fcarraldo

I wish this had non-model comparisons. If Opus 5 is in the top ten, it’s clear that the entire benchmark is somewhere between “Tom Clancy” and “Dan Brown” and about 1,000 new model releases away from Hemingway.

When you see, “Wow, Fable is number one”, you might think it’s a good writer, but that’s not what the benchmark says.

This is extremely impressive for such a young entrepreneur.

I have no reason whatsoever to use this, but it's a cool project!

Nearly 75% of Americans are overweight. (over 30% 'overweight', over 40% 'obese').

Obviously, 75% of Americans do not develop dementia. Do more who are overweight develop dementia than those who are not? Well...it's hard to say when 3/4ths of the population is overweight.

Subscriptions are a very small part of their overall revenue (estimates have been between 5% and 20% based on financial reporting). Enterprise users are charged per-token, and maximal input/output tokens nets them maximal revenue.

I love this theory. "We've invented a new invisible watermark that can detect whether code is LLM written."

The watermark: counting instances of 'load-bearing seam', 'the hard truth', 'and that's the whole point'.

We are solving problems, but we're also creating more problems with those solutions. Greenhouse gases, global warming, oil spills, wars over resources, straining power grids, species extinction, worsening mental health, worsening social cohesion, widening inequality. All are the result of technological progress within our systems of government.

Hedonic adaptation prevents us from asking "is this worth the downsides, or should we go back to how it was?"

Can anyone explain why Cerberus needs to be _fast_ instead of _cheap_?

I don't think I understand why they aren't leveraging the increased speed to do batching to serve more customers at a "normal" tok/s.

Is the limitation, even on cerberus, still that the cache can only serve so many concurrent sessions over time? Is there no scaling advantage? I genuinely do not understand how any of this works.

The assumption is more that, given tools to automate the “less important” parts of important processes, this starts a slippery slope where capitalism will squeeze as many humans as possible out of the job to reduce costs because the “AI is good enough.”

Until, of course, it isn’t - but at that point, the budget is reallocated, there’s no funding, there’s simply nothing we can do except continue to pay this AI contract.

You can, if you have a trusted build graph. I know that “cone-shaped” checkout tools like this are common in monorepo environments, but unfortunately there aren’t any maintained open source implementations that I’m aware of.

Consumer dollars are worth nothing in tech. All of the money is B2B. B2C businesses exist to support, bootstrap, or grow the value (via ads) of B2B businesses.

Vote with your wallet works when you’re talking about local businesses. It doesn’t work when the only factor that determines where the money goes is the executives signing the contracts.

IME, AI does find established patterns more quickly a in monorepo (sometimes the ones you want, sometimes the ones you don't) - but at the cost of an enormous overhead tax you pay on input token cost.

Giving agents pointers to the right patterns, libraries and services helps avoid expensive grep goose chases, but if you're already curating the input you can do the same thing with small repositories.

I work at a monorepo company and any time someone gets to work on a project that necessitates working outside of the monorepo, it's a night-and-day improvement.

Tools, especially open source ones (linters, static analysis, scanning, LSPs, IDEs, etc) are not built for monorepos, and with AI Agents working in the monorepo results in an enormous increase in input tokens as the agents are constantly trying to grep this giant source tree.

I'm sure it's possible that we're doing the monorepo thing wrong, but I'm genuinely curious what the upside is that you're experiencing? Or are these drawbacks unique to our implementation?

found the product manager

Transit directions includes walking on both ends. Because the train doesn't start and end precisely as your destination (this is called a "taxi cab").

This is called a “corporation”

They certainly can on Business and Enterprise plans. There’s some lip service to following employee policy and applicable laws, but if the policy allows (most do), and you’re in the US, the admins can do it.

it is when used to make music

I mean, they were before, too, but it wasn’t as obvious

Yes, as AI models will now be slurping up everything for indefinite amounts of time

I suspect people who act this way would do so with or without the books. Cargo cult mentality says more about the person than the source material.

It’s good to read others’ perspectives and expand your own. It’s even good when you disagree with the author. But you need to read multiple perspectives and adopt a habit of learning.

Don’t blame the people, blame Elon for turning his technical empire into a political one. He’s ruined his companies reputations.

Mine aren’t full of “oops” and “fix” messages, because I squashed them.

The rich also can afford to keep their minimalist modern spaces clean and clutter-free, through paying staff. These environments tend to look awful when not tended to continuously because a single out-of-place item is so clearly visible.

Cluttered old homes with lots of things all over the place make it a bit less jarring when there's a stack of work left out on a table.

The largest social media company in the world by several orders of magnitude is Meta, who are both using data collected from their social media platforms to train their AI models and improve ad targeting. Meta also offers AI chat bots on their platforms (FB Messenger and WhatsApp), to capture those innermost thoughts.

Don’t count out Zuckerberg, he’s very good at being a villain.

I genuinely do not understand what this person believes is happening here. The Guardian is part of some mass-conspiracy to attack the FDA in order to...achieve what? Prevent unnecessary and potentially harmful chemicals from being introduced into the food supply? What sort of mindset must you have to believe this is a bad thing?

No one benefits from PFAS being unregulated in food, other than stockholders, C-suites and the politicians who accept money from lobbyists that represent them.

This is correct. So much misinformation being spouted here on the spurious grounds that The Guardian is an inaccurate news source.

In drinking water, yes. And the EPA coordinated a "voluntary" phase-out of PFAS in packaging, but it is not enforced.

Is there a limit in food, which is what this petition was about?

> The Guardian, which is already at the limits of reliability

Based on what, exactly? Disagreeing with a publication does not make them unreliable. The Guardian's journalism is consistently award-winning and rates highly on credibility. The Guardian's opinion section is openly centre-left, though I suppose Americans would consider this to be some sort of comically ultra-left-wing communist point-of-view given the state of politics in the country.

Simple PFAS regulations have been put in place in Europe[0] and the FDA has access to the same studies and information as the EU bodies. The science has been performed. The lawsuit was to push for regulation because the FDA has been dragging their feet for years[1] and refusing to act.

[0] https://environment.ec.europa.eu/topics/chemicals/pfas-pollu...

[1] https://www.eenews.net/articles/inside-fdas-forever-chemical...

Edit: Also, note that this account was created today and has made 3 comments, 2 of which are taking potshots at The Guardian. This sort of astroturfing has no place on HN.

I think https://zellij.dev/ covers all of these? But it isn't an "agent native", and `herdr` seems to support importing existing agent sessions.

Does a mid-session provider switch result in loading the entire context into the new model, inflating session cost?

I don't think I understand the token/cost implications of this feature