Practical (a.k.a. Actually Useful) Architecture • Stefan Tilkov • GOTO 2023

It's speech This is the last for Stefan Tilkov at the goto conferences, where he began to speak at 2012 year. I remember being in 2019 During the year, he personally attended his presentation on architecture in Berlin at the Software Architecture Conference. It seemed to me then that he was a very cheerful speaker and consultant, who would be speaking for many years to come, writing many articles and books. 2023 It's been a year. Goto did it. page in memory of his legacy, and then I'll tell you about his last performance in which he gave 10 Advice for pragmatic architectural work

1. Choose your perspective(s) consciously Stefan suggests choosing a perspective in a meaningful way. It shows different perspectives: domain architecture (with major domain services and junctions with external systems)macro-architecture (Main domains and their interactions within)microarchitecture (What technologies are used for these services). 2. Explicitly architect your team setup Stefan actually refers to Conway's reverse maneuver and goes on to mention Team Topologies. 3. Match your organizational setup to project size, etc. - I was talking about it. report At YaTalks, Stefan's focus is on what architecture looks like in each case. 4. Don't be afraid to decide things centrally Effective autonomous teams require a number of centralized decisions. (Which is ironic.). Stefan talks about what those decisions are. 5. Pick your battles wisely It’s about change management and choosing the changes you need right now. It is important not to make too many changes. 6. Enforce the least viable amount of rules, rigidly Here the author shows the importance of not having too many rules, otherwise we may lose flexibility and become too bureaucratic. 7. Balance prescriptive vs. descriptive architecture Here, the author says a good thing, that documenting architectural decisions is not the same as making them.) Often this documentation is the first step, but not the end goal. A good architectural solution can not be liked by everyone, and if you managed to make a decision that satisfies everyone, then most likely it is a bad decision. (You can watch a clip on this topic. B2 - Compromise) 8. Don't aim for perfection - iterate An iterative approach to improvement n 9. Architect for delivery flow as much as for runtime quality Stefan explains that when making architectural decisions, you need to look not only at functional and non-functional requirements, but also at how this affects the development processes. That's what I said in the report, too.How we changed the development of the best mobile bank to the requirements of business" 10. Be boring & do more with less Here Stefan talks about technological stability and the level of insanity of companies, which depends on the amount of money they are willing to spend on consulting:) In the end, the advice from Stefan sounds like this.

Prefer simple, straightforward solutions to overly clever and complicated, cool approaches and Focus innovation on the domain, not technology

Well, the final chord of the talk is devoted to the fact that architecture is still the most interesting area in IT because it matters - if you make it bad, everything will fall apart, but if you do it right, then success is not guaranteed:)

P.S. RIP Stefan. You were a great architect, consultant and speaker.

#Architecture #Software #Engineering