A SaaS marches towards production; defining an engineering culture.
By Jeffrey M. Barber
Good news!
I’m now testing the integration of the components that I’ve been talking about by building a staging/integ environment. This new environment looks and feels like the future production site except I play fast and loose. As I move fast, I reveal problems which will be problematic, and I’m taking notes. Fortunately, most things are working as intended. My gossip is gossiping. My routing tables are routing. Deploy kind of works.
Gossip as a Failure Detector
By Jeffrey M. Barber
The other day, I sat down and wrote a gossip failure detector for the Adama SaaS that I’m building. What I love about gossip is how resilient it is to failure. Currently, the gossip failure is working in unit tests and a few small scale tests. I’m running out of innovation tokens, but I feel confident implementing a gossip detector from scratch.
Errata on the WebSocket Woe
By Jeffrey M. Barber
Woe (noun, literary): great sorrow or distress.
There’s great fun when you post on HN and get those precious internet points in bulk, but the real value comes from more data. There were things that I failed to mention along with confusion.
The path towards reliability (and scale) for great sleep
By Jeffrey M. Barber
As I have committed towards an open source SaaS, I am designing how to handle the failure modes since I like sleep. I’ll have much to say about the role of design on great sleep, but today I’m looking at the path towards launch and asking if there is an opportunity to provide trade-offs to the future consumer as this unfolds along with the related business model.
Woe unto you for using a Websocket
By Jeffrey M. Barber
Woe (noun, literary): great sorrow or distress.
Anyone building products on the web with any interactivity, collaborative, real-time, or reactive features will realize that we live in a cold dark world of anguish. My team’s SOSP21 paper on BladeRunner glosses over the depth of this topic, and today I intend to walk through my thoughts as I must build yet another streaming proxy. Perhaps, this wandering of thoughts helps you on your journey as well.
I retired so I can focus on my open source project
By Jeffrey M. Barber
Adama is about to get a whole lot of love, and I intend to talk strategy today. As a side project, I could wander the desert for a while without much progress. It was easy to forgive as there was other shit distracting me, and my role as principal engineer was already hard enough.