Incremental Design, DevOps, Microservices & CICD • Michael Nygard & Dave Farley
Keywords
Summary
160 words
Critical Evaluation
The interview provides valuable insights into the practical challenges of managing data at scale in a microservices architecture. Michael Nygard’s experience at Nubank offers a concrete case study, illustrating how principles of modularity and coupling apply to data. The discussion is technically sound, with both speakers demonstrating deep expertise. However, the format is conversational, which limits the depth of analysis. Some topics, such as the specifics of their ETL pipeline, are only briefly touched upon. The sources cited are primarily the speakers’ own books and websites, which are authoritative in the field but not independent. The title accurately reflects the content, and the conversation stays on topic. The main strength is the practical, real-world perspective, which is often missing in theoretical discussions. The main weakness is the lack of formal structure and the absence of concrete metrics or case study results. Overall, the interview is a valuable resource for practitioners, offering actionable ideas and reinforcing established principles.
157 words
Title / Content Match
The title accurately reflects the content, which covers incremental design, DevOps principles, microservices, and CI/CD, as discussed by the two experts.
Quality & Reliability
8/10
The interview features two recognized experts in software engineering, Michael Nygard and Dave Farley, discussing practical approaches to data management and microservices. Their insights are grounded in real-world experience at Nubank and extensive industry knowledge. The conversation is technical and specific, with no unsupported claims. However, as an informal interview, it lacks formal citations and peer review, so a score of 8 reflects high credibility but not absolute certainty.
Key Moments
Markers derived by PSI from the transcript: the creator did not define chapters.
- Introduction and context of the interview at GOTO Copenhagen 2024.
- Discussion of Nubank's challenge with 300,000 datasets and the need for incremental changes.
- Exploration of modularity, cohesion, and separation of concerns in software and data.
- Introduction of data domains and their alignment with domain-driven design.
- Discussion on microservices and data ownership, including automatic ETL generation.
- Challenges of coupling between microservices and data extraction.
- Plans for source-aligned data products and giving developers control over data representation.
- The need to merge operational and analytical worlds through domains.
- Discussion on platform engineering and the difference between complex subsystems and true platforms.
- Conclusion and final thoughts on continuous delivery and incremental design.
Cited Sources
- Dave Farley's website — Dave Farley's personal website, likely containing information about his books and consulting.
- Michael Nygard's website — Michael Nygard's personal website, likely containing his blog and publications.
- Dave Farley on Bluesky — Dave Farley's Bluesky profile, a social media presence.
- GOTO on Bluesky — GOTO conference's Bluesky profile.
- GOTO Copenhagen 2024 — The official website for GOTO Copenhagen 2024, where the interview took place.
- Session abstract — The abstract for the session, providing more details about the talk.
- GOTO Tech — GOTO's website for tickets and events.
- Dave Farley on LinkedIn — Dave Farley's LinkedIn profile.
- Continuous Delivery Ltd. — Dave Farley's company website.
- GOTO on LinkedIn — GOTO conference's LinkedIn page.
- Michael Nygard on LinkedIn — Michael Nygard's LinkedIn profile.
- GOTO Conferences YouTube channel — The YouTube channel where the video is published.
Concurring Sources
- Release It! 2nd Edition by Michael Nygard — Michael Nygard's book on production-ready software, likely supporting his views on design and reliability.
- Continuous Delivery by Dave Farley & Jez Humble — The foundational book on continuous delivery, supporting the DevOps practices discussed.
Contribution & Novelties
The interview provides a practical perspective on applying software engineering principles to data management, particularly through the concept of data domains and automatic ETL generation. It offers a real-world example from Nubank, illustrating how to handle massive data sets incrementally. The discussion also critiques the term ‘platform engineering’, distinguishing between complex subsystems and true self-service platforms.
Pour aller plus loin :
- Data Mesh — A concept by Zhamak Dehghani, relevant to the discussion of data domains.
- Domain-Driven Design — The basis for the ‘domain’ concept mentioned in the interview.
- Continuous Delivery — A core practice discussed, with Dave Farley as a pioneer.
102 words
Radar Profile
The radar profile shows high scores in quality of information and reliability, reflecting the expertise of the speakers. The quantity of information is moderate, as the interview format limits depth. The technical level is high, suitable for experienced practitioners.