danmaz74

Ah, so both Claude and Codex are going to have the same exact reduction for their two equivalent plans - 20x to 10x, at the same previous price - at the same time? I already knew about the ChatGPT announcement, but the claude one is completely new to me and completely unexpected. So much for competition...

An expression that stuck with me from the Domain Driven Design book is "supple design". Achieving supple design (and thus supple code) is a goal, but it doesn't come for free - it requires a lot of work. I'm trying hard to get agents to create supple design - or at least not to make my own designs less supple - but it's not easy, because they have a tendency at overcomplicating and overengineering code.

The trade-off between quality and speed of development has always been a tenet of software engineering. Agentic coding changes the equation a lot, but the equation is still there.

By that analogy, Anthropic is doing more like 0.000001% of the work of learning all we learned, and writing it down so that Anthropic at al can condense it into their models.

Higher TPS by itself wouldn't remove the humans in the loop. Humans in the loop aren't there to make things faster, but to make them right - current LLMs can't do that on their own in most real world projects.

On the contrary, I would say that higher TPS would make humans in the loop much more productive, because the current long wait times mean that you lose focus and flow.

A US-led NATO may very well not survive two more years of Trump presidency.

T-bills are offering higher yields, that helps with demand.

This is pretty a common misconception. Schengen area only refers to the fact that there are no border checks between two countries which are in that area - it's just like an internal movement. This applies to anybody, so for example if a Canadian enters the Schengen area, they can move to any other country within that area without checks and without requiring an additional visa.

What you're thinking about is the "single market" - which, even if the name could deceive, also applies to people, including the right to go work anywhere within the single market for citizens of single market countries. Which is what many brits always hated, as they wanted single market for services and goods, but not for people.

Single market doesn't imply Schengen though - eg I as an EU citizen have the right to move to Ireland for work, but I still need to go through border checks as Ireland is outside of Schengen (because they have free movement with NI).

Schengen looks like a stretch, considering the geographic situation, but free movement is achievable, if both want it.

As someone who is very much in favor of this, let's remember that for now we're only at the announcements level. Any binding changes in the EU require a long process - if anything else, because there are 27 member states, each of which have veto power. But I'm hopeful that facts will follow these announcements.

At this stage it's not about how much it's worth, but how much investment is required to reach their goals.

Yes, that could be a way to achieve that. After all, even humans have short term and long term memory, and it looks like sleep is a very important "tick" to connect the two, so it's not all just continuous.

For sure, doing it for all users would be economically unfeasible. I wonder if the labs are experimenting with something similar, though.

I have no idea about timelines, but the current LLM architecture has no mechanism which could emulate online learning in a way similar to how it happens in humans and other animals. As another commenter wrote, there is no neuroplasticity while these systems interact with their environment in normal usage; adding information/constraints to the context partially mitigates this, but it's a completely different mechanism.

The RL phase is the most similar mechanism I know of that comes to my mind, but I'm not sure it could be adapted to fill that gap.

My totally unsubstantiated theory is that this is the missing link towards what most humans would consider AGI. I don't see this as intractable, but it may require some substantial change in architecture.

Given how important matrix multiplication with a huge number of fixed parameters is becoming, there is an enormous incentive to design much more efficient architectures where this very simple compute is colocated with memory. Inference cost would come down a lot.

Yes, but that should be on a case by case basis. Blocking everybody is just an ideologically stupid choice, besides being inhumane.

Memory fabs take years to build, they cost a lot and there is no guarantee that the demand will stay as high years in the future. It's all very complex.

How long has this war been going on? A year is a totally realistic time frame if you evaluate his words.

30-40 per day means between 10,000 and 15,000 Palestinians killed per year. Just think about that number for a minute. I think that you're giving Ben Gvir way too much benefit of the doubt here.

And he is a minister in Israel's government, not some rando. Israeli governments and actions in the last many years are simply indefensible.

I've been using it to vibe code a personal application while I'm using a more "powerful" setup for my main agent assisted coding, and I have to say that it's been very good so far. I just wished that it was easier to hand off some tasks to codex web, like I tried to do today as I was starting a boat trip.

Come on, it's not just MBAs. I remember when patio11 (nothing against him personally, it's just the best example I can remember) became a star here on HN by saying that the solution to every problem as a consultant was to hike your rates. The idea of just asking as much money as you can get away with, which was seen as reprehensible by most of the population (my guesstimate, I don't have hard data) until more or less the eighties, Reagan and Thatcher era, has now permeated Western society at most levels, not just MBAs.

PS Chinese companies were (and probably still are, when they can) stealing IP, just like the USA did when it had more IP to steal from Europe than to protect from others. Western companies were very happy to let that happen when in exchange they could make more profits in the short term by using cheap Chinese labor and infrastructure.

> hope their bet on JPY vs EUR pays off in the long-term

Which bet?

The plug in "balcony panels" are usually limited to 800W here in Europe, so having them without a battery can make sense because you're probably going to use that power on the spot most of the time.

Russia has been doing the same for years - especially trying to destroy all electrical and heating infrastructure during Winter. Ukraine has the right to retaliate, now that they're able to.

For many of us, it was our first computer. It was very simple to understand at every level and very hackable. I was very young when I used it, but I learned so much about the internals of computers.

You won't get 12% return if your panels generate electricity which is only paid between 18 and 19, because there is already overcapacity between 16:30 and 18.

I've read several comments in the last few months citing this kind of approach. What I'm trying to do is to make the implementation phase more reliable.

By the way, if you're not doing that yet, something that can really help when doing UI/UX work is to have the agent create some mockups, and then tests based on those - I'm using Cucumber with some extra sauce for this. It's a very nice way to guide the agent in a falsifiable way.

Wait until you hear about the giga-factories...

But seriously, if you care, this is just like using an existing library to do the lower level development work which I think is already pretty well done by the existing agents. It's not a design decision.

Interesting. From time to time I'm thinking about completely replacing the agent layer with my own, so that I would be able to enforce some rules at a lower level, but I'm wondering if it's worth it. What's been your experience with that?

Yes, like many others I've been experimenting a lot. What I've got so far is a harness-of-harnesses - ie, a harness which sits on top of Claude Code, Codex or OpenCode. I still use Claude Code or Codex directly for the initial planning of features, to investigate issues, and for small fixes, but whenever there's something even just a bit complex to do, I use my second-level harness.

Summarizing it a lot, what it does is:

* help you make better plans

* split plans into iterations, in a module-aware way for projects which have strict modularity (for now I'm doing this specifically with TypeScript and dependency cruiser) - this helps a lot when a project becomes complex

* ask an agent to implement an iteration, and then programmatically run a lot of checks after each iteration - not just regression tests, but also checks against project principles and conventions

* when possible, automatically fix deviations; when not possible, raise them to myself for an end-of-plan review

In this way, instead of having to constantly be engaged with the chat interface, with all the shorter or longer wait times which break my flow, I spend a lot of highly focused time during initial planning and final review. A plan implementation can go on for hours, and the various anchoring mechanisms added to the tool keep drift to a minimum.

At some point I'm planning to release this tool as open source. As this is the result of months of trial and errors, dogfooding, and vibecoding on the tool itself, the codebase is chaotic and the UI is still full of experiments I mostly basically abandoned, and I'm not used to releasing stuff in this status. But perhaps, in this brave new world, I should just do it and see what happens?