Switching to Signed URLs Killed My Image Cache — Decoupling expo-image Keys from the URL
Signed URLs rewrite their query string on every expiry, so a URL-keyed cache never hits. Here is how to pin the key with source.cacheKey, what the cache read and write APIs actually return, and what changed when I re-ran the measurements.
Cutting MCP Tools Didn't Make Anything Lighter — 2,377 Bytes of Definitions vs 110,298 Bytes of Response
I wrote a minimal MCP server for a wallpaper catalog and measured the byte cost of tool definitions against the byte cost of responses. Here is which side actually matters, and the real reason to merge tools.
Deciding overlay text legibility at ingest time instead of on device — four metrics measured side by side
Moving the question of whether text stays readable over a wallpaper out of the device and into the content pipeline. Four candidate metrics measured across 240 images, including what downscaled judging actually computes.
A JSON File 19% Smaller That Ships Only 3% Smaller — Measuring Catalog Payloads Against gzip
Five payload formats measured on a 20,000-item wallpaper catalog. Key shortening saves just 3.2% after gzip, while a columnar layout cuts 26.8%. Real numbers for Brotli, paging, and JSON.parse time.
When Two-Character Queries Silently Return Nothing: Measuring Japanese Search Indexes in an Expo App
SQLite FTS5's trigram tokenizer returns zero rows for Japanese queries shorter than three characters, without raising anything. I benchmarked linear scan, a bigram inverted index, and FTS5 over a 20,000-item catalog to find the real threshold.
Every Experiment Kept Landing on the Same Devices: Hash Choice and Salt Position, Measured Across a Million IDs
On-device experiment assignment looked fine in isolation, then collapsed the moment two experiments ran side by side. Here is what one million IDs revealed about four hash functions, why the culprit was the concatenation order rather than hash quality, and the implementation I settled on, with the measured numbers.
What Renovate may bump in an Expo app, and what it must never touch
Turning on automated dependency updates in a Rork-generated app also hands Renovate the 123 packages Expo SDK 57 pins. Measured on 2026-07-30, six of them sit a full major version behind npm latest. Here is how to generate the ignore list from the SDK instead of maintaining it by hand.
Counting what prebuild --clean will erase before you upgrade to Expo SDK 57
A raw diff between two generated ios/ trees showed 649 changed lines; only 3 were real edits. How to count what prebuild --clean erases, and move it into a config plugin.
When to Raise Your Minimum iOS Version — Count Leftover Branches, Not User Percentages
Judging a minimum OS bump by usage share produces the same answer every year, so the decision never happens. Here is the annotation convention, the sweep script that counts how many iOS and Android branches each candidate floor would retire, and what to watch for 30 days after.
What Filtering Out Prerelease OS Devices Cost Me — Separating Telemetry Tracks Instead of Dropping Them
Filtering prerelease OS devices out of telemetry left me with no warning signs on GA day. Here is the rebuild: an os_track dimension, a remote GA baseline, split alert thresholds, and a release gate that separates blockers from fixes.
Collapsing Duplicate Requests Into One: A Reference-Counted Single-Flight Layer
When several components fire the same API call at launch, you get a burst of identical requests. Here is a single-flight layer that shares one in-flight promise instead: how to build the key that decides the folding, the trap of handing out a failed promise forever, the trap of one caller's abort cancelling everyone, and the test that keeps it all from regressing, with the real network numbers alongside.
Resolving App Config in Three Layers: Merging Defaults, User, and Remote With Bounded Overrides
A single type-safe layer that merges compiled defaults, user preferences, and remote config. So a broken remote value never takes your app down, each key gets its own override strength, plus schema validation and range clamping, built from a real production incident.