Alexander Petros explaining how single page applications had a detrimental effect on maintainability. Emphasis mine.

So, the title of this talk is based on something that a friend and mentor of mine said to me on the phone a week ago. He said, “When I started working in the civic technology space, modernization meant replacing systems that were built before I was born. So, you’re thinking your COBOL, your mainframes, your—you know—big computers in rooms, and turning those into, you know, Java, VBs. But today, modernisation has come to mean replacing systems that were built 10 years ago.“

And there’s a very specific reason for that, and I’m going to address it and come up with some ways that we can fix that problem. So, the main reason is that 15 years ago, roughly, we started building websites as single-page applications by default, and this has had enormously detrimental consequences in our ability to maintain those websites over a long period of time and update them. And I have a very specific example here, which is the ability to apply for food stamps for the state of California. So, for SNAP benefits, until very recently, California had two web portals for doing this: one was BenefitsCal, which was the one-stop portal for a whole bunch of benefits, which included food stamps, and that was a single-page application. And then there was getcalfresh.org, which was a traditional HTML service built on Ruby on Rails by Code for America, and that was specific to food stamps. And I’m going to just show what it looks like to load these two applications on a relatively low-performing device. This is a P75 device, which means that it’s at the 25th percentile on capabilities for network and processing power.

What happens next will blow your mind.

Paul Hockenos talking about how having eco-friendly energy generation capability helps you when your country is at war.

US environmentalist Bill McKibben argues that Ukraine’s autonomous power sources are ‘comparatively invulnerable to attack’. In our new era of drone warfare, writes McKibben, centralized and complex energy facilities are now ‘sitting ducks for drone attack’. Smaller, ‘scattered’ generation sources with batteries and independent transmission grids, like the Mykolaiv fields, are not only harder for Russia to strafe, but they’re much more quickly repaired, too. This is in contrast to petrochemical refineries, which McKibben calls ‘one of the most complicated machines humanity has ever constructed’, and often cover hundreds of acres. These are ‘filled with highly complex equipment and highly flammable hydrocarbons; hit one corner with a drone and the flames and the damage are likely to spread quickly’.

‘A coal power station [is] a large single target that a single missile could take out,’ says Jeff Oatham of DTEK, Ukraine’s largest energy investor. ‘You would need around 40 missiles to do the equivalent amount of capacity damage at a wind farm.’ And, like wind parks, a solar field functions even when part of it is out of operation.

Ukraine is revamping its energy system at warp speed for the purpose of energy security, not climate protection. ‘Ukraine’s energy transition is not a slogan,’ says Levgeniia Kopytsia, a Ukrainian energy analyst at the think tank Institute for Climate Protection, Energy and Mobility. ‘It’s a security-driven transformation, unfolding under extreme constraints. It priorities decentralization, flexibility, and speed of recovery.’ In wartime Ukraine, renewables and local grids are coldly calculated security assets, Kopytsia says, that can boost the system’s resilience – and in Ukraine this translates into saving lives.

AWS and GCP introduced spend limits now. In 2026. Why so late? From Hacker News.

joshdavham: It’s incredible that in 2026, AWS and GCP are only just now introducing this. It’s possibly one of the most obviously needed features for a cloud provider.

Also, does anyone know why it’s taken this long? I suspect it’s a technical reason. While one could be cynical, I doubt it’s an intentional business/product decision. Hard spending caps are both excellent product differentiators and could possibly save these providers money as they don’t have to forgive their users when they accidentally over-use a service.

Anon1096: It’s one part technical, one part a product decision. The technical part is that billing is not actually instant. As a most basic example, a VM reports its billing units every X period of time it is active. If there is some network blip but it’s still running, then that billing data could be delayed.

The product level decision is that “shut down everything” is something the customers you want to target don’t actually want. Are we including deleting RDS data? S3? Glacier storage? If so then the headline will just change from “Hobbyist got charged XXXXXX on AWS” to “Business literally had all their data deleted because a hacker took over their VM and mined bitcoin”. The only people who really want this are hobbyists and it’s not a market segment that’s worth chasing. Easier to do the status quo of forgive afterwards then even open the can of worms of deleting all of a business’s data and all their backups just because they had a 100k overrun.

Do check out the entire thread for counter points.

This post on Thoughts of Widdershins highlighting how people are moving out from social media to go small and more focused.

What strikes me is that hardly anyone is leaving for something big. They’re going smaller. A personal site, a self-hosted app, a capsule nobody’s going to stumble onto by accident. Less reach, more control. A few years ago that would have sounded like giving up. Now it sounds like the sensible option.

I don’t think the big platforms are going anywhere. But the people who made them interesting in the first place keep quietly walking out the side door, and writing a post about it on their way out.

Simon Willison arguing that hard budget caps need to be turned on by default.

Here’s a product feature which the world is going to need a whole lot more of over the coming months and years: default hard budget caps. I’m talking about the feature of pay-by-usage services and APIs that lets you say “after $X/month, cut this thing off and return errors”. These need to be hard limits. Soft caps, “after $X/month, send me a warning email”, will not cut it.

Coding agents, and personal agents (coding agents wrapped in a less threatening UI), greatly reduce the friction of spinning up code that can do useful things. Sometimes those things cost money—calls to paid APIs, or hosted web applications, or systems that can bill for additional storage and compute.

Nobody wants to wake up to an email sent at midnight warning about a budget limit and find that, while they slept, their rogue service had consumed several hundred (or several thousand) more dollars of usage.

Martin Fowler arguing that AI needs to have conscience rather than consciousness.

Which all makes me think that when it comes to AI and LLMs: why are we wondering if they have consciousness – when we should be wondering why they don’t have a conscience?

After all, the AI labs trained them to be super-persistent, why did they not train them to be well-behaved? It’s as if a human trains a dog to bite children, and we blame the dog rather than the trainer.

A decade ago, a dog ran in front of me while I was riding my bike, putting me into the hospital with a broken arm and a broken face. Massachusetts law makes owners strictly liable for what their dogs do. That meant I didn’t have to prove the owners were negligent in how they controlled their dog or fenced their property – they had to pay my hospital expenses (in practice their insurance company paid my insurance company). There should be something along these lines for LLMs. Those that train the LLM should be responsible for what it does, after all if it has such a galaxy brain it should be able to tell if it’s doing something wrong and either stop or get a human’s explicit approval.

Sunil Pai’s suggestion for senior engineers:

I think in moments like this, the thing you need to do is actually a little counterintuitive. but you have to start from a position of assuming that everyone around you is operating in good faith, okay? like, they hired you for you. they didn’t hire you for who you think you’re going to be in six months. they like you for who you are right now, bro.

so the thing to do is actually to drop a level. not to try to be a higher-level engineer, but to almost drop a level and simply become the greatest teammate for a while. you’re still responsible for your work, you’re just trying to get moving again. you know, “I’m just here to help and support my teammates. I’m going to do bugs, the annoying things on somebody else’s plate or that no one has been able to get to. do grunt work, organizational stuff, write-ups. just things to help people out.”

and the reason you want to do this is you want to move away from an outcome-based mindset to momentum-based. like, you want to build a routine, because momentum is everything in this, and relationships are an even bigger thing in this. and by getting momentum going and fixing your relationships with other people, the thing you get is stamina. and that’s when you realize the real lesson, which is that big projects are not done with big efforts. it’s a big marathon, just slow, steady, incremental work. forget about a weekly basis, like a daily basis, an hourly basis, to the point that it’s muscle memory.

you’re in automatic mode. you wake up, you make yourself coffee, you get to work, and you just do the work. and then 30 days later, you look back and you see the immense volume of work that you’ve done just by doing a little bit every day. because the thing you’re going for is reliability, fixing your relationships. you want to take the burden off your teammates and your manager, because then, once you build that trust, they will give you bigger work. that’s the move here.

remember, you are in the reputation-building business, and software is actually downstream of that. godspeed and best of luck.

This will never stop blowing my mind.A 12-year sequence of real telescope images of a star 133 lightyears away.And around it, four planets slowly orbiting.Four real exoplanets, each more massive than Jupiter, imaged as they move around their host star.

— Paul Byrne (@theplanetaryguy.com) 2026-10-02T02:14:05.749Z

Mat​hia⁠s S​chäf⁠er highlights the impact of AI on web development. She cites various examples where earnings have fallen off a cliff in a short span of time for anyone to react and adapt. Below is the experience of Axel Rauschmayer.

The income from my book sales went from being enough for me to live off (2024) to zero (2026).

The traffic to my blog and my books (which were free to read online) increased beyond what I can currently afford. Virtually all of it comes from AI crawlers, so there is no ad income.

Therefore, I’m taking my blog and my books offline so that I can decide what to do next – especially w.r.t. AI companies stealing my work.

Jeff Atwood sharing the email he received which highlights the loss of belonging, community, connection—or whatever you want to call it—in the world on AI.

Sometimes I would get stuck, and I needed someone to just nudge me in the right direction.

I actually come from a family with a few software engineers. I have friends who were software engineers. I reached out for help. The standard response from everyone fell between ignoring me or telling me I need to “do it myself”.

You know who did help me? Random strangers on Stack Overflow. People who had nothing to gain from helping me other than just trying to help someone in need.

That meant so much to me. I felt a connection to people through a community. 

And it meant the world.

An LLM would have done exactly what I wanted back then. But would it have given me what I needed? 

At the time, I needed to know somebody cared enough to help me.

You, your colleagues, and random strangers in the Stack Overflow Community, really helped me. It was more than answering my questions. You gave me your time. That made me feel like I mattered. The fact my questions could help other people in my predicament made me feel like I was part of something.

An LLM can’t do that.