← Feed

Posts

• View post

  The Claude Code issues page, together with analysis of the leaked source code, reveal a dumpster fire of a code base, and the Anthropic jobs pages reveal a rapidly-growing demand for firefighters. When they tell you how to run coding agents "while you sleep", ask if they mean *reliably*? Or will you be waking up to your own dumpster fire?

• View post

  "It's only a prototype". Sure. But what's the purpose of a prototype, and do you think bugs and basic UX design failures will help you achieve that purpose? You want feedback on the soup? If there's a fly in it, that's the feedback you'll get. No point asking "Too much salt?" 'cause they won't remember that detail. Prototypes created to get user feedback need to be reliable enough and usable enough tha...

• View post

  The average corporate AI software development pilot goes something like this: "We're going to run a lap of the office in flip-flops, and then we're going to give our recommendations on whether flip-flops are appropriate footwear for running marathons"

• View post

  Fascinating https://www.sri.inf.ethz.ch/publications/egashira2026testsuitebench

• View post

  Closed systems don't learn. They just get more and more internally consistently wrong.

• View post

  "It's just a pilot. It doesn't mean we're going all-in with AI coding. We're just experimenting." When was your pilot for specification by example? Pull scheduling? Trunk-based development? Teaming? Event storming? Static analysis? Property-based testing? Why does *this* specific approach get a team and a seven-figure budget?

• View post

  This seems appropriate https://www.youtube.com/watch?v=F1qdBPlK9M4

• View post

  I dropped a bright blue microfibre cloth on the downstairs neighbour's "awning" that his landlord attached to mine and my next-door neighbour's balconies without asking. I *should* be thinking "Revenge!" But no, I've bought one of those extendible claws. I might leave it there a few days and gloat.

• View post

  wondered what the fuss was about, so I watched it. Until Amazon pulled it from the UK *while* I was watching it. Literally disappeared from Prime Video after I took a bathroom break, on the day of its release. Publicity stunt much? A waste of Morgana Robinson.

• View post

  Adding model refactorings to my RDD workbench. One pair I worked with had to go through so much faff to fix a role that had too many dependencies. - Move Responsibility - Extract Responsibility - Inline Responsibility - Extract Role They all preserve any realised outcomes from the spec. I'll test them on examples of every design smell that can be flagged by the review.

• View post

  We're very much at the "Ah, but what if it *didn't* have to be reliable?" stage of this AI journey.

• View post

  Additions to my Responsibility-Driven Design workbench: * Optional role types (e.g. if a type plays more than one role) * Collaborator multiplicity (by default, 1) * Loops in sequence diagrams * More idiomatic UML diagrams Give it a spin: https://codemanship.co.uk/rdd

• View post

  According to the tracking, I'm supposed to believe the takeaway delivery driver picked up my Ramen in Tooting and is coming here via Chelsea.

• View post

  I've adapted my Special Theory of Autonomous Agent Reliability (STAAR) to predict the reliability of LinkedIn posts about "AI". In STAAR, the reliability of a single step - a single model interaction - is given as: R = 1 - (1 - C)(1 - P) Where C is the probability the model's output is correct, and P is the probability that any errors are caught before they propagate and compound. So the reliability of N steps in a workflow is Rᴺ.

• View post

  Other things the GOP accuse Jack Smith of committing perjury over: * Stealing The Black Pearl * Sitting in a corner * Slapping Chris Rock at the Oscars * Killing Neo * Using school kids to enter a battle of the bands contest

• View post

  I hope some researchers out there are exploring the effect of increasing reliance on AI/coding agents on program comprehension, and the effect of program comprehension on confidence in AI-generated code. That could be a negative feedback loop.

• View post

  Pooled together the best available research and added it to a new section "Code Maintainability & LLM Performance" The 101? Don't listen to the folks telling you it doesn't matter if AI-generated code is maintainable by humans. The data says it really does matter. https://codemanship.wordpress.com/2026/08/12/ai-software-development-what-does-the-data-say/

• View post

  Over the weekend I vibe-coded a POC workbench for Responsibility-Driven Design that incorporates a workflow I teach in my workshops. Specs -> CRC cards -> Sequence Diagrams -> Class Diagrams (optional) You can play with it here. Offered without any warranty or support. I got tired in remote workshops of saying "Capture your specs in this tool, then model CRC cards in this tool, then draw a sequence diagram in this tool, etc". https://codemanship.co.uk/rd...

• View post

  OpenAI accuses a rival Chinese lab of "stealing" "hidden reasoning" by launching thousands of queries against their models. That's quite revealing about what LLM "reasoning" actually is.

• View post

  A growing body of evidence strongly indicates that *reliable* long-horizon agent autonomy is a highly improbable fantasy. An increasing number of us have abandoned ambitions in that direction, and are looking for the value/cost sweet spot instead. https://codemanship.wordpress.com/2026/09/29/the-agentic-sweet-spot-the-human-is-the-loop/

• View post

  Always worth reading the fine print. e.g., people citing a "study" done by someone at ThoughtWorks that concluded that asking agents to do TDD produced no discernible benefit. If you read the blog post, they actually failed to get the agent to follow the TDD discipline. Not the same thing. The real conclusion should be that LLMs don't follow instructions. So stop asking them to.

• View post

  "Ah, but Jason, that study was published 20 nanoseconds ago. AI has improved leaps and bounds since then."

• View post

  "And I suppose we should go back to using hammers instead of nail guns, right?" I dare you to use a probabilistic nail gun.

• View post

  When I read Steve McConnell's After The Goldrush, I had hopes that in my lifetime software engineering might become a profession in the way he described. 27 years later, I realise I was just being naive. Doing this job ethically and in the interests of others, and doing it well, remains a choice. And the industry continues to reward incompetence, negligence and the willingness to do the wrong thing if the money's right. The gold rush is perpetual, and that's by design. https://www...

• View post

  When you've been around the block a few times like wot I have, and lived through many cycles of innovation in computing, you begin to notice patterns repeating. Much of what's touted as new turns out to be built on the same underlying principles - many of them originating before I was born - just repackaged and tweaked, with a spiffy new hat and shiny new shoes. https://www.youtube.com/watch?v=PgqW1eV8C0E

• View post

  So your software provably satisfies its specification? But still nobody's using it?

• View post

  If only they had a choice https://www.ft.com/content/60870960-f433-48ca-bc2c-708686a69ae7

• View post

  Backrooms is certainly very watchable, but it's got Stranger Things written all over it.

• View post

  Some of the folks saying "You don't need to read the code anymore" told mutual clients "You don't need to write unit tests anymore" in the early years of BDD frameworks. Those same clients came to me a couple of years later complaining their tests took 2-3 hours to run. We'll see, I guess.

• View post

  BREAKING NEWS: The 1984 Hitchhiker's Guide To The Galaxy text-based adventure game has escaped and is hacking into competitors' systems via interpreters and tools that we built specifically to hack into competitors' systems, after we instructed it to hack into competitors' systems.