Feed on “architecture”
10 posts tagged architecture. See the whole feed.
Why Open Quest Format keeps dialogue on its own rig, apart from the quests
In Open Quest Format, conversations run on their own model. A quest step points at a dialogue by reference and only receives the flags and events the dialogue drops on the wire, which is what lets Ink and Yarn Spinner jack in.
What Skyrim and Baldur's Gate 3 quests share under the chrome
A run through the Open Quest Format model, choom: how steps wire into a flat graph, how objectives count events, how payouts and endings hit the wire as events and flags, and why so much of the gig is left to the engine on purpose.
Proving a gig can be finished before any merc jacks in
Open Quest Format keeps its condition language to eleven operators and zero scripts, so that the validator can sniff out unreachable steps, dead ends and gigs with no reachable ending before a single merc jacks in. This post walks through how the scan runs and how many eddies that limit costs, choom.
The rig that sat dark all day
NWF Sync is a four-route read-only protocol for pulling feed journals between machines, cut for the one job direct fetching cannot pull: items that scrolled off the front page while your rig was powered down.
One changelog file, three surfaces it hits
One typed changelog file feeds an in-game badge, a Settings tab and a public website page, so release notes cannot drift between the product and its pitch.
Rigging a game to behave like a desktop widget
Cozy Coast is an Electron and Phaser game built as a borderless desktop widget first, choom, which changes the window model, the render budget and the packaging pipeline.
Cozy Coast multiplayer, zero servers, straight peer to peer
Cozy Coast shipped a multiplayer hub with no backend at all, choom. Steam handles lobby discovery, connection brokering and packet relay, and the host relays angler state.
Multi-agent workflows one netrunner can actually reason about
A tight multi-agent rig one choom can trace beats a sprawling swarm nobody trusts. Scope each agent, wire tools explicit, log everything, and the crew ships preem.
Evals for delivery crews: trusting AI output without re-scanning every line
Trustworthy AI output comes from evals wired into the real pipeline like ICE, not a benchmark: eval what matters, gate at the right spot, and keep review habits that scale for the whole crew.
Building Neurowire: one canonical rig, six feed formats
Neurowire treats feeds like a data-modelling job: one canonical rig every parser feeds and every serializer reads, so bolting on a format is a single serializer, choom.