> I re-read this comment and it's embarassing I ever have to write this, because after 15 years of Redis history at such scale and popularity, pretty much every

> I re-read this comment and it's embarassing I ever have to write this, because after 15 years of Redis history at such scale and popularity, pretty much everybody that was seriously exposed to Redis knows those stuff. Is tech culture really degraded so much that we have to restate the obvious? I know you wrote OG and did a lot of Redis but... Yes, Tech culture is that fucked. In my past job I was hired semi-specifically to deal with a concern, namely that their use case 'fit well with Kafka' but the latency for their case sucked at least as much as the API pain. (Yes I did something better. No, IDK if folks will ever see it). Now? Now I spend my days trying to 'paper-over' patterns that drive me to insanity just trying to make it work from a 'people need to learn why things work on a starship' level [0]. On a real level you didn't fail, Redis has lots of great patterns. On a -practical- level it's a shitshow because you now have lots of folks 'glue-gunning' the Redis API on use cases that probably need tweaking or aren't the right fit at all, alas they all worked off the same example on GH/SO/etc and then did their own "this wasn't even the right way to do this so I'm adding glue, what could possibly go wrong" case. (That said, Nats has decent stuff for this in form of KV CompareExchange style APIs, and I see the inspiration there, so that's something to feel good about.) [0] - Namely, if anyone has a good prompt for taking a photo of someone and doing img2img of 'Person in astronaut uniform preaching from an ivory tower', that would be a plus