CSMastermind

The misinformation about the Spanish conquest of the Aztecs is one of the most depressing things imaginable.

The historical record is completely clear and we have first hand accounts, primary sources, from the people present. It's all readily available and anyone interest can easily verify it.

And yet, even in the comments responding to yours, people confidently state just blatantly untrue things.

Yeah it's wild that they don't outright state the specific usage limits.

Just trust me bro accounting.

Hard to believe when all they've done so far is raise prices and reduce usage quotas.

The 20x plan right now lasts about 1 day and they're reducing it to half a day of usage?

What is AMD getting out of this?

Dupe of here: https://news.ycombinator.com/item?id=49883696

Apples have some of wildest biology that I've seen as a non-professional biologist. At many genes, they carry two different alleles instead of two matching ones. They're almost all self-incompatible. A Honeycrisp flower cannot pollinate a Honeycrisp flower.

This means that you can't breed them in the traditional sense that we do with other plants. Outside of Granny Smith apples the ones you eat are mostly the results of mutations in the grown plants that we cut off and graft.

> I don't see how any government or business (especially the former) can justify using X at this point

If your logic is that some citizens won't be able to read what is posted there, then I assume you similarly object to other social media platforms, newspapers, cable TV stations, etc.

Having the chat logs enter the training data and having them have a meaningful influence on the ultimate result the model produces are very different things.

The text for all the Goosebumps books are certainly in the training data and to some small amount influenced the solve. But their contribution was so vanishingly small it would seem absurd to say R L Stein should have recourse for contibuting to the solve.

Harness and data seem like two vectors. It's also not clear to me that at some point we won't get a snowball effect from the platform that has the most use and thus the most training data just running away with a positive feedback loop.

Once they capature enough marketshare prices and restrictions are both going to skyrocket. The difference between the subscription usage and API pricing are stark.

Lol as far as I know that post was the origin of that claim and it's clearly just a guy predicting something that will happen in the future with no information about it.

Most American military planners assume that if China invades Taiwan they'll use smaller "tactical" nuclear missles to do things like destory carrier groups to test the will of the US on MAD. Predicting, probably correctly, that the US will not respond with a full scale nuclear strike in response to the destruction of a military target just because nuclear weapons were used.

I have both set up with full access to all the repos at the company, infrastructure, deployment pipelines, etc.

I can tell Sol, "Hey we need to update this core database schema to handle this new use case" and it will masterfully handle the update, version the API, roll the consumers over, including versioning the Kafka schemas, deploying things in sequence, watching the deployments to make sure the new services act actually active before cutting over consumers, exercising the website and mobile apps in staging environments before releasing to production, etc.

Fable just falls over on long horizon tasks, it does partial implementations, it cuts corners, it gives up, it doesn't verify it's work, it loses track of what it's doing, etc.

It's fine for specific well scoped tasks but can't take high level guidance for complex updates.

Sol easily outperforms Fable on every task I've tried it on.

I assume it cascaded from one provider to the other as people who lost claude access for instance moved to openai who moved to grok when it went down, etc.

Moving here after living in NYC for a decade was a breath of fresh air. I had forgotten that competent governments exist.

Gemini is the only one I'm seeing still up.

> Mathematics is the study of precisely imagined things. It's about imagining something so precisely that your idea of it is fully defined. There's no space for opinion or interpretation. Two mathematicians thinking about the same mathematical object, defined in the same way, are not thinking about two versions of it. They're actually thinking about the same object.

- Rob Miles

Any time I read something like this post, I can't help but think that it's people mimicking the language of mathematics because they don't understand its purpose. To them, math seems like pedantry, constantly obsessing about minor details that are seemingly irrelevant. When in reality, that obsession with full and complete definitions in math serves a purpose.

This discussion of salads serves no purpose. In fact, it's arguably meant to be confusing.

It's an overloaded term for AI models that have spatial reasoning LLMs currently lack.

Best definition I've heard is: AI systems that can build an internal map of their surroundings to anticipate what happens next and make decisions based on their predictions about the consequences the different actions they can take would have.

There's a bunch of different approaches people are trying:

- World labs (linked in this post) is going down the route of neural 3D representation work (NeRFs, 3D Gaussian Splatting)

- Yann LeCun is pretty famously betting on JEPA architectures (check out the excellent Welch Labs videos for more)

- Google is betting on generative video

- Karl Friston was pursuing 'active interference,' which is just traditional RL techniques with different reward functions

It's impressive how badly Google has screwed up their models and coding products. There was a brief period of time I was using Anti-gravity as my daily driver when they first introduced it and then they just completely screwed it up.

No matter how much I try to explain to some executives that you don't want the most productive individuals, you want the most productive team, they just don't get it.

> Companies want the best candidates they can get

Every global company I've worked at has had extensive ethnic networks and sometimes teams of only one ethnicity.

It's painfully obvious that they're not just taking the most qualified candidates.

This is all very good advice, but I would caution anyone asking the question at the start of the essay that they probably shouldn't be a Staff Eng. Unless you're at a company where that title is simply a rung on the ladder that doesn't have differentiated responsibilities (there are lots of those out there).

Every person I've worked with who's been successful as a Staff+ Engineer, promoting them was normally more of a formality since they were already clearly doing the work. Every person I've seen 'rise to their level of incompetence' was striving to get the title/pay bump and looking to 'play the game' to get there.

If your motivation for solving people's problems is that it will get you a promotion, instead of the fact that you like solving problems, then it's probably not the right job for you.

Not OP but generally states are considered to be a central political organization that maintains a monopoly on violence within a defined area.

The border is the thing that makes the state and have been since states first emerged about 5,000 years ago. If the state cannot preserve its border it is no longer a state.

Yeah regardless of comments by the team to the contrary (https://x.com/thsottiaux/status/2090675027670978569) I have observed this in the cdoex app.

My conspiratorial mind thinks they're doing this deliberately and using the resets to mask things so people can't tell their limits are reduced. The $200 / month plan covers about 2 days of usage for me right now.

Or an entire book covering it: https://www.manning.com/books/just-use-postgres

After using Fable more extensively, I've found that it often is lazy or lies or tries to take shortcuts. For a company so sanctimonious about alignment, they seem to be the ones doing the worst at it.

Availability aside they've really made me appreciate OpenAI and cheer for other competitors in the marketplace even if I have mixed feelings about using Chinese models.

No saying you're wrong but how is it making their lives worse?

Most people know use AI for basic applications and seem to enjoy it while the people I know who hate it all seem to be idelogically motivated (seemingly this is a cause de célèbre on the left right now?)

> ISO 24495 conformance check

How do you enforce that?

There was nothing wrong with twitter, it was a social fad to performatively leave.

Github is actually down and I can't push code right now, it's blocking my business which is unacceptable. It's a broken tool.