After oodles of research & experimentation, I've got a pretty good handle on what kind of input tends to produce the best results in AI coding. The hard part was coming up with a catchy mnemonic. Introducing C.R.E.S.S principles for context engineering. codemanship.wordpress.com/2026/05/04/c...

Product design students have been raising issues with us that the media is telling them they have to use AI and they’re not going to have jobs. Then when they go to use it, they find it gets things wrong, wastes their time, requires rework. They’re feeling gaslit and stressed out.

www.microsoft.com/en-us/micros... is it good to build an entire economy and software infrastructure on this

If you don't want GitHub Copilot trained on your code, you've been automatically opted in. You have until April 24th to opt out.

Spring's RestClient throws exceptions on HTTP errors by default, but that's rarely what you want in production. Here's a beginner-friendly walkthrough of error handling best practices so your API calls fail gracefully. https://youtu.be/MuYzEZk6-zI

I keep hearing that AI has made Agile irrelevant. Okay. I'll throw in the towel. Let's dump Agile altogether. So, instead, here is by far the best AI-focused way of working that I know: First, - The way we treat people and work with them is more important than processes and tools

I keep seeing the claim that "AI" multiplies the productivity of senior developers repeated as justification for FOMO. The best available data clearly shows that, in terms of *outcomes* at the *team* level, that multiplier is - at best - about 1.1x. And for the majority, it's more like 0.9x.

4. A brilliant concept from George Fairbanks: write code in a way that makes the architecture obvious. If the code you work in closely represents the intended architecture, the shared mental model of the entire team stays up to date. 7/9

github.com/making/nulla... A Maven plugin that configures ErrorProne and NullAway for nullability checking. It replaces the ~40 lines of maven-compiler-plugin boilerplate typically required to set up NullAway.

Whenever I mention Agile Software Development, inevitably someone will say "It can't be done" or "Nobody's done it". As someone who's spent more than 25 years turning "It can't be done" into "We did it!", I have a few thoughts on that. codemanship.wordpress.com/2026/02/12/a...

It's 25 years to the day that proponents of iterative, "lightweight" development methodologies came together in Utah to work on their ski tans and pen the Manifesto for Agile Software Development. I have thoughts, which I've typed out to make them visible. codemanship.wordpress.com/2026/02/11/t...

A large-scale study of developer IDE activity found that only 8% of devs who claimed they did Test-Driven Development did anything even close in reality. Were 92% lying? I suspect not. More likely, they just didn't know they weren't doing it, because they didn't know what TDD actually looks like.

Here are the (slightly revamped) slides for the JUG Switzerland and Voxxed Days Ticino editions of “Domain-centric? Why Hexagonal And Onion Architecture Are Answers to The Wrong Question – And What To Ask Instead”. #jugch #vdt26 🍃📦

Demanding an estimate guarantees the estimate will not be useful to the business and will always drive costs up. Software engineers all know that estimates are useless. 1/6

Modern microservices and moduliths can scale—but should they? 🤔 Christopher Curtin shows how Cell Architectures help you scale smarter, isolate clients, and manage cost and reliability at #Devnexus. devnexus.com/events/intro... 🎟️ Get tickets at devnexus.com 📨 Sign up atlj.ug/DevnexusConn...

We Need to Stop Calling Everything a Mock! | @emilybache.com 📽️ AVAILABLE NOW youtu.be/RvKPOjlQKyM

Let’s pause to reflect on that word, “iterative”. Iteration without convergence is chaos – literally. That rather begs the question, converging on what, exactly? codemanship.wordpress.com/2025/12/22/a...

"If your default choice is a distributed system, you’re likely doing it wrong." sashafoundtherootcauseagain.substack.com/p/the-case-a...

A "user story" (the unit of work) describes a user's problem, not a solution. Anything we build that moves us toward a solution to that problem is fine.

I would like to propose SOSS: Sponsored Open Source Software. If your company uses OSS, you sponsor the project $100/year. Not per seat, or per team, just $100/year, all in. - 30 developers using Ghostty? $100 - 300,000 engineers using tmux? $100, total

Upgraded my Spring Modular Monolith sample to Spring Boot 4, and Spring Modulith 2. github.com/sivaprasadre... #Java #SpringBoot #SpringModulith

Sadly, all the Learn Blender people were right. You couldn't pay me to go back to Maya or 3Ds Max. And I couldn't pay me either, because their license costs are insane Support Blender today!!!

Software Design Is NOT Software Architecture (& That’s A Good Thing!) | @emilybache.com 📽️ AVAILABLE NOW Watch HERE ➡️ youtu.be/idGM8fvtT8U

Wow! @adambien.blog held a magnificient talk + live coding tonight and showed us why #Java is the most productive platform for LLM-Driven Development. My key takeaways 👇

We all adhere to a *system* of work. That system may or may not be chaotic, but it's a system nonetheless: a collection of components and subsystems that work in concert to achieve a single goal—in our case, to produce a product. 1/10

Mapstruct is nice when it fits your code, but these days I prefer event-sourcing. Completely sidesteps the need for mapping! #EventSourcing

Spring Boot 4 is moving to Jackson 3 (already in M3). 👉 Check out the Jackson 3 Migration Guide: github.com/FasterXML/ja... 👉 Also useful, the OpenRewrite recipe: docs.openrewrite.org/recipes/java... More insights coming soon in the official Spring blog post next week.

#JUnit 6.0.0 is released! ✨ Java 17 and Kotlin 2.2 baseline 🌄 JSpecify nullability annotations 🛫 Integrated JFR support 🚟 Kotlin suspend function support 🛑 Support for cancelling test execution ⏭️ New `--fail-fast` mode for ConsoleLauncher 🧹 Removal of deprecated APIs docs.junit.org/6.0.0/releas...

I think I unconsciously perceived it as subject.verb(object), however "In most OO languages, you read it backwards: object.action()." Think that's where Manager classes came from.

#java JEP 401 on value classes is submitted : openjdk.org/jeps/401

Putting all your controllers in one directory and your views in another Is like putting the cortisone and hemorrhoid cream in the drawer with toothpaste because they're all tubes.

The Brutal Secret To All Prioritization Schemes My latest video short for the Modern Software Engineering channel @modernswe.bsky.social www.youtube.com/shorts/go4rW...

"Can we make this smaller?" is a very different, and much more useful, question than "How big is this?" People say that the discussion surrounding estimation is more important than the estimate itself. 1/5

Yet another defense of story points over on the Lean-and-Agile group [https://t.ly/BrLp9] got me thinking I should repeat the following as a public service announcement: Re story points (vs story counting) don't work. Story points require estimation. 1/9

@maven.apache.org 4 (from 4.0.0-rc4 on) contains the Maven Upgrade Tool with which you can automatically upgrade your Maven project. I finally found time to write a small article about it's feature and how to use it. Please test the tool and give feedback :) maven.apache.org/tools/mvnup....

80% of estimates are wrong in the best case. It's just math. Assuming you have good-faith estimates, the map of real-to-actual time forms a normal bell curve: 40% or so will be early, 40% will be late, and the top of the curve is within acceptable limits. 1/7

Executives: if I'm working for you I want you to delegate management & retain leadership. Definitely not the other way around.

The key characteristic of Agile development is delegated management. Any framework that offers full supervision to upper management is not Actually Agile. (definitions from Robert Austin: www.dorsethouse.com/books/mmpo.h...)

When a “stand up” involves people working on different goals/projects reporting their status, that is just a status meeting.

Manchmal sind es die kleinen Dinge, die einem das Leben versüßen oder auch 100. @bahn.de @tagir-valeev.bsky.social

The technique problem with legacy code is that since it wasn't written with tests, it's hard to write tests without changing the code, and it's scary to change working code without having tests. Writing tests last forces you into this legacy code situation.

The DORA research found that there is NO TRADE-OFF BETWEEN SPEED & QUALITY Teams with good scores on both, spend 44% more time on new features!!! Read “Accelerate” by Nicole Forsgren et al. 4/4

Feels like 10 years at least since I RT’d posts like this on test-driven development (been more like 2 or 3 really). Why TDD Doesn’t Lead to Dumb Code by @kentbeck.com tidyfirst.substack.com/p/why-tdd-do...

What is design, really? Grady Booch offered a timeless, elegant answer in 2006 that still resonates deeply in software, systems, and beyond. A thread 🧵 on “On Design”: web.archive.org/web/20141105...

wrote up some thoughts on how our interfaces are getting flatter, and how we might design with 💃 our bodies 🤸 in mind. 🧶 went ham on little yarn worlds wattenberger.com/thoughts/our...

Here's my problem with Microservices. 🧵 1/9

Let me see if I've got this straight: you don't trust your developers to make decisions, but you trust hallucinating autocomplete? Righto.

I have just merged into Spring Framework main branch a first-class support for programmatic bean registration for Java and Kotlin. Will be part of the upcoming Spring Framework 7.0.0-M3.

Interesting links about SAFe and why it is not agile.

The A-Z of Code Craft: A is for Agility codemanship.wordpress.com/2025/01/21/a...