Yeah but this is the thing, that's just text. If I tell someone "you can't post on HN anymore", whether they won't is entirely up to them.
Permissions in context or text are weak, these tools - especially the ones that operate on untrusted input - need to have hard constraints, like no merge permissions.
To be clear - the text I pasted is config for the Github actions workflow, not just part of a prompt being given to a model. The authors seemingly understood that the LLM could be prompt-injected run arbitrary code so put it in a workflow with read-only access to the repo.
Maybe. The reality of software engineering is that there's a lot of mediocre developers on the market and a lot of mediocre code being written; that's part of the industry, and the jobs of engineers working with other engineers and/or LLMs is that of quality control, through e.g. static analysis, code reviews, teaching, studying, etc.
And those mediocre engineers put their work online, as do top-tier developers. In fact, I would say that the scale is likely tilted towards mediocre engineers putting more stuff online than really good ones.
So statistically speaking, when the "AI" consumes all of that as its training data and returns the most likely answer when prompted, what percentage of developers will it be better than?
That's not how modern LLMs are built. The days of dumping everything on the internet into the training data and crossing your fingers are long past.
Anthropic and OpenAI spent most of 2025 focusing almost expensively on improving the coding abilities of their models, through reinforcement learning combined with additional expert curation of training data.
Have you seen how many people are talking about the November 2025 inflection point, where the models ticked over from being good at running coding agents to being really good at it?
Doubt it”s sustainable. These big models keep improving at a fast pace and any progress like this made in a niche would likely get caught up to very quickly.
It was ZZT for me, no idea how old I was, probably 8-10 or so.
But when you take a bird's eye view, it's interesting and great to see how over the years, games where you can build your own games remain popular and a common entryway into software development.
But also how Epic went from ZZT via Unreal to Fortnite, with the latter now being another platform (or what Zucc wanted to call a metaverse) for creativity.
Other notable mentions off the top of my head where people can build or invent their own games (in-game, via an external editor or through community support) or go crazy in besides Roblox are Second Life (...I think), LittleBigPlanet, Warcraft/Starcraft (which led to the genre of MOBAs), Geometry Dash, Mario Maker, TES, Source engine games, Minecraft, etc etc.
I heard the same thing from a colleague who worked on a Dutch banking app, they were quite diligent in fixing logic bugs but said that once you fix all of those, the rest is space rays.
As an aside, Apple and Google's phone home crash reports is a really good system and it's one factor that makes mobile app development fun / interesting.
Self-help books are based on first telling you you have a problem, then selling the solution. I get that some are actually correct, but the industry as a whole can only be sustained by inventing new problems and/or making sure newer generations learn about existing ones.
Ferris is a self-help book author, and while I kinda get where he's driving at, it also feels like he's just doing the same thing again, but meta - overconsumption of self-help book is like a dog chasing its tail (or a snake giving himself a BJ?), here's a solution. I'm somewhat surprised it's just an affiliate link blogpost instead of a whole book.
Tim Ferriss has always been good at identifying the next trends in the self-help space and positioning himself as an expert for trending themes. I think he might be acknowledging that self-help markets are saturated with all of the life extension influencers trying to one up each other with their protocols and supplement stacks.
This might be a trial balloon to see if this topic has legs for another book or product.
Makes sense, the iphone has only one port after all. Interesting that it supported a second one though, or maybe that's the Pro revision designed for this use case?
The second port is likely necessary for USB hubs that rely on both ports. I had one for my M1 Air. I assume it'd still work with the 2 different speeds, but I'd be curious to try it.
I'm going to get a Neo for my wife once it's available in my country.
Exactly, simplicity is a subjective term; some think of it as in Clean Code where codebases end up as oneliner functions or overly formal lasagna code with many clean-feeling layers, but they can't see the resulting complexity in the overarching architeture.
Depends on the management and whether they're involved in coding. Any engineering manager, architect, senior / lead developer etc should appreciate lower complexity.
Of course, if it's the person in charge introducing said overengineering there is a problem.
they can recognise on the informal level, but you can't put it into end of the year review document. What it will be? "Kept N PRs from introducing cruft into our systems?". Fixing or building things is much more visible, than just maintaining high standards.
Worse, to suggest a simpler approach checking existing products/APIs or even preparing toy prototype is required to be confident in own advice. This hidden work is left entirely unnoticed even by well meaning managers/engineers: they simply don't know if you knew or had to discover simpler solution.
Permissions in context or text are weak, these tools - especially the ones that operate on untrusted input - need to have hard constraints, like no merge permissions.
reply