Hacker Newsnew | past | comments | ask | show | jobs | submit | richard_cory's commentslogin

This is consistently reproducible in completions API with `gpt-5-chat-latest` model:

``` curl 'https://api.openai.com/v1/chat/completions' \ --header 'Content-Type: application/json' \ --header 'Authorization: Bearer <your-api-key>' \ --data '{ "model": "gpt-5-chat-latest", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "How many times does the letter b appear in blueberry" } ] } ], "temperature": 0, "max_completion_tokens": 2048, "top_p": 1, "frequency_penalty": 0, "presence_penalty": 0 }' ```


hilarious if true, their "gpt-oss-20b" gets it right - however, it still fails on e.g. the German compound word "Dampfschifffahrt" (Dampf-Schiff-Fahrt, steam-ship-journey/ride) because it assumes it's "ff" not "fff"


On the second try gpt-oss-20b gave me "The letter b appears once in the word blueberry."


> because it assumes it's "ff" not "fff"

Funnily enough—and possibly related—this was correct before the German orthography reform of 1996 [https://en.m.wikipedia.org/wiki/German_orthography_reform_of...]


The "gpt-5-chat" model is a non-reasoning model and these struggle because of tokens.


> For instance, you cannot write this in Haskell:

> http://krue.net/avrforth/

From what I understand about the link, you likely meant that one cannot write an interpreter for avrforth in Haskell which reads avrforth source code and executes it on bare metal, because such an interpreter will need to directly access the hardware to be able to manipulate individual bits in memory, access registers, ports, etc. and all of this is not possible in Haskell today. If this is not the case, please feel free to correct me.

However, if my understanding is correct, I don't see how this is a problem of Haskell being a functional or "leaning more towards computability theory" language rather than a mismatch of model of computation between the language and the hardware. Haskell can perform IO just fine by using the IO monad which uses system calls under the hood to interact with the hardware. If a similar mechanism is made available to Haskell for accessing the hardware directly (e.g. a vector representing the memory and accessible within the IO monad), it should be possible to write an interpreter for avrforth in Haskell. This means that the current constraint is a tooling/ecosystem limitation rather than a limitation of language itself.


> for-profit advertising itself is systematically unethical

Very interesting. Can you please expand a bit more on why do you think this is the case?


Not oc, but I share some of this sentiment. Modern advertising is heavily based on behavioral science, psychological and especially emotional manipulation. This is on top of extreme methods to hijack your attention at all cost. It might sound like hyperbole but if you read marketing case studies you realize this isn't only the norm, it's something they take pride in, especially when it appears to work (which it does).

In my view, blocking this isn't just morally just, it's absolutely necessary. I deliberately choose not to partake in this and not be a target for manipulation to the best of my ability.

Maybe there was a time when advertising was more about creating awareness instead of feeling and making you want the product, but advertising changed dramatically over the 20th century. There's quite a lot of reading material out there if you're interested.


Also not oc, but…

Without advertising, “content marketing”, and paid placements/reviews people would buy things when they desire or need them.

They’d ask friends, compare specs, and read/watch reviews before determining what to buy.

That is: without ads, people would gravitate towards buying what fits their needs best. They would make generally rational choices given the information available.

Advertising’s job is to subvert those rational choices and make people buy something, whether it’s the best option or not. In fact, even when they don’t actually want or need anything at all.

It causes people exposed to it to spend money unnecessarily, and on the wrong products and downright bad products. Some are more susceptible than others, but in the end it’s an illegitimate tax levied every time you buy something. Even if you didn’t respond to advertising when making a purchase, advertising is so ubiquitous and necessary in most markets that the price you paid probably contributed to the advertising the manufacturer had to deploy to keep up with the arms race.

There’s nothing ethical or necessary about any of this.

Ideally there would be legislation that would force business models to change, but while there is not, ad blocking is absolutely an imperative.


"That is: without ads, people would gravitate towards buying what fits their needs best. They would make generally rational choices given the information available."

Not to sound snarky, būt have you met humans?If sociology and economics have shown anything, itš that human do NO make rational consumption choices


I agree with the sentiment but I’d wager the choices they make are more rational without ads than with, which is all the argument needs.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: