aivarsk
During SREcon EMEA observed how vendors compete to ingest more, reduce storage size and price. I decided to try another way and create an old-grumpy-dev version of OTEL that is simple and just stores everything to files for me to grep. I had some ongoing Python zipapp projects with SQLite and dummy HTTP server so I used that as the skeleton. Fun fact: Python ships HTTP clients, servers and the asyncio loop, but no async HTTP server or client. I stayed with the batteries included and relaxed durability on SQLite to reach 3500+ ingest requests per second. AI generated the dashboard for young-happy-devs to feel familiar. You can do curl+bash install (I wouldn't) but I added docker to make it safe, you can build it yourself and some curl or --demo to fill in sample data. Today added Sentry errors as jsonl files.
There are more otelites on github, this is my take on what "lite" means. That's my Futurama Popplers and Zitzels moment.
Modular Monoliths and Other Facepalms – Kevlin Henney – NDC London 2026
Code Cleanliness On the origins of 'clean code'
"Clean Code" a Timeless Truth or a Myth We Keep Telling Ourselves?
Solid is not so solid
TigerBeetle as a File Storage
Running TigerBeetle without a control plane database. Part one
The Lost Art of Semaphores
Pycon UK 2025: Principle Misunderstandings, Kevlin Henney [video]
Ring Buffer in the Database
Discussing "99 Bottles of OOP": Our First DNF (Did Not Finish) [video]
Solid Deconstruction – Kevlin Henney [video]
Handmade Hero – Getting rid of the OOP mindset [video]
A broker-less distributed messaging system from the previous century
Python JSON encoder apples and oranges and making it faster
Ledgers are simple, stop spreading FUD
Making sense of card range query EXPLAIN plans
Pessimism, optimism, realism and Django database concurrency
The COST of double-entry accounting with a PoC in Django
Not-So-Solid OO Principles
The O of Solid
The S of Solid
Building a Python ORM from scratch in 25 minutes
Building a Python debugger in 20 minutes
Accounting Systems Before TigerBeetle
Async Python is slower than "sync" Python under a realistic benchmark. A bigger worry is that async frameworks go a bit wobbly under load. https://calpaterson.com/async-python-is-not-faster.html