwarves. It did get up to close to 200 at one point, which was causing issues with the food supply, but I exiled all the unhappy dwarves and lowered the population cap for a few years while I built things up.

    The only thing is… the game seems not very difficult actually? Before I intentionally tunneled into the underworld and lost the fort to demons, my militia was pretty good and could easily take down Forgotten Beasts and Titans. To defend against sieges (not that any happened), I could just raise my drawbridge. I was totally self-sufficient, I just ignored the trade caravan when it came. And even when I did tunnel into the underworld, only two demons came out! A far cry from the “hundreds” the wiki says. However, two were enough.

    I think my next fort needs to be somewhere more exciting where there’s actually an external threat I have to defend against. Maybe I should embark on a haunted glacier.

Roleplaying Games

Another good year for roleplaying games. I’ve played 8 systems and run another 5 (unless I’ve forgotten some we just used for a one-shot or something), with a total of 20 people across 4 groups:

  • Tuesday Group

    • Run: Ars Magica
    • Played: Eclipse Phase, Castaway (one-shot)

    Unfortunately, I ended up leaving this group back in March. After I finished running the Ars Magica campaign, someone else started an Eclipse Phase campaign but ultimately we decided nobody was really into it, and so I bowed out when they decided to start a new campaign of D&D 5e (since I was already in two other D&D-like games at the time).

  • Work Group

    • Run: Old School Essentials

    I started this group after some not-very-subtle hints that a bunch of people really wanted to play D&D but nobody was willing to run a campaign. So I started a fortnightly game with 6 players that was supposed to be easy for people to drop in and out of, since I didn’t expect attendance to be very good for an after-work game.

    Unfortunately, attendance was even worse than I expected, and after several frustrating instances of cancelling due to only 1 (or 0!) players being available, I decided to drop the campaign and we’re going to regroup in January to see who wants to play and can commit.

  • Saturday Group

    • Run: Old School Essentials, Traveller (one-shot)
    • Played: Ecopunk 2044, Monster of the Week, Simple World, Stars Without Number

    My first group, still going strong.

    We did go through a rough patch where a bunch of sessions got postponed or skipped on short notice, but we solved that by switching to a more open schedule: when you have the next session of your campaign prepared, you can claim the next available Saturday on the list, and you can’t claim more than one Saturday in advance. Ostensibly, we’re now playing four campaigns, but in practice it’s just mine and one other alternating weekly, with one of the other two GMs occasionally running a single session of their campaigns.

  • Sunday Group

    • Run: Forbidden Lands (one-shot), Traveller, Wicked Ones
    • Played: Genesys (one-shot), Whitehack

    My second group, also going fairly strong. This year one player dropped out, and we haven’t managed to get a replacement yet. Though, recently, a friend has been listening into our sessions. But we’ve not managed to convince her to actually join in as a player yet.

    It used to be me and another GM alternating games, but the other GM had to take a break so it’s just been me for a while. She doesn’t seem to have any plans to start a new campaign soon, so I guess it’s just me running weekly games for the foreseeable future, which is fine by me.

Right now I’m running two campaigns:

  • An OSE Dolmenwood hexcrawl for my Saturday group, in which the players are cartographers charged to map out the forest and learn its secrets. Which really is just a good excuse for them to wander around and get into adventures. Dolmenwood is a really great setting, it’s not just your standard D&D-flavour fantasy, it’s really evocative, weird, and fairytale-like. And the books are just so well-written and easy to use, I barely need to prepare anything beyond finding or reskinning the occasional dungeon to drop in at a convenient location.

  • A Traveller “Milieu 0” campaign for my Sunday group, in which the players are private contractors doing scouting and surveying work of systems which haven’t been visited since the collapse of interstellar civilisation some 1,500 years ago. All set against an exciting backdrop of the formation of the Third Imperium and the interstellar power-shakeup that results. We only just started this a few weeks ago, and have got through the first adventure (a murder mystery in which they had to prove their innocence), and will be picking this up again on the 8th of January with a survey mission.

I really like both OSE and Traveller. They and Call of Cthulhu are definitely up there as my favourite systems.

Of the other systems I’ve run this year:

  • Ars Magica was just too complicated. I feel you’d need to basically study the rulebook to be able to run or play it without needing to constantly look things up.

  • Forbidden Lands didn’t click for me. I didn’t like the hexcrawling or stronghold rules, which are basically the only reasons you’d use it over another fantasy system.

  • Wicked Ones just wasn’t really my thing. Too narrative, actually challenging the players (without just deciding up front that they were going to lose) was really difficult.

Still, I don’t regret trying any of them.

Finally, this year, I switched from Roll20 to Foundry. It felt a bit silly paying $9.99/month to Roll20 to get… frankly, not very much. I’d used Foundry as a player a fair bit, so I decided to take the plunge and try it out as a GM. It worked very well, and I’m now happily self-hosting it.


I got back into baking bread this year. I kind of fell out of the habit when flour first became unavailable in 2020, but now I’m making nice loaves of sourdough every week or so. Also, I do sourdough now, previously I used instant yeast.

Here’s my recipe.


  1. Mix half cup rye flour and half cup water in a glass jar, lightly cover with a tea towel and leave at room temperature for 24 hours.
  2. Discard half the mixture, mix in another half cup rye flour and quarter cup water, re-cover and leave for another 24 hours.
  3. Repeat step 2 for 4 days.


  1. Prepare dough:
    1. Mix 6 cups bread flour with 2 cups water, 2.5tsp salt, and half cup starter.
    2. Lightly cover dough with a tea towel and leave at room temperature overnight.
  2. Proof:
    1. Move dough onto a lightly floured tea towel, and shape into a tight ball.
    2. Leave dough-ball at room temperature for 2 hours.
  3. Pre-heat oven to 200C with a dutch oven (or cast iron pan) (with lid) inside.
  4. Bake:
    1. Move dough ball from tea towel into hot dutch oven (with lid), and bake for 45 minutes.
    2. Remove lid and continue to bake until golden brown (10 to 15 minutes).
    3. Remove bread from dutch oven and leave to cool.

If your kitchen is cold, for the proofing step you can fill the sink with warm water, put the tea towel-wrapped dough-ball in a bowl, and float the bowl in the sink.