Platform Strategy - Gregor Hohpe & James Lewis - GOTO 2024 (Category Architecture)

It's interesting. interview The guys from the GOTO book club with Gregor Hohpe, who recently wrote a very interesting book, "Platform Strategy." Gregor's a grated kalach, and I'm kind of already. told He is the author of the book, the author of the book.Enterprise Integration Patterns" (my post about the book), "The Software Architect Elevator" (my post about the book), "Cloud Strategywhich I have not yet read andPlatform strategyThe one I’m reading right now and the one I’m talking about.

The main points in the interview are as follows: Platforms are important because they help reduce the cognitive burden on engineers. Classification of platforms: base platform (like AWS)As well as custom platforms within companies, which should not repeat AWS, and close specific scenarios for the company. Gregor wrote a paper on the subject, "Build abstractions not illusions." told. In general, platforms can create illusions of simplicity, but can actually be complex and require decision-making. Platforms should provide useful abstractions for app creators and not be misleading. Here the author recalls the book "The Software Architect Elevator" and says that if the correct use of the platform requires diving into the details of the implementation of abstraction, then something is wrong with the platform. Gregor talks about renaming various systems into platforms. And in order to understand the platform, you need to remember that platforms don’t have to do everything for you, they have to make things easier. When building platforms, it’s important to communicate with development teams and observe what they’re doing to understand how they solve problems. Gregor often helps teams understand what they’ve actually done and how it can be used in other contexts. Metaphors can help people think differently and increase transparency. In fact, Gregor himself uses automotive metaphors to talk about platforms - the automotive industry has long begun to use platforms for the chassis, and on top sculpted slightly different bodies. The cost of creating a chassis is actually divided between different models of cars. Good platforms are harmonizing the foundation, and on top of it, they allow innovation to blossom – that’s what happened in the world of cars, and that’s what we see in the world of cloud platforms. Standardization can be a tool for innovation, but it is important not to make it an end in itself. It’s just a tool to achieve goals. When we think about change, we need to understand how restrictions that are being removed can affect people’s worldview and work. This transition to a new paradigm can be difficult for people, but it can lead to new opportunities and limitations. At the same time, it is important to ask questions about what new constraints come with new technologies. When creating an internal development platform in a company, you need to create a platform that will be useful for business, not just for engineers. It should focus on what is important to the business, not the technical aspects. The economy of an in-house platform may be worse than that of a cloud provider due to its smaller scale. In addition, it can be related to questions of the value and durability of products.

In the future, Gregor plans to write a book about API strategy and integration, which will consider boundaries and help people internalize those concepts. This will be a continuation of the book.Enterprise Integration Patterns"which came out bigger" 20 years ago.

#Conference #PlatformEngineering #SystemEngineering #Software #Architecture #DistributedSystems #Management