Dec 5, 2018 · 22m · mad
A New Kind of Logging System // Zach Sherman & Ben Johnson, Timber (FirstMark's Data Driven NYC)
gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions
At Data Driven NYC, Timber.io co-founders Ben Johnson and Zach Sherman discuss the architectural challenges of handling high-volume telemetry data and present the concept of an observability data routing layer. They demonstrate how decoupling data streams into specialized storage engines dramatically improves system reliability, operational flexibility, and infrastructure cost efficiency.
How this conversation actually went
Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. Matt holds 7.5% of the talking time here. How this is scored →
speaking balance: gold is Matt, purple is the guest (3 minute bins)
Responding to an audience question, Ben directly reframes expectations, clarifying that Timber does not offer processing guarantees as durability relies on Kafka.
Hardest push from Matt ▶ 20:15 Matt asks guests to ground complex technical terms for audienceMatt pushes past the dense data engineering talk, steering guests to explain fundamental definitions like metrics so everyone in the room can learn.
Biggest teaching moment ▶ 20:25 Ben educates on how aggregated metrics cut storage costsBen explains to the host and audience how rolling raw events into single numeric aggregates radically cuts storage and processing expenses.
Matt holds his own ▶ 17:36 Matt frames the burgeoning complexity of data engineeringMatt displays domain familiarity by highlighting how tool proliferation created the whole specialized discipline of data engineering, which the guest validates.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | Matt as informed peer | Guest teaching | Guest disagreement | Matt pushing back | Why |
|---|---|---|---|---|---|---|
| The Scale and Challenges of Telemetry Data | 0 | 2 | 0 | 0 | This is a solo presentation monologue by Ben Johnson outlining Timber's scale and logging challenges. The host does not participate during the presentation, so host metrics are zero. | |
| Diversifying Data Demands & Navigating Tool Overload | 0 | 2 | 0 | 0 | Ben continues the presentation monologue, explaining the shift from Elasticsearch to multi-storage routing and referencing Matt Turck's landscape slide. Host does not interact. | |
| Defining Routing Layer Requirements | 0 | 3 | 0 | 0 | Zach Sherman takes over the presentation monologue to explain technical routing requirements and control/data plane concepts. Host is silent during the presentation. | |
| Timber's Modern Routing Architecture & Cost Reduction | 0 | 3 | 0 | 0 | Zach presents Timber's modern stack built with Rust and Lua, along with dynamic configuration benefits. Host remains silent as this concludes the talk. | |
| Real-World Use Cases, Key Takeaways, and Presentation Wrap-Up | 3 | 3 | 0 | 1 | Matt Turck conducts a live Q&A session, guiding the talk to cover basic definitions, ideal customer ICPs, and industry trends. The dynamic is fully collaborative and informative. |