31
Strangler pattern, a powerful simple concept to do refactoring
Inspired by the strangler figs in Australia, a strangler fig application gradually draws behavior out of its host legacy application
Martin Fowler
A website on building software effectively
- Popular posts
- 84
- Total points
- 17,806
- HN comments
- 10,168
- Rank
- #33 personal
Visit martinfowler.com → On Hacker News Jun 2009 – Oct 2026
Software Engineering 66Career & Management 20Tech Industry & Business 15AI & Machine Learning 13Society, Policy & Culture 10Systems & Performance 8
32
The LMAX Architecture - 100K TPS at Less than 1ms Latency
LMAX is a retail financial trading system that can handle that can 6 million orders per second on a single JVM thread. The business logic runs in-memory surrounded by disruptors using event sourcing.
33
Microservice Trade-Offs
The microservices architecture can be useful, but to know if it's suitable in your context you need to understand its costs and benefits.
34
The Future of AI Software Development
Detailed report on the Thoughtworks AI retreat, should we write a manifesto for AI-enabled development, is spec-driven development the return of waterfall, LLMs perform better in healthy codebases
35
Building your own CLI coding agent with Pydantic-AI
How to build a CLI coding agent
36
Building reliable agentic AI systems
AI helping pharmaceutical researchers query decades of information buried in PDF reports
37
The one true REST explanation
Leonard Richardson's model for understanding a RESTful architecture as steps in a maturity model. Levels are (1) resources, (2) HTTP verbs and return codes, and (3) hypertext controls.
38
What I'm Up to Now
a short post
39
Exploring Mastodon
With the current uncertainty over Twitter, I'm starting to explore using Mastodon
40
Current thoughts on social media
My view on the state of social media in late 2023
41
TeamTopologies
A model that helps people design effective teams for software development.
42
Design Token-Based UI Architecture
Describing established patterns for organizing design tokens and practical approaches for automation
43
Building Boba AI: Lessons learnt in building an LLM-powered application
A series of patterns for building an AI co-pilot application.
44
A Guide to Threat Modelling for Developers
Threat modeling is a risk based approach to cyber security requirements analysis.
45
Privacy Protects Bothersome People (2013)
We need to support privacy, not for those of us who have "nothing to hide", but for bothersome people like investigative journalists and activists, without which our democracy would crumble
46
Your Organization should run its own Mastodon server
Run your own Mastodon server to control your social media presence
47
Ship / Show / Ask: A modern branching strategy
Ship/Show/Ask is a branching strategy that helps teams wait less and ship more, without losing out on feedback.
48
On Pair Programming
A guide to effective pair-programming
49
Martin Fowler on Software Patents
a bliki entry for Software Patent
50
Microservices and the First Law of Distributed Objects
In P of EAA I said "don't distribute your objects". Does this advice contradict my interest in Microservices?
51
Conway’s Law (2022)
Systems are constrained to follow the communication patterns of their designers.
52
DataLake
A Data Lake is a store that hold raw data as a source for data scientists to explore ways to gain information. It should not be accessed by end-users or used for system integration.
53
The Learning Loop and LLMs
How continuous learning shapes effective software development with LLMs
54
GUI Architectures (2006)
Describing the way architectures for UIs have evolved over the years, in particular the many and often misunderstood definitions for Model-View-Controller.
55
Martin Fowler: An Open Letter to Pearson about SOPA/PIPA
To: John Isley, Executive Vice President, Higher Education & CEO Pearson Technology Group
56
The consequences of task switching in supervisory programming
the future of senior developers, junior developers, more on cognitive debt, DevEx versus AgentEx, the role of IDEs, consequences of task switching in supervisory programming
57
Understanding Spec-Driven-Development: Kiro, Spec-Kit, and Tessl
Notes from my colleagues on AI-assisted software delivery
58
Cannot Measure Productivity
It is impossible to measure the effectiveness of tools and techniques in software development because we have no way of measuring our output, and thus our productivity.
59
One Line of Code That Compromises Your Server
Demonstrating how a weak session secret can allow an attacker to take control of a server
60
DSLs Enable Reliable Use of LLMs
Why Domain-Specific Languages and good abstractions make LLM code generation reliable and reviewable.