What I really want is the ability to do more operations which are decoupled from the editor. Even just being able to reliably type check Unity C# code without a running editor would be great, or running unit tests. Would really improve things like CI as well as the ability to run parallel development sessions.
They do disclose that they scored much lower than Fable on SWEBench Pro, which is a pretty high-quality benchmark. I think it's partially just about what they choose to emphasize...
To clarify, the more accurate description would be "Testing how well LLMs can follow the rules of Magic", right? There is no actual evaluation of how "well" they are playing?
This is basically just the normal dynamic with American tax law where tax jurisdictions are terrible at coordinating, so they end up approving things and agreeing to tax things at a very low level in order to win the competition. Even when states/counties try to work together on this stuff there's a huge defector problem, like "hey I can back out of this multistate tax compact agreement and get 500 new jobs which will let me win local reelection".
I suppose you can reduce a lot of both good at bad things about the country to "because federalism".
I wonder how Paradox handles this stuff in their games like Europa Universalis. Have they ever made a "the pope wants you to switch calendar systems" event which changes the actual in-game date?
I really fundamentally do not understand what problem Gas City solves that is not already solved by normal subagent orchestration patterns. If you want to call your main LLM session the "mayor" and have it delegate its work out to planners and coders and reviewers and QA and so on, this is already a thing you can do! If you want to do this in a reusable way you can create skills and subagent definitions and use /commands, etc. Why do we need hundreds of thousands of lines of opaque Go code to accomplish any of this?
There's a real hobbyist vs professional distinction with Claude Code. For professionals, including when I use it at work, we're generally super happy to have Claude spawn as many subagents as possible and burn more tokens to get a better result. Hobbyist users on a $20/month plan, though, generally want more conservative behavior.
It's hard for Anthropic to cater to both sets of users with one model.
No "max" or "pro" equivalent? I wanted to get a new Macbook Pro, but there's no obvious successor to the M4 Max available, M5 looks like a step down in performance if anything.
Does the "caching containers for Codex Cloud" mean I have some chance of being able to reuse build artifacts between tasks? My Rust project takes around 20 minutes to set up from scratch in a new Codex environment, which seems extremely expensive.
- Superintelligence poses an existential threat to humanity
- Predicting the future is famously difficult
- Given that uncertainty, we can't rule out the chance of our current AI approach leading to superintelligence
- Even a 1-in-1000 existential threat would be extremely serious. If an asteroid had a 1-in-1000 chance of hitting Earth and obliterating humanity we should make serious contingency plans.
Second question: how confident are you that you're correct? Are you 99.9% sure? Confident enough to gamble billions of lives on your beliefs? There are almost no statements about the future which I'd assign this level of confidence to.
Also copyright duration when Star Wars was created was a maximum of 56 years, and obviously George Lucas felt that was sufficient incentive to create it!
this doesn't seem to provide much of a justification for an extraordinary claim like "the widely accepted number for the age of the universe is wrong by 4 billion years"?
If you've hired an intern before, think of Cursor like that. When you come up with an intern project plan, you usually need to give a very clear specification of what you expect, and you usually need to have the project be pretty self-contained. A lot of real problems are really bad intern projects, and a lot of problems are a really bad fit for AI. You have to be strategic about it.
Liking it so far, although the 'task' progression isn't my favorite. Feels like there are way too many tasks to do that seem like tedious busywork.
Honestly a really simple game design fix for this would be to unlock tasks more slowly as the player demonstrates more engagement with the system. That way if you are like me and mostly find them boring and repetitive, you don't feel as bad about not getting them done.
The queryable expression thing is something I struggle with all the time in Rust. It's especially bad in that language because (unlike in e.g. Java or C#) there is no way to view the Debug representation of your types in the debugger, you just get the raw memory layout which adds a huge barrier to 'what is going on with this code?' and requires you to dig around through countless nested layers to understand it.