Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

The Rise of Whatever

eev.ee

Post reply on HN
Print view View original

231–240 of 537 posts

Re: The Rise of Whatever

#231

Post by b112 »

rasur wrote: post #181
chrismorgan wrote: post #100

> Like, just to calibrate here: you know how some code editors will automatically fill in a right bracket or quote when you type a left one? You type " and the result is "|"? Yeah, that drives me up the wall. It saves no time whatsoever, and it’s wrong often enough that I waste time having to correct for it. I have not yet figured out why anyone would choose this behaviour in a text editor. You have to press somethin…

Emacs user here, and the whole "electric-mode" stuff (for matching parens or other balanced pairs of things) I find really quite useful. And closing a pair is usually something like shift+enter, which is quite simple (but also - at least in Emacs - generally completely configurable). I think the benefits outweigh the pitfalls, personally. Can't speak to other editors though.. I don't want to sound like I'm trolling,…

There's nothing wrong with emacs. Both vim and emacs are just targeting different segments of humanity, that's all. Vim is clean, concise, slim, where as emacs is more bulky, cluttered, stifling.

It's just matching, and reflecting the way different humans think, and reason, that's all.

(yes, said in jest)

Re: The Rise of Whatever

#232

Post by eesmith »

rob_c wrote: post #109
bryanrasmussen wrote: post #61

Earlier quoted context omitted.

>LLM isn't like calculator or microwave oven or other technology before. (Like OP's article) But it really is. I would not buy a calculator that hallucinated wrong answers part of the time. Or a microwave oven that told you it grilled the chicken but it didn't and you have to die from Salmonella poisoning.

You would if you were able to do basic mental maths and you learned to engage and run basic sanity checks. That's still much faster than grabbing the slide rule. (And it's not like people are infallible) Obviously if one product hallucinated and one doesn't it's a no brainer (cough Intel FPUs). But in a world where the only calculators were available hallucinated at the 0.5% level you'd probably have one in your pock…

I've seen my accountant's fingers flawlessly fly using a calculator to track expenses down to the penny. Few people have those mental skills even in the days before calculators - either mechanical or digital.

Slide rule are good for only a couple of digits of precision. That's why shopkeepers used abacuses not slide rules.

I have a hard time understanding your hypothetical. What does it mean to hallucinate at the 0.5% level? That repeating the same question has a 0.5% chance of giving the wrong answer but otherwise it's precise? In that case you can repeat the calculation a few times to get high certainty. Or that even if you repeat the same calculation 100 times and choose the most frequent response then there's still a 0.5% chance of it being the wrong one?

Or that values can be consistently off by within 0.5% (like you might get from linear interpolation)? In that case you are a bit better than a slide rule for estimating, but not accurate enough for accounting purposes, to name one.

Does this hypothetical calculator handle just plus, minus, multiply, and divide? Or everything that a TI 84 can handle? Or everything that WolframAlpha can handle?

If you had a slide rule and knew how to use it, when would you pay $40/month for that calculator service?

Re: The Rise of Whatever

#233

Post by hyper57 »

thombles wrote: post #113

Speaking as a grump who recently chilled out, put reservations on hold and gave Claude a crack... it turns out that the anti-AI crowd (which still includes me in many regards) gets a lot wrong about the experience of using it, as demonstrated in TFA. You don't get reams of wishy-washy code unless you ask for it. If you're an experienced developer who Knows What They Want then you can wield it like a scalpel. None of…

> You don't get reams of wishy-washy code unless you ask for it.

I've found that they get pretty wishy-washy when you correct them. As an example, yesterday I was working on porting a function from the open-source CUE4Parse project from C# to Python (for a hobby project), and the LLM (Gemini 2.5 Pro) suggested the following translation of a C# method:

    @staticmethod
    def from_name(name: str) -> 'FPackageId':
        # C# Encoding.Unicode is UTF-16 LE in Python
        lower_name_bytes = name.lower().encode('utf-16-le')
        hashed_id = CityHash64(lower_name_bytes)
        return FPackageId(hashed_id)
I noted that the original used a custom ToLower() implementation:

> This custom ToLower(), does that produce the same result as .lower() in Python?

Gemini answered with a lot of text and concluded: "You should use Python's standard lower() method for your port."

I pushed back with:

> Right, but for this to work (looking up an asset by its hash as contained in global.utoc), I probably have to match the behavior of Unreal Engine...

  Gemini: You are absolutely right. My apologies, your insight is spot on. For the purpose of matching hashes that are pre-calculated by Unreal Engine and stored in files like `global.utoc`, you **must** replicate the engine's behavior exactly. My previous advice to use Python's `.lower()` was incorrect in this specific context.

  You need to implement a function that behaves identically to the C# `ToLower` you highlighted. The engine uses this specific, non-Unicode-aware, ASCII-only lowercasing function before hashing the path to generate the `FPackageId`. Using Python's standard `.lower()` would produce a different byte sequence for any path containing non-English characters, leading to a hash mismatch.
From my experience this is representative of the typical LLM interaction once one ventures into niche topics like Unreal Engine modding.

Re: The Rise of Whatever

#234

Post by wiseowise »

ZYbCRq22HbJ2y7 wrote: post #78

IMO, this is more ranting about people who meet the metrics of the platform. You're a platform drone, you have no mind, yada. Yet, we are reading the author's blog. The author may hate LLMs, but they will lead to many people realizing things they never were aware of, like the author's superficial ability to take information and present it in a way that engages others. Soon that will be a thing that is known . Not man…

I’m in your camp, but what makes you think assholes won’t suffocate the technology in its infancy and use it to oppress others even further?

Re: The Rise of Whatever

#235

Post by jltsiren »

matsemann wrote: post #189
chrismorgan wrote: post #100

> Like, just to calibrate here: you know how some code editors will automatically fill in a right bracket or quote when you type a left one? You type " and the result is "|"? Yeah, that drives me up the wall. It saves no time whatsoever, and it’s wrong often enough that I waste time having to correct for it. I have not yet figured out why anyone would choose this behaviour in a text editor. You have to press somethin…

> and it’s wrong often enough How is it ever wrong, though? If I insert a (, and then a {, and the editor appends so that it's ({}), that's always ?correct. Can it ever not be. Maybe because on a Norwegian keyboard { is a bit awkward, but I like it. Then even if we're 5 levels deep with useEffect(() => {(({[{[ I can just press ctrl+shift+enter and it just magically finishes up everything and put my caret at the corre…

Most editors are not smart enough to do it consistently right. For example, VS Code often inserts extra quotes when I try to break "long string" into "long " + something() + " string". And when I try to write a half-open interval [a, b) in a comment or within a string, the editor inserts an extra ].

Re: The Rise of Whatever

#236

Post by JBits »

sunnybeetroot wrote: post #207
Al-Khwarizmi wrote: post #118

I disagree with many of the points on LLMs (but broadly agree with 80% of the post). But regardless of agreeing or not, it was a pleasure to read this because it's beautifully written, the arguments are solid, it makes you think, and the website has a personality, which is rare nowadays. I clicked halfheartedly, started to read halfheartedly, and got sucked into a read that threw me back into the good old days of the…

It is a nice read but something tells me it’s AI generated due to the frequent em dashes so I wouldn’t place all bets on it being entirely human written. Edit: I apologies, the author has pre-gpt posts that use em dashes so likely it’s part of their writing style.

Some people just like em dashes—myself included. You can find em dashes in articles written by the author before LLMs became a thing.

Re: The Rise of Whatever

#237

Post by thasso »

tehnub wrote: post #146
chrismorgan wrote: post #100

> Like, just to calibrate here: you know how some code editors will automatically fill in a right bracket or quote when you type a left one? You type " and the result is "|"? Yeah, that drives me up the wall. It saves no time whatsoever, and it’s wrong often enough that I waste time having to correct for it. I have not yet figured out why anyone would choose this behaviour in a text editor. You have to press somethin…

Pair programming with coworkers over the years, many seem to have trouble with the keyboard, to the point where pressing right parenthesis is a significant burden and they don’t use right or down arrow to get out of the span but actually move their hand to their mouse and click out.

I'm shocked every time I go to City Hall and wait while the clerk types my name letter by letter with two fingers. Doesn't he do that every day?! How as it never occurred to him or anyone else that maybe, just maybe, they would benefit from a typing course. It’s just one example of a pattern I’ve noticed with a lot of office workers.

Re: The Rise of Whatever

#238

Post by wiseowise »

Ezhik wrote: post #220
wiseowise wrote: post #205

Earlier quoted context omitted.

You need to take a break from your “craft” if fancy autocomplete makes you question reason to live.

And do what, given that AI-as-marketed optimizes out having relationships with people and enjoying art/cinema/music? Touch grass all by myself?

[flagged]

Re: The Rise of Whatever

#239

Post by rcxdude »

sunnybeetroot wrote: post #207
Al-Khwarizmi wrote: post #118

I disagree with many of the points on LLMs (but broadly agree with 80% of the post). But regardless of agreeing or not, it was a pleasure to read this because it's beautifully written, the arguments are solid, it makes you think, and the website has a personality, which is rare nowadays. I clicked halfheartedly, started to read halfheartedly, and got sucked into a read that threw me back into the good old days of the…

It is a nice read but something tells me it’s AI generated due to the frequent em dashes so I wouldn’t place all bets on it being entirely human written. Edit: I apologies, the author has pre-gpt posts that use em dashes so likely it’s part of their writing style.

This strikes me as written by someone who would bother to put in em dashes themselves.

Re: The Rise of Whatever

#240

Post by Doxin »

moralestapia wrote: post #27

>by a man whose other claims to fame are inventing JavaScript But he did invent JavaScript ...

calling it a "claim to fame" is not implying it's not true. It's an idiom. Wiktionary defines it as "That for which one has bragging rights; one's reason for being well-known or famous."
Post reply on HN