819 karma · joined October 20, 2022
We have a monorepo, we use automated code generation (openapi-generator) for API clients for each service derived from an OpenAPI.json generated by the server framework. Service client changes cascade instantly. We have a custom CI job that trawls git and figures out which projects changed (including dependencies) as to compute which services need to be rebuilt/redeployed. We may just not be at scale—thank God. We're a small team.
- portable, reproducible environments: I suppose you could achieve this with a docker setup. If I jump to a different workstation, I want to be able to load my current project without playing setup wrangler.
- license management like a dotfile database: all my licenses are fettered to the wind across two or three email addresses, and every time my PC crashes (twice in the past 5y) or something breaks I have to go recollect them. It’s a quest. Quests suck.
- remote or cloud processing: connect to your workstation if you’re on the road, or a cloud cluster running DAWs. Sure, this introduces some lag which can force you into drag to piano roll contexts, but other times you’re just messing with mixes. But gaming giants like Sony figured it out for PS4/5 titles.
- shareable projects with some innovative business solution to the license barrier. I want to be able to access somebody else’s project and load it in its entirety—whether I have Neural DSP’s latest Archetype or not. Whether I have Serum2 or not. Apple managed to get bands onto iTunes. We seriously can’t get Waves or Neural DSP to go the same route? Some royalty-based approach?
I know these are outlandish requirements in some scenarios but I feel like this would fix the misery of music making in the era of Windows and macOS being goblinware operating systems.
A todo list that feeds into a calendar (with a high degree of flexibility) that feeds into a tada list.
I’ve been working on this casually for the better part of a year and hope to release something that is home-hostable later in 2026, once I’ve lived with it for a bit.
Like I don’t want to remember that I need to do something, I just want a time slot in my calendar that auto populates against a set of constraints, so I can go from moment to moment in a flow. I want to be able to control that todo list and my goals’ time/space constraints with natural language—not some godforsaken form that gives you carpal tunnel. And then I want to see how my progress grows and how much I’m committing to what I want to do.
And like the proton, this statement is somehow heavier than the entire article. What an absolutely bizarre, arrogant choice of words.
“I’m not defined by my career Greg”
“Greg see a therapist”
But I’d be glad if somebody was competitive enough to force Elon to behave again.
When you pay an arm and a leg every month for Netflix, you're funding a studio that pumps out the equivalent of screeching goblin sounds. Makes no sense.
-Iran
For example, I work in a small company with a data processing pipeline that has lots of human in the loop steps. A monolith would work, but a major consideration with it being a small company is cloud cost, and a monolith would mean slow load times in serverless or persistent node costs regardless of traffic. A lot of our processing steps are automated and ephemeral, and across all our customers, the data tends to look like a wavelet passing through the system with an average center of mass mostly orbiting around a given step. A service oriented architecture let us:
- Separate steps into smaller “apps” that run on demand with serverless workers.
- avoid the scaling issues of killing our database with too many concurrent connections by having a single “data service”—essentially organizing all the wires neatly.
- ensure that data access (read/write on information extracted from our core business objects) happens in a unified manner, so that we don’t end up with weird, fucky API versioning.
- for the human in the loop steps, data stops in the job queue at a CRUD app as a notification, where data analysts manually intervene.
A monolith would have been an impedance mismatch for the inherent “assembly line” model here, regardless of dogma and the fact that yes, a monolith could conceivably handle a system like this without as much network traffic.
You could argue that the data service is a microservice. It’s a single service that serves a single use case and guards its database access behind an API. I would reply to any consternation or foreboding due to its presence in a small company by saying “guess what, it works incredibly well for us. Architecture is architecture: the pros and cons will out, just read them and build what works accordingly.”
https://lukemuehlhauser.com/bostroms-unfinished-fable-of-the...