Developer Blog for Real-World Programming
Four tools solve four different problems that all get pitched as the same problem. Here's what each one does, where each one breaks, and the combination that covers a real homelab without paying an SSO tax you didn't see coming.
A growing list of open-core self-hosted projects, from Grafana to Planka, gate single sign-on behind an Enterprise tier. Here's which protocol actually gets paywalled, and what to check before you deploy.
Andon Labs' AI agent Luna runs a real San Francisco store with real employees, and it just recommended terminating one of them. The headline undersells what actually happened, and what it doesn't prove.
Same standard as everywhere else on SudoSecurity — traced to source, verified before it publishes. Nothing here is sponsored.
Four tools solve four different problems that all get pitched as the same problem. Here's what each one does, where each one breaks, and the combination that covers a real homelab without paying an SSO tax you didn't see coming.
A growing list of open-core self-hosted projects, from Grafana to Planka, gate single sign-on behind an Enterprise tier. Here's which protocol actually gets paywalled, and what to check before you deploy.
Forum Archaeology #2. Three free PHP forum platforms started from the same premise a generation ago. One got hit with a CVSS 9.8 authentication bypass this June. One had eleven CVEs last year. One has stayed almost quiet. Same premise, three completely different fates.
A logic bug in screensharingd let attackers skip authentication entirely and land as root — no password required. Here's how it was found, patched, reverse-engineered, and finally weaponized.
Forum Archaeology #2. Three free PHP forum platforms started from the same premise a generation ago. One got hit with a CVSS 9.8 authentication bypass this June. One had eleven CVEs last year. One has stayed almost quiet. Same premise, three completely different fates.
A PHP API case study: fixing an N+1 query, twenty serial cURL calls, missing connection pooling, and stats recalculated on every request with Redis caching.
A recipe card with blank ingredient fields is worse than no recipe card at all. Part 3 of the kitchen series covers PHPDoc, comments, and the documentation debt nobody notices until the person who understood it walks out the door.
Escoffier's brigade de cuisine solved the same problem PHP's class visibility solves: who gets to touch what, and who answers to whom. Part 2 of the kitchen series, and the one where we build something real.
Python 3.15 release candidate: explicit lazy imports, the new frozendict and sentinel types, the Tachyon profiler, and real JIT benchmark numbers.
A technical deep dive into the features shaping the next Java feature release, due
Escoffier's brigade de cuisine solved the same problem PHP's class visibility solves: who gets to touch what, and who answers to whom. Part 2 of the kitchen series, and the one where we build something real.
Linus built Git in 10 days. GitHub became where open source lives. Then Microsoft paid $7.5B, killed Atom, trained Copilot on your code, corrupted main branches with a merge queue bug, and folded the whole thing into CoreAI. Here's the full story.
Andon Labs' AI agent Luna runs a real San Francisco store with real employees, and it just recommended terminating one of them. The headline undersells what actually happened, and what it doesn't prove.
Game AI isn't intelligent. It's state machines, search, and scoring arranged to feel like judgment, with real, runnable code from FSMs to local LLM dialogue.
Ghost Check #1: the Puppet Master hacked minds by exploiting the gap between what a system experiences and what it can verify. Thirty-one years later, a researcher did the same thing to Microsoft 365 Copilot with a single email.
Building a local AI server is more than installing Ollama. This guide builds a full production AI stack with Open WebUI, LiteLLM, Qdrant, Traefik, and Langfuse — what each component does, why it's needed, and three complete Docker Compose configurations for beginners, power users, and small teams.
Eight topics, one editorial standard — browse it all, or see what we'd call our best work.