Last week I was preoccupied with SGDQ so didn’t write anything — this week was genuinely exhausting in a nice way.

• SGDQ was tonnes of fun, and in particular Ocarina of Time: Triforce% is one of the genuinely coolest things I’ve ever seen. Also after contributing for years, for the first time one of my donations got announced out in front of 100k people: It felt weird.
• During the work week, the recently formed API Rationalisation team ran a in-person office-hours where they talked about their current thinking. Really cool, not only because it’s similar to the approach we took in FT Specialist and it’s always nice being validated, but because their communication style is brilliantly engaging & transparent in a way unlike anything I’ve ever experienced. Hoping for an eventual write-up!
• Every year the FT does rounders in Regents Park and this year I went along for the first time. Genuinely fantastic seeing so many people (many for the first-time) and discovering how competitive some teams were. Highlights included light gossip, discovering a secret jägerbomb bar in the bushes and being massively sunburnt. Fairly sure I got recruited as a wedding DJ.

The area I’ve been working on — enabling parallel streams of product development — is taking shape: Our ideals & constraints greatly mirror Sam Newman’s principles of microservices and coincidentally some suggestions from a recent workshop are similar to current thinking on micro-frontends, leading to an interesting aside:

• Somehow synonymous with “multi-framework”, micro-frontends have become a slightly controversial topic, however, the two ideas are unrelated. Micro-frontends involve similar decisions & patterns to those of micro-services, with similar pitfalls. If you’re using a separate design-system or component library, what you’re doing is probably considered micro-frontend development.
• This estate already deploys micro-frontends, enabling autonomous product-teams who own vertically partitioned front-end micro-services. At its core, the challenge can be reframed as improving the ability for product development to happen horizontally whilst protecting this gift of autonomy. Naturally technology doesn’t negate communication & workflow improvements and so those are happening too.
• Like all early thinking in the open, this may be naive or otherwise wrong, so next steps include much more research. The field has come some way beyond where it was years ago, encouraged by improvements in Web Components and the rising popularity of Islands Architecture. More organisations like Zalando, DAZN & edX have adopted some manner of micro-frontend architecture and are reporting highly positive outcomes!

At the start of the year I promised that I wouldn’t try to learn a language, but quickly broke that promise to restart Japanese. If it goes anywhere I’ll write up my study-plan - most recently I’ve begun focussing on grammar & vocabulary.

• I’ve relearnt Hiragana: Granted I hadn’t completely forgotten them, but coming at it “from scratch” has made me learn some of the more visually similar letters which previously tripped me up like: め (me) and ぬ (nu), or, ち (chi) and さ (sa).
• Grammar can be difficult in any language and Japanese vocabulary is famously expansive. With Bunpro & Wanikani respectively I’m now making progress — through the magic of structured spaced repetition over small breaks in the day (already spent reading and doing crosswords).
• So anyway, at time of writing according to Wanikani I now know 12 of the kanji. Only ~20K more to go!