实时热榜
daily.dev · Popular
- 01Why use ORMs if LLMs write code?An ORM skeptic argues that with LLMs now capable of writing boilerplate code, the traditional justification for using ORMs (avoiding tedious SQL mapping code) largely disappears. The author contends that ORM query optimization already requires deep expertise and workarounds, undermining the promise of abstraction, and suggests it may be simpler to have an LLM write parametrized SQL directly rather than fight with an ORM's implicit API, while cautioning about SQL injection risks.
- 02WebSocket vs SSE vs Long Polling: The Real Cost of 1,000 EventsA hands-on benchmark built with a Node harness and a byte-counting TCP proxy measures the real wire cost of WebSocket, SSE, and long polling delivering 1,000 events. WebSocket used 119,692 bytes, SSE 131,596 (10% more), and HTTP/1.1 long polling 884,698 bytes (7.4x), mostly re-sent headers. On HTTP/2, long polling drops to 182,475 bytes (1.56x payload). Latency differences at low event rates were under 0.3ms across all three; long polling only degrades when events arrive faster than a round trip
- 03Neon Functions: backend logic next to your dataNeon has launched Functions, a serverless Node.js 24 compute layer that deploys directly onto a Neon Postgres branch in the same region as the database, with DATABASE_URL and other backend credentials (Object Storage, AI Gateway, Auth) injected automatically. Unlike typical Lambda-style handlers, Neon Functions are long-running, supporting minutes-long agent tool loops, WebSockets, and SSE, and they branch alongside the rest of the Neon stack - creating a child branch spins up an isolated copy o
- 04Blocked aria-hidden: The Warning is Right, and Every Fix You’ve Found is WrongThe Chrome console warning about 'blocked aria-hidden on an element because its descendant retained focus' is correct, not a nag, and every popular workaround for it — blur() with no follow-up, setTimeout delays, stripping aria-hidden, or modal={false} — actively harms screen reader users by leaving focus stranded on or removing the modal focus trap entirely. The root defect is an ordering bug: code hides a region (via aria-hidden) while focus is still inside it, creating 'ghost focus' where a c
- 05GitHub - debba/free-my-arch: Interactive disk cleanup skill for Claude Code on Arch Linux. Scans package caches, dev junk, Docker leftovers and oversi...free-my-arch is a Claude Code skill for Arch Linux that automates disk cleanup through a four-phase process: scan, report, confirm, and clean. It targets package caches (pacman, AUR/yay), dev caches (npm, pip, cargo, Go), Docker leftovers, and oversized home directory files. Nothing is deleted without explicit user approval — cache groups use a multi-select prompt, personal files are confirmed individually. Safety rules prevent glob-based rm, protect configs and browser profiles, and flag app da
- 06Transition Kit: Opensource library of page and theme transitionsIntroducing Transition Kit Opensource library of page and theme transitions, currently in beta: • 30+ transitions available • 4 theme toggle components with inbuilt theme transitions • install the theme toggle components via shadcn registry Try it here: link Please give a ⭐ if you like it GitHub: link
- 07HTTP QUERY Shipped. Your Cache Did Not Get the MemoRFC 10008, published June 2026, introduces the HTTP QUERY method — the first new HTTP method since PATCH in 2010. QUERY is safe, idempotent, and cacheable like GET, but carries a request body like POST, solving the long-standing problem of structured search payloads too large for query strings. The critical infrastructure catch: caches must key on the request body and related metadata, which no browser and few managed CDNs (e.g., CloudFront only allows a fixed seven methods) support by default y
- 08Shape how the future daily.dev features look likeWe’re exploring a new way to help developers discover and keep up with the technical content they care about. We’d love to understand how you search, learn, and stay informed today and what you’d want us to build next. The survey takes around 4 minutes.
- 09I burned $157,143 in tokens last month. I paid $400.A developer shares that they processed $157,143 worth of tokens in a single month while only paying $400, using two $200/month subscriptions — OpenAI's Codex plan and Anthropic's Claude plan. The core argument is that AI labs are currently subsidizing usage heavily due to competitive pressure, and developers should take advantage of this pricing gap before it closes. The advice is simple: stop treating tokens as scarce, use flat-rate subscriptions aggressively, and ship projects now while the ec
- 10Why Go is an Ideal Language for AI-Assisted Software EngineeringAs AI coding assistants shift software engineering from writing to reviewing code, Go's design philosophy becomes increasingly valuable. The post argues that Go's strict compiler, unified toolchain (gofmt, govulncheck, native fuzz testing), opinionated simplicity, and strong backward compatibility make it uniquely suited for AI-assisted development. Key points include: Go's static type system catches AI hallucinations at compile time rather than runtime; its comprehensive standard library reduce
- 11The Appwrite CLI is now written in GoAppwrite has rewritten its CLI from TypeScript/Node to Go, shipping a single native binary with no runtime dependencies. Key improvements: startup time drops from 207.6 ms to 11.1 ms (up to 22.5× faster), installation shrinks from 330 npm packages (209 MB) to 2 packages (13 MB), and peak memory during push drops from 283.5 MB to 28 MB by streaming archives via io.SectionReader instead of buffering. The binary size goes from 66 MB (Bun) to 14 MB. All user-facing interfaces — flags, exit codes, ap
- 12Day 13/30 — Roles Are Not Enough“Admin” and “User” work until the business asks a real question. Can a finance manager refund this invoice? Can a tenant administrator edit users from another tenant? Can an author update only their own draft? Roles describe broad identity. Permissions and policies describe actions on resources within a context. A stronger model checks: actor + action + resource + context Avoid scattering user.role === ... checks throughout controllers and services. When did your application outgrow basic role-b
- 13What I learned from vibe coding a small web project with GitHub Copilot and GPT-5 miniA developer shares firsthand experience vibe coding a small web project using GitHub Copilot with VS Code Agent Mode and GPT-5 mini, on a stack of Go, SQLite, React, Redux, and Sass. The positives include fast boilerplate setup, dependency installation, and incidental learning of CLI tricks like lsof and curl. The negatives are significant: the AI struggled with more than a handful of moving parts, hallucinated fixes (nearly wiping a database), installed outdated dependencies, and produced code
- 14Rust is the future of LinuxGreg Kroah-Hartman, the Linux kernel stable branch maintainer, has publicly endorsed Rust as the future of the Linux kernel. The core argument is not just memory safety, but that Rust's compile-time enforcement of data validation, memory lifetimes, locking rules, and error handling reduces the burden on the kernel's ~150 maintainers who review code from ~5,000 contributors. Greg claims 80% of Linux security vulnerabilities would have been impossible if the code were written in Rust. The current
- 15TanStack Did It Again... They Fixed ChartsTanStack Charts is a new JavaScript charting library using a grammar-of-graphics approach (similar to ggplot2 in R), treating charts as compositions of marks, scales, and mappings rather than fixed chart types. It supports ~55 mark types including bars, lines, areas, sankey diagrams, tree maps, radar charts, world maps, and more. Key features include SVG and canvas output, animations, tooltips, interactions (brushing, zooming, keyboard selection), export to SVG/PNG, and a three-level styling sys
共 15 条热搜查看完整榜单