dmeijboom

My point of view (after 18 years of programming): DO use frameworks (compile-time checked queries if you can) but skip ORMs that hide/obfuscate SQL completely as it will result in slow queries, extra round-trips, etc

Makes sense. At brainpod.io we moved from Next.JS to Deno Fresh. Not 100% happy but definitely an improvement. Also via Vite (since V2).

I loved NextJS when it was relatively new but with turbo pack, the app router (and the insane amount of bugs that came with it) it's no longer my go-to frontend framework.

We built Steiger after getting frustrated with Skaffold's performance in our Bazel-heavy polyglot monorepo. The main pain points were:

The TAR bottleneck: Skaffold forces Bazel to export OCI images as TAR files, then imports them back into Docker. This is slow and wasteful

Cache invalidation: Skaffold's custom caching layer often conflicts with the sophisticated caching that build systems like Bazel and Nix already provide.

Currently supports:

Docker BuildKit: Uses docker-container driver, manages builder instances

Bazel: Direct OCI layout consumption, skips TAR export entirely

Nix: Works with flake outputs that produce OCI images

Ko: Native Go container builds

Still early days - we're planning file watching for dev mode and (basic) Helm deployment just landed!

Happy user since day one. Since adopting Orbstack most of our frustrations with Docker on Mac OS are gone.

Thanks! I was looking for this blog post for a while now

I don’t get the appeal of Traefik. If you want an easy to use reverse proxy that works well, pick nginx. Want something simple for self-hosting? Take a look as caddy. For Kubernetes, try out Envoy Gateway.

Try OrbStack ;) It’s the only docker daemon alternative for Mac that makes working with docker acceptable.

We’ve been using Zalando’s Postgres Operator before in production and we recently switched to cloudnative-pg. We didn’t experience any issues so far and I’m a big fan of their design choices (where you have a single user, single database for each micro-service that requires one).

This happened a while ago and was related to the trademark/copyright issue. It’s an immature move and I can’t take this seriously. All this does is creating unnecessary drama surrounding the Rust foundation.

I was literally looking for a lightweight Elasticsearch alternative for a couple of days now... This is great, will set it up asap.