On the GOTO Convention in Copenhagen in 2025, Kent Beck and I spent a while on stage speaking and answering questions from the viewers – a format I check with as “two previous geezers on a park bench”. We discuss our experiences with LLM-augmented programming (at that time – October 2025), we present our frustration that issues we’ve been saying for thirty years nonetheless should be stated, we are saying how something like a manifesto reunion must be led by a youthful technology, and opine on what junior builders ought to be specializing in of their profession.
❄ ❄ ❄ ❄ ❄
Ian Johnson has written a collection of posts about restructuring a gnarly codebase
The story follows an actual Laravel + React codebase over ~3 months and ~258 commits from a legacy monolith with no checks to a well-structured utility with automated high quality gates, a React SPA migration in progress, and an AI agent that reliably ships manufacturing code with minimal supervision.
The collection covers the steps in respectable element, and his method follows the sorts of steps I’d use. First get all the things beneath the management of respectable characterization checks, add static evaluation, introduce the suitable patterns to make issues stream simply.
With all of this, is his use of AI, which modified throughout the train:
For the primary two months of this venture, I used Claude Code with auto-approve turned off. Each file edit, each terminal command, each change… I reviewed it earlier than it executed. […] The outcomes had been good. The code was clear. However I used to be doing a lot of the considering and half the typing. The agent was a flowery autocomplete with higher strategies. I wasn’t getting the leverage I’d hoped for.
I learn an article about “on-the-loop” versus “in-the-loop” human-AI collaboration. The framing clicked instantly […] I used to be micromanaging as a result of I didn’t belief the agent to do the suitable factor. And I didn’t belief the agent as a result of there was nothing forcing it to do the suitable factor.
His early steps put in checks, static evaluation, and the suitable architectural patterns. With these in place, he might let the agent do extra work.
My function shifted from author to curator. I don’t write a lot of the code anymore. I Outline the patterns […] Overview the take a look at specs […] Overview the output […] Replace the harness […] Make strategic selections […]
He finishes the collection with conclusions about how he’d generalize his expertise to different circumstances.
❄ ❄ ❄ ❄ ❄
Again within the land of my start, there was some notable groans when the Nationwide Well being Service determined to shut practically all of their Open Supply repositories, supposedly to the safety menace of LLMs. Closing repos like this isn’t an efficient counter to LLM-augmented attackers. I think it’s no coincidence to see GDS (Authorities Knowledge Companies), the highly-regarded IT enablers within the UK authorities publish their place
Transferring code from public to personal as an alternative choice to funding in secure-by-design supply, possession and remediation is a warning signal as a result of it reduces sharing and scrutiny, can sluggish coordinated enchancment throughout authorities and suppliers, and doesn’t take away the underlying weaknesses in a working service.
Terence Eden memorably sums up his view on this:
Inside the UK’s Civil Service you sometimes hear the expression “being invited to a gathering with out biscuits”. It implies a slightly frosty dialogue with none of the well mannered niceties of a standard assembly.
❄ ❄ ❄ ❄ ❄
I’ve seen a number of circumstances the place these builders who’re most concerned in working with LLMs discover they’re working into an issue with cognitive endurance, Adam Tornhill has joined this group:
One of many huge wins with brokers is that they allow us to stick with the higher-level drawback for longer. We get much less sidetracked by particulars, dependency cleanup, and related secondary duties that used to interrupt focus.
However there’s a price we’re nonetheless underestimating. Agentic coding is mentally costly.
I can normally maintain the tempo for a few hours. Then I want a break. The tempo is just too intense. And based mostly on conversations with different engineers, I don’t suppose I’m alone in that.
He explains that working with The Genie means we’re making extra selections in much less time, this enhance in choice density is tough on the mind.
He responds by preserving agent duties small, automating all the things he can, and accepting that he received’t know each line of code so long as he has good verification mechanisms in place.
Notably, he has not gone within the route of doing his work with swarms of brokers that he coordinates. As a substitute has one long-running activity that he babysits and one focus activity
That final level is necessary given the running-twenty-agents-in-parallel hype. I can’t even take into consideration twenty significant issues to construct, and even much less so concerning the ensuing cognitive tax of the seemingly interruptions. It’s precisely the improper factor to even think about. At the least for people. (And sure, I perceive sub-agents and machine parallelisation. That isn’t what I’m objecting to. It’s the parallelisation of human consideration that doesn’t scale).
I appreciated that he included some ideas about what people can do in time exterior this intense programming time. Not simply “have a espresso” (though he contains that) but additionally about studying concerning the area that the software program helps.
❄ ❄ ❄ ❄ ❄
A few pithy quotes from social media
“Metaphor debt” is when all your metaphors contain the idea of “debt” as a result of you possibly can’t consider every other metaphors anymore.
❄ ❄
If a vegan crossfit fan is utilizing Claude to jot down Rust, which factor do they let you know first?
❄ ❄ ❄ ❄ ❄
Karl Bode reacts to audio system getting booed when mentioning AI throughout graduation addresses. He factors out that youthful people are more and more sad with the tech oligarchy and their fruits.
The factor is the children aren’t silly. They see the sector clearly. They see the distinction between what’s being bought to them by tech firms, the press, and graduation audio system, and what they’ve repeatedly seen with their very own eyes.
They’ve watched tech oligarchs spend the final decade mired in scandal after scandal, hype cycle after hype cycle, steadily enshittifying all the things they contact alongside the best way.
[…]
The proportion of Gen Z that suppose AI’s advantages don’t counterbalance the dangers now sits round fifty p.c, up 11 share factors in simply the final yr. Eight out of each ten consider that utilizing AI makes the method of precise studying harder.
He sees younger folks saddled with the notion of coming into a worsening world –
which leads them to rage towards this newest fruit of the tech oligarchy. A rage
that’s simple for folk like me
– with a snug retirement off-ramp – to correctly respect. A rage that might have marked political and social penalties.
❄ ❄ ❄ ❄ ❄
Related to those issues are a few gadgets in final week’s Economist newspaper. The newspaper argues that traditionally main technological advances haven’t led to important unemployment or drops in wages (paywalled article). The closest was the unique industrial revolution in nineteenth Century Britain. There was a stagnation in wages throughout this era, however there was additionally an enormous enhance in inhabitants, from 4½ million to 12 million.
It additionally factors out that we’ll most likely solely perceive the total penalties of all this when a recession hits, as that is when most unproductive jobs are typically flushed out of the system.
A second article (additionally paywalled) signifies that AI is having some impact on graduate hiring. They did an evaluation of surveys of current graduates, trying to see if employment different relying on a job’s publicity to AI. The least uncovered quintile of topics noticed employment charge fall by 1.5% during the last couple of years, whereas essentially the most uncovered quintile’s drop was 6.6%.
❄ ❄ ❄ ❄ ❄
Lawfare isn’t impressed with the newest efforts by the US Authorities to control AI.
On [last] Wednesday, the White Home invited leaders of OpenAI, Google, Anthropic, Meta, and Microsoft to the Oval Workplace for a signing ceremony the next afternoon. President Trump was to signal an government order on AI and cybersecurity—the administration’s most formal effort but to ascertain a voluntary course of for reviewing frontier fashions earlier than their launch. However roughly three hours earlier than the ceremony, when some firm executives had been already within the air to Washington, the White Home referred to as it off.
They see the proposed laws as delicate, and together with some precious measures to harden defenses towards cyber threats.
However it’s price underscoring the implications of suspending (if not outright canceling) this order, which, by its personal phrases, was about as modest a frontier-AI intervention because the federal authorities might placed on paper: voluntary, targeted on the federal government’s personal defenses, and explicitly barred from turning into a licensing regime. The objection isn’t a lot about authorities coercion as concerning the authorities having any settled function in any respect. Voluntary, in different phrases, isn’t the ground of frontier AI coverage on this administration; it’s the ceiling.
It is a questionable place on condition that the issues animating this draft order will seemingly develop within the close to future. It is usually self-defeating for many who applauded the order’s delay or demise. Removed from resolving the danger of presidency meddling in AI, killing the order simply leaves in place what Ball has described because the “opaque and basically lawless” various: authorities entry taking place via again channels, on phrases set case by case, with no secure guidelines in any respect.
One of many issues here’s a distinct lack of governmental experience, both in AI or in software program usually. An excessive amount of is being determined on the whims of the tech oligarchy, there isn’t any try to have interaction within the broader points at hand. That’s not fully a nasty factor, making an attempt to control one thing that’s nonetheless evolving so quick is normally a idiot’s errand – however the issue right here is the influence of AI is so huge that there’s actual hazard in being too far behind.
❄ ❄
Which leads me to a uncommon factor, an endorsement of a candidate for political workplace. In case you are voting in congressional district MA-06 (North Shore of Massachusetts), I’d critically have a look at Beth Anders-Beck, who’s working for congress in that district. Beth has a protracted background in software program growth (together with growing the notion of Forest and Desert), so would introduce experience that Congress desperately wants. I’ve identified Beth for many years, and have a excessive opinion of their intelligence, judgment, and skill to work with others. Congress doesn’t deserve Beth, however it does want her.
