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

If only they managed to tell the mobile app to tell the model to reply in English to English prompts.

I suffix everything with "Reply in English", and even so I‘m getting lots of Chinese.


I think their system prompt is in Chinese and probably has instructions to prioritize answering in Chinese, since this has never happened to me via API, where I (or the coding harness) set the system prompt.

I'm starting to have chinese characters bleed into claude as well. Perhaps a sign of the times. Understanable for a chinese first model but an english first (supposedly) model? wild stuff.

I also love the gaslighting of some models, like ChatGPT mixing in words with cyrillic letters and when asked about it answers: "it can look as Slavic to the eye" and "sorry that it came across as Russian"

Funnily, one of the annoying writing quirks of Claude/GPT in Russian is that it constantly mixes in random English words

"i'm sorry, i left the task 半done"

I just started learning Chinese instead, like they want us to

seriously


English isn't the first language for me as well so I don't see any problem with that

Yes, this is one of the few issues with Deepseek; their chat pages and the app all respond in Chinese. However, i think i have only had it happen once when using the API, and im using it for hours each day for the last... couple of months?

last couple of weeks, before they've unified instant and expert the former always replied in chinese unless steered, expert was by default english

Same issue on desktop. Would be nice be able to set a prefix or postfix for every prompt.

that's my only gripe with deepseek honestly

I occassionally get Chinese characters interlaced with English in Google AI Mode, too.

nothing to do with mobile app, I have same issues while using it on desktop browser, it will never remember to use English permanently, even within one conversation

Karma farming with predictable URLs about to become interesting. Simply flag it.


This page does not exist at the time of this comment, and I‘m so sick of people submitting predictable URLs before they go live, just to be first. We also see that with LLM model releases regularly.

Licenses are contracts, just like an EULA.

And if you expand the acronym EULA you will actually find that it is… a license agreement.


Yes, and the point is that in order to obtain and keep the license, you have to fulfill additional clauses that you agreed to by accepting the EULA.

If EULA is the same as a license, why would MSFT not just use a license? The intellectual contortions of the AI boosters here are stunning.


"an end-user license agreement is the same as a license agreement, so why do they use an end-user license agreement instead of a license agreement?"

I don't understand this question.


No one but, you, an AI moron, has asked this question. Get your AI girlfriend to explain it to you.

And the AI swine reacted by downvoting the top comment again now that arguments for an EULA are made. Fucking fascists.

> BUSL is OSI-approved

Of course it isn‘t.


I was a bit surprised to read that there was an OSI-approved license with usage restrictions and wondered if I had missed the memo. But you seem to be correct and it's not OSI-approved and therefore is not generally considered to be an open source license.


> Quarto 2 is a full rewrite of the Quarto CLI, written from the ground up in Rust

Richard McElreath: https://bsky.app/profile/rmcelreath.bsky.social/post/3muofvc...

> In the actual candy study, the participants were all students, status was manipulated in the lab by telling some students they were better than others, and "nearly twice" means 1 piece vs 2 pieces.

> oh and the researchers told participants they could take some candy. There was no stealing!


Good luck finding an advisor/a department that awards PhDs for that.

“It’s not novel”.


I think you're right that this will be the most common objection, but wrong that it will be exceedingly difficult to find an advisor that would support this. I assume that if you _can_ overturn some foundational result, you won't face criticism over novelty (potentially you can argue that overturning important accepted results is inherently novel). Then the rest is just math:

Assume that n% of foundational studies are demonstrably incorrect if replication were attempted. Then if you replicate X foundational studies, there's a 1 - (1-n/100)^X chance that you'll show one foundational study to be incorrect. Then I think you just need to find an advisor who believes the incorrectness rate n, and the replication attempts X is high enough for you to revert a foundational study over the course of your PhD.


https://de.wikipedia.org/wiki/Konrad-Zuse-Museum_H%C3%BCnfel...

Apart from a quite a few museums having his machines (or replicas).


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

Search: