Practical notes on design and code.
AI
Use `response_format` when the model should answer in a fixed schema. Use function calling when the model should ask your app to do work. This article shows the contract boundary, a local request-shape test, and the failure cases where the wrong choice adds confusion.
About 10 min read
Programming
`filter(Boolean)` removes falsy values at runtime, but TypeScript still needs a callback shape it can treat as a narrowing predicate. This article shows the difference, proves it with a small compiler fixture, and explains when a truthiness filter is the wrong tool because `0` or `""` would disappear too.
About 6 min read
CSS
See why `content-visibility: hidden` skips rendering without collapsing the box, then compare it with a visible state and the cases where `display: none` is the real fix.
About 6 min read
Coding
Angular's `track` expression decides which DOM row belongs to which item. See why unstable keys recreate rows, how to preserve input state, and where the fix stops helping.
About 6 min read
Programming
`Promise.any()` gives you the first fulfillment, but it does not convert a fully failing batch into the first rejection. This article shows why all failures are collected, when `AggregateError.errors` matters, and how to test the boundary.
About 9 min read
AI
`previous_response_id` can continue a conversation without replaying the whole transcript, but it does not carry every kind of state. This article shows what the API keeps, what it drops, and when you still need your own transcript store.
About 8 min read
CSS
Set `min-width: 0` on the shrinking flex item so the filename can clip instead of forcing the row open.
About 8 min read
Programming
`Promise.allSettled` keeps one result slot per input promise, so completion order and result order can differ. See the guarantee, the reason it exists, and a regression test that catches the wrong assumption.
About 10 min read
DevOps
Use `concurrency` to keep release branches in order and let preview branches cancel stale deploys, then verify the policy with a local regression test.
About 14 min read
CSS
Show why a clipping wrapper can steal sticky behavior, then compare `overflow: hidden` with `overflow: clip` in a controlled card fixture.
About 7 min read
CSS
Compare `overflow: clip` with a real scroll container, then verify why a clipped panel hides content without exposing a usable scroll offset.
About 8 min read
CSS
Reserve gutter space for the scroll container so the panel keeps the same usable width before and after it starts overflowing.
About 6 min read