SShortSingh.
TopTechnologyAI & MLStartupsProgrammingScienceBusinessFinanceCrypto & Web3World
Back to feed
Programming

Developer Releases Browser Game for Practicing Vim Commands

0
·1 views

A developer has created hjkl, a small browser-based game designed to help players learn and remember Vim commands. The game is meant to be played casually as a way to unwind while getting some Vim practice. The creator says they plan to add many more commands over time. The project was shared on Hacker News, where it received a few points and no comments at the time of the post.

Read the full story at Hacker News

This is an AI-generated summary. ShortSingh links to the original source for the complete article.

Discussion (0)

Log in to join the discussion and vote.

Log in

Related stories

0
ProgrammingDEV Community ·

Cómo un espejo en Netlify mantiene webs accesibles durante bloqueos de LaLiga a Cloudflare

Las operadoras españolas bloquean durante los partidos de fútbol direcciones IP de Cloudflare por orden judicial, lo que deja inaccesibles también cientos de miles de webs legales que comparten esas mismas IP. Los autores de un curso sobre MCP describen cómo siguen operativos: durante cada partido el DNS se redirige automáticamente a un espejo estático en Netlify. El sistema consulta los horarios de Primera División, abre ventanas de bloqueo alrededor de cada encuentro y renueva los registros DNS y el certificado TLS de forma programada. Las rutas dinámicas del servidor MCP y el inicio de sesión se reenvían mediante un proxy firmado al Worker de Cloudflare, que solo atiende peticiones validadas. Los datos citados de OONI indican que, entre enero y junio de 2026, más de 500.000 dominios de Cloudflare sufrieron bloqueos en al menos una ocasión.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Seeding Random Number Generators to Make Game Bugs Reproducible

Randomized systems in games such as shuffles, spawns, and dice rolls often produce bugs that occur once and cannot be recreated, which leaves testers unable to prove what happened. The article argues that a generator's starting seed can restore repeatability, since identical seeds yield identical number sequences, though full replay also requires the same code, engine version, and input order. Recommended practice is to log the initial seed and the build for every run, so automated tests can record failing seeds and be re-examined on demand. The author notes that seeded determinism covers logic only and does not address timing, cross-device floating-point variation, or intermittent issues like race conditions. Beyond debugging, fixed seeds also let designers compare balance changes across the same sets of runs.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

L3G-World uses latent 3D Gaussian memory for faster real-time world generation

Researchers introduced L3G-World, a system that generates interactive 3D worlds in real time by storing VAE latent features in 3D Gaussians. The authors report that rendering these latents directly raises VAE-conditioning throughput by 349 to 359 times, avoiding costly RGB re-encoding used by earlier methods. The approach uses up to 38.6 times fewer primitives than the NeoVerse baseline and achieves 3.2 times faster memory updates. The evaluation reached about 9 FPS and was limited to two indoor datasets, leaving scalability to outdoor and higher frame-rate settings unproven. The method also depends on a pretrained VAE, which may constrain use with new asset domains.

0 comments Read more at DEV Community
0
ProgrammingDEV Community ·

Autonomous Coding Agents in Pure Python: AST Parsing, Self-Healing Loops and Sandboxed Testing

A DEV Community engineering article describes building a sovereign autonomous coding agent in pure Python, avoiding recurring SaaS fees for commercial coding tools. The approach pairs probabilistic LLM reasoning with deterministic checks, using Python's ast module to validate syntax, extract symbols, measure complexity, and flag dangerous operations. Test suites run in isolated asynchronous subprocesses with strict timeouts, so untrusted code cannot corrupt the agent's process. Failing test output is condensed into structured diagnostics, such as the failing test, exception type, and offending frame, to keep LLM context focused. The article frames this as a closed loop in which the agent runs tests, formulates repair hypotheses, and applies verified patches.

0 comments Read more at DEV Community