lomereiter

It's not just memory savings, it's also about CPU cache. A frequently accessed tree/graph-like data structure can fit into L2/L3 cache so much easier if it uses 32-bit offsets instead of pointers.

IMO the actual difference between introverts and extroverts is that the latter use @here and @channel a lot more frequently

...or Korg NTS-1 (or even both together), which is not as modular but fun and hackable in its own way

The community can agree to always call it "Redpanda Connect™ (former Benthos)"

Don't you ever get inspired by reading good books? You can learn new tricks and apply them elsewhere, and it's much easier than actually contributing to a big open-source project.

If this were indeed the average attitude, there wouldn't be this big of an outcry with regards to the move from fully open-source to source-available licenses (Mongo/ELK/Hashicorp/etc.)

"Large teams" is the key, there has to be enough services and/or language diversity to justify the extra complexity.

To set up an open-source service mesh, the infra team anyway has to configure a private certificate authority and cert-manager to create k8s secrets for the service mesh components. From there, it's straightforward to extend the common deployment template (hopefully there is one) to mount a volume with an auto-rotated certificate. All an application developer has to do is to use that certificate, which is much less effort than what you are implying.

Reminds me of a Russian joke.

A secretary applies for a job; the interviewer asks her: "In your CV you claim that you can type 1000 characters per minute - for real?!" "Yes!", she replies, then adds in a low voice: "but such nonsense comes out..."

Arrow format is not intended for storage, it's for in-memory data exchange between different libraries and languages.