Hacker Newsnew | past | comments | ask | show | jobs | submit | dayvid's commentslogin

I upgraded from a 13 to a 17 pro. Barely notice a different, my old phone was just dying

I got a battery replacement instead, and still use 13 pro, works like new.

The 13 is the last generation of the mini and I still have mine. Good to know I'm not missing anything. Battery capacity isn't even below 80% so I can't replace it under Applecare yet.

13 to 17 pro is a huge leap in performance, battery life, and camera features.

It also documents itself extensively

But then you’ll use it to change something, and it won’t update any of the extensive documentation. So then it’s filled with lies.

Or worse

It'll mention the old version then the change, and the relating tickets.


That actually hasn’t been my experience. For all that I dislike about LLM code, it maintains comments well.

You need a better harness. LLMs are significantly better at this than humans. Add it as a step in your LLM process and your documentation will always be up to date.

Sounds like human generated code and docs. TBH

I fucking hate it.

It encodes context it needs into doc comments and regular comments that make them in insufferable to read and so verbose because the comments contain info on other parts of the code base (that might have changed since that comment was written) but were useful to the model for its implementation. I have so far found no way to stop claude from doing this. It will sometimes do it with hand holding but the moment a task takes a slightly large amount of context its back to the word barf.


This is certainly a problem with out-of-the-box usage. But easily overcome. Spend time updating your harness. This is not a hard problem to solve and there are many helpful guides out there.

We basically took our documented (human) SDLC process and applied it to the relevant harness hooks. Since our SDLC docs talked about what good comments looked like and what to avoid, we basically had the direction for the tools. Opus 5 did throw us a loop and increase the verbosity and decrease usefulness of the text, but refining our documentation cleared that hurdle within about a day.


Large businesses who choose enterprise solutions over cheaper indie/open-source solutions for similar reasons (dedicated support, sticky products designed for their industry, less regulatory concerns, etc.)

Do they not measure returns on their investments? It's not like Anthropic is the only enterprise out there, any other offering is significantly cheaper. Like I said, I do not believe there are agentic pipelines out there built on their flagship models, if there are then they are eating the costs and not profitable or they have found some serious market that makes no sense to me.

There's no mercy for people learning the ropes now. The lower latency of LLMs make people impatient to deal with growing pains.


Just pay them well enough. Or pay the people at the top a good amount of money to keep the people below in check and have some hope you can rise to the top for a small portion of the people.


If civilization collapses, money is useless


Pay them with chocolate rations or higher status bunk beds.


Many businesses have information with other parties they're contractually obligated to not share. That's on top of insider information or plans that would hurt shareholders or the business if it was out in the public


You can have really stupid parents who ingrain suboptimal behavior patterns and interests to you. Things like poor eating, attitudes and beliefs about work and what’s possible, abuse, poor ways of coping with stress and conflict. Throw into that a hopeless environment with no immediate opportunities, bad peer group, and early pregnancies and you have a recipe for disaster


I've been remote for ~6 years. I can't stand being in an office when I don't need to be there and looking busy. On top of that, commute savings are great. I work out of a coworking space a walking distance from where I live. I also had a child recently, and being able to pick him up easily after work is priceless. Forcing most tech talent to be in the same 2-3 cities is one of the worst things in recent history.


It puts a premium on your concurrency skills. I'm constantly setting up agents and moving from agent to agent; answering questions, evaluating work, providing feedback, finishing up items. Builds a different muscle and there's still more ways to improve workflow


It's not even that much money for the reputational damage. Wild.


Reputational damage has not been relevant in US politics for a decade. It simply is not a thing anymore


What reputation?


Yes because Truth Social and its owner/head troll both had a stellar reputation to begin with.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: