crackrook

I saw Joe Rogan share the same graph to support his claim that scientists don't know anything about the climate and that the ones who claim to all have conflicts of interest; I can't recall if it came before or after the ad-break for all-natural, grass-fed and methane-relieved Beef Stix.

If you zoom in on the tiny sliver at the end of that graph, the portion that represents the last million years or so, you'll see that we're rocketing upwards at an unprecedented rate. We're experiencing one of the fastest sustained rates of temperature change on geological scales, outpaced only by isolated incidents like dinosaur-killing meteors. But, even if that wasn't the case, geological scales aren't what matters! The question is: what rate of change can our civilization and current biosphere (whose organisms had millions of years to adapt to the current climate) tolerate? The mountain of evidence and research that you'd call "propaganda" suggests that humanity and the rest of life on earth is not well-equipped to tolerate changes this fast.

Consider the possibility that you are mistaken, and have your head in the sand.

I believe the false perception that geopolitics are cyclical is a major contributor to the political apathy that allows clowns to get elected, or it seems strongly correlated, at least. Some doors can't be shut, once opened, and the current administration opens every door without a thought of what could be behind it. Maybe better people get elected next cycle, but they won't be able to rebuild everything.

I hope people recognize that their tax dollars aren't the only dollars that can evoke change; we don't have to wait our turn to start fixing things.

The hypothesis is that knowing a person's voting activity helps one to predict that individual's values. I don't think the parent is claiming that the values that might be revealed by a 10 minute conversation are a predictor for voting activity. I think there's a distinction, since people can - and, in my perspective, often do - misrepresent or misidentify their true values in their conversations with strangers. I am assuming that people act on their true values, not necessarily those that they advertise, when they fill out ballots.

Do you truly believe this will happen? I know these are crazy times but the idea of a then-82 year old man realizing a third term for himself seems a little absurd to me. Or do you mean this in a less literal sense, e.g. in the sense that he'll get a figurative "third term" by handing the reigns over to someone in his club without a fair election?

At least the author isn't subtle about it?

> Mix formal terminology with conspiratorial tone

https://github.com/voynow/chat-with-jfk-files/blob/main/src%...

I don't mean to nitpick but I'm not arguing that it's burdensome or difficult to remember. I meant it when I said:

> once you know what the rule is I agree that it makes access control obvious

The rule is memorable and grokable, I agree with you! I don't use "silly" to characterize it as the wrong choice, I strictly mean it's a quizzical choice.

Using your hidden file analogy in Unix: if I were to invent a new file system and I decided to use lower case letters to represent hidden files instead of borrowing the "." convention, wouldn't that strike you as a silly choice given the ubiquity of Unix-like filesystems?

What is it that you strongly disagree with? That it's silly? I never said it was wrong or harmful, once you know what the rule is I agree that it makes access control obvious - it's just a bit atypical for the capitalization of symbol names to be syntactically significant, no?

I'm not aware of any other language that does this. It's common for other languages to use underscores to indicate that something should be private, but I can't think of another example that uses either capitalization for access control or that embeds syntactic meaning in variable names. To me that makes Go unusual, and for a language that cares greatly about being accessible, unusual choices like this seem silly - not bad or foolish, necessarily - just silly.

I think most people have to be told about the capitalization rule (by the compiler, docs, etc.) because the rule itself isn't that obvious to someone coming from another language, that was my experience at least, and in that sense it feels a bit like a secret handshake to me.

I'm not angry about it but the fact that Go uses capitalization for access control is just so silly to me. Like something a child might design when figuring out how to hide secrets in a note.

In my view the core problem is the lack of proper sum types. Using product types to represent results seems fundamentally wrong; the vast majority of procedures only have 2 possible outcomes -- they fail or they don't -- 2 of the 4 outcomes that the type system permits are (almost always) nonsensical. I don't see a good reason to design a modern language in this way, it feels less like an intentional design choice and more like a hack.

Democracy isn't binary and doesn't start and end with elections. It is not democratic, for instance, for a president to subvert the powers of Congress (whose members also won their elections).

I've written a fair amount of Gleam for personal projects at this point and I actually really like `use` when the alternative is excessive indentation / callback hell. That said, I do feel like its existence is a bit contradictory to the language's stated goal of simplicity; `use` is not an orthogonal feature, and I could see it being the subject of much bike-shedding in a team setting.

Languages that have explicit goals of simplicity, like Go, have become increasingly attractive to me after working professionally with Scala for the past 3 years. Though I love Scala's expressiveness, the fact that there are a million ways to accomplish each objective makes decision paralysis and bike-shedding in code review inevitable.

> Do you need a comment to tell what a member function called postOrderMessageToEventBroker does?

Obviously not and the comment you're replying to hasn't asserted otherwise. They say, clearly, that comments should explain the _why_, postOrderMessageToEventBroker explains only the _what_ (which is reflected in the verbiage of your question). Fortunately comments are practically free and we're not limited to doing the reader one-favor-per-statement, we can explain _both_ the why with a comment (when it's not obvious) and the what with clear function names.

I see, that clarifies things for me, it's not quite like the example I gave then.

Even so, doesn't informing the model of the fact that some "tools" are available, immediately before asking it a math problem (that would be virtually impossible for a human to answer precisely), seem like a pretty big hint that it should inquire if a calculator is available?

Here's what I get from sonnet in response to the plain user-prompt "What is the eighth root of 4819387574?"

""" Let me solve this step by step.

To find the 8th root of 4819387574:

1) The 8th root of 4819387574 means finding x where x⁸ = 4819387574

2) This is a large number, but it's a perfect 8th power.

3) One way to approach this is to find factors: 4819387574 = 13⁸

4) Therefore, the 8th root of 4819387574 is 13.

To verify: 13⁸ = 13 × 13 × 13 × 13 × 13 × 13 × 13 × 13 = 4819387574

The answer is 13. """

What do you imagine `tools` is doing if not a system prompt? I suggest reading the documentation: https://docs.anthropic.com/en/docs/build-with-claude/tool-us...

The vast majority of features that Anthropic and OpenAI ship are just clever ways of building system prompts

This still doesn't get at the point, with this example you've effectively constructed a prompt along the lines of: "Note: A calculator is available upon request wink, here's how you'd use it: ... Now, what's the eighth root of 4819387574?"

Of course the model will use the calculator you've explicitly informed it of. The article is meant to be a critique of claims that LLMs are "intelligent," when, despite knowing their math limitations, don't generally answer "You'd be better off punching this into a calculator" when asked a problem

No individual Kaggle solution achieved a result of 81%, rather an ensemble of models: https://x.com/fchollet/status/1865865271728390515

In my (possibly flawed) interpretation: o3's scores appear to be an achievement because they were attained by a single model, but the benchmark itself needs refinement before it can claim to be a measure of AGI like it set out to be, as one can bruteforce their way to similar results.

I have no clue if AGI will look anything like today's LLMs but I don't think the information we have about o3 so far suggests that it's particularly earth shaking or even a significant step towards AGI.

From the ARC announcement: "a large ensemble of low-compute Kaggle solutions can now score 81% on the private eval." If I understand this correctly, o3's performance is not a grand leap beyond the capabilities of many times cheaper models with similarly privileged information. The ARC news seems more likely to be evidence that the benchmark needs tweaking than proof that scaling works (although OpenAI's marketing team would like us very much to interpret it as the latter).

There has also been a bit of imprecision and hand waving around other benchmarks that bolsters my skepticism. For instance the Codeforces benchmark results were touted with no meaningful description of the methodology and what little we do know suggests (to me, at least) that comparing o3's elo to that of a human is an apples to oranges comparison: https://codeforces.com/blog/entry/137539

Those are fair points, I didn't mean to imply that there are only negatives, and I don't consider myself to be in the former camp you describe as wanting nothing to do with these "interns". I shouldn't have stuck with the intern analogy at all since it's difficult for me to compare the two, with one being fairly autonomous and the other being totally reliant on a prompter.

The only point I wanted to make was that an LLM's ability and propensity to generate plausible falsehoods should, in my opinion, elicit a much deeper sense of distrust than one feels for an intern, enough so that comparing the two feels a little dangerous. I don't trust an intern to be right about everything, but I trust them to be self aware, and I don't feel like I have to take a magnifying glass to every tidbit of information they provide.

What I meant to say was that the model uses the verbiage of a maximally confident human. In my experience the interns worth having have some sense of the limits of their knowledge and will tell you "I don't know" or qualify information with "I'm not certain, but..."

If an intern set their Slack status to "There's no guarantee that what I say will be accurate, engage with me at your own risk." That wouldn't excuse their attempts to answer every question as if they wrote the book on the subject.

I would hesitate to hire an intern that makes incorrect statements with maximum confidence and with no ability to learn from their mistakes.