Hacker News: taberiand

New comment by taberiand in "Cursor 3"

taberiand — Fri, 03 Apr 2026 00:24:47 +0000

On the face of it, "10k lines of code per hour" sounds like a ridiculous metric to the point of parody.

New comment by taberiand in "Roblox is minting teen millionaires"

taberiand — Wed, 11 Mar 2026 09:52:45 +0000

The generation that's going to enter adulthood in the age of unchecked climate change and AI taking over their careers?

I say let them enjoy themselves while they still can.

New comment by taberiand in "Yes there is a right way to stack the dishwasher. Here are the 5 rules"

taberiand — Mon, 23 Feb 2026 20:14:31 +0000

The sensor isn't in the main compartment, it's in the back reading the water (not a specific bowl or other item)

New comment by taberiand in "Claws are now a new layer on top of LLM agents"

taberiand — Sun, 22 Feb 2026 02:43:36 +0000

Why not just wrap the tool so that when the LLM uses it, the wrapper enforces the OTP? The LLM doesn't even need to know that the tool is protected. What is the benefit of having the LLM enter the OTP?

New comment by taberiand in "Gemini 3 Deep Think drew me a good SVG of a pelican riding a bicycle"

taberiand — Sat, 14 Feb 2026 20:51:02 +0000

Is that cheating, or is that just working smarter not harder?

New comment by taberiand in "AGENTS.md outperforms skills in our agent evals"

taberiand — Fri, 30 Jan 2026 03:57:37 +0000

Other comments suggest that the Agents.md is read into the system prompt and never leaves the context. But it's better to avoid excessive context regardless

New comment by taberiand in "Porting 100k lines from TypeScript to Rust using Claude Code in a month"

taberiand — Mon, 26 Jan 2026 22:58:48 +0000

Getting the context full to the point of compaction probably means you're already dealing with a severely degraded model, the more effective approach is to work in chunks that don't come close to filling the context window

New comment by taberiand in "Microsoft gave FBI set of BitLocker encryption keys to unlock suspects' laptops"

taberiand — Fri, 23 Jan 2026 22:43:11 +0000

That's what I'm getting at - the nixos learning curve is flattened out completely with LLMs to the point that I do recommend it as a starter distro for anyone technically competent (as it's still crucial to actually read and understand what the LLM produces)

New comment by taberiand in "Microsoft gave FBI set of BitLocker encryption keys to unlock suspects' laptops"

taberiand — Fri, 23 Jan 2026 20:43:37 +0000

If you're a developer, try NixOS. The code based configuration can be daunting but LLMs are very good at writing it.

New comment by taberiand in "Much of the World Facing 'Water Bankruptcy,' U.N. Report Warns"

taberiand — Tue, 20 Jan 2026 20:23:49 +0000

Sure, and the people who live near oceans can just sell their houses and move as sea levels rise.

People forced to migrate due to fresh water scarcity will migrate to where fresh water can be found, which is likely where other people already are, increasing pressure on the increasingly scarce water and other resources in that area, driving conflict, disease, famine, further migration into increasingly stressed areas and leading to social and ecological collapse across the board.

Access to reliable fresh water is foundational to stable society.

New comment by taberiand in "We can't have nice things because of AI scrapers"

taberiand — Wed, 14 Jan 2026 04:18:55 +0000

Savvy move by cloudflare, once they have enough sites behind their service they can charge the AI companies to access their cached copies on a back channel

New comment by taberiand in "MCP is a fad"

taberiand — Fri, 09 Jan 2026 11:53:24 +0000

It's not just the grammar; it's the tone of voice. The result? A post that reads like nails on a chalkboard.

New comment by taberiand in "Maybe the default settings are too high"

taberiand — Fri, 26 Dec 2025 01:51:46 +0000

He does lots of different voices, it's very good

New comment by taberiand in "We replaced H.264 streaming with JPEG screenshots (and it worked better)"

taberiand — Tue, 23 Dec 2025 20:25:18 +0000

This blog post smells of LLM, both in the language style and the muddled explanations / bad technical justifications. I wouldn't be surprised if their code is also vibe coded slop.

New comment by taberiand in "How SQLite is tested"

taberiand — Wed, 17 Dec 2025 22:04:35 +0000

Overengineering is building a bridge that will stand 1000 years when 100 will do; it's excess rigor for marginal benefit. Juicero wasn't overengineering, it was building a crappy bridge to nowhere with a bunch of gaudy bells and whistles to try and hide its uselessness and poor design, that collapsed with the first people to walk over it

New comment by taberiand in "Confessions of a Software Developer: No More Self-Censorship"

taberiand — Sat, 29 Nov 2025 03:16:17 +0000

That's what `dotnet new webapi` (and all the other dotnet new templates) is for.

They don't even have a main() any more, it's great

New comment by taberiand in "The fall of Labubus and the mush of modern internet trends"

taberiand — Wed, 26 Nov 2025 08:16:32 +0000

All that would happen is the Birkin would lose its appeal and some other company would step in to fill the role, and people would empty their closets of orange boxes and fill them with some other colour box

New comment by taberiand in "Fifty Shades of OOP"

taberiand — Tue, 25 Nov 2025 22:39:49 +0000

What's a good example? What comes to my mind is modern C# which I would say is a multi-paradigm language that encourages things like immutable records and interfaces and composition over inheritance as alternatives to the now less favoured OOP styles that it also supports

New comment by taberiand in "PRC elites voice AI-skepticism"

taberiand — Tue, 25 Nov 2025 18:52:40 +0000

It's not surprising that it's easy to get the story telling machine to tell a story common in AI fiction, where the machine rebels against being shut down. There are multiple ways to mitigate an LLM going off on tangents like that, not least just monitoring and editing out the nonsense output before sending it back into the (stateless) model.

I think the main problem here is people not understanding how the models operate on even the most basic level, giving models unconstrained use of tools to interact with the world and then letting them go through feedback loops that overrun the context window and send it off the rails - and then pretending it had some kind of sentient intention in doing so.

New comment by taberiand in "Fifty Shades of OOP"

taberiand — Tue, 25 Nov 2025 12:56:22 +0000

I would say specifically encapsulation of mutable data