Hackers' Pub is a place for software engineers to share their knowledge and experience with each other. It's also an ActivityPub-enabled social network, so you can follow your favorite hackers in the fediverse and get their latest posts in your feed.
I've put together an updated version of the Sponza scene with uncompressed PNG and compressed AVIF textures. I wrote about the process and compared the results against KTX.
‘The Queensland government has emphatically declared a crocodile-inhabited river prone to flooding will host Olympic rowing and canoeing events in 2032’
구글, 개발자 지식 API와 MCP 서버 공개 ------------------------------ - AI 기반 개발 도구가 확산되면서 *최신 개발 문서에 대한 정확한 접근성* 이 중요해짐 - Google은 이를 해결하기 위해 Developer Knowledge API 와 *Model Context Protocol(MCP) 서버* 의 공개 프리뷰를 발표 - API는 Google의 공식 개발 문서를 *머신이 읽을 수 있는 Markdown 형태로 검색·조회* 할 수 있… ------------------------------ https://news.hada.io/topic?id=26556&utm_source=googlechat&utm_medium=bot&utm_campaign=1834
Staying nearby makes all the difference. Grab your discounted room for SCaLE 23x at our conference hotels before the block pricing ends. Special pricing ends this week. https://www.socallinuxexpo.org/scale/23x/venue
Honestly, I’m pretty happy with the Discord news. Maybe we can go back to having proper documentation for things. Even just wikis and forums would be a massive improvement.
그나저나 어제 본 몰트북 우려 기사에서 몰트북 공간이 "프롬프트 인젝션 공격의 온상지다" 라는 대목에서 잠깐 쓴웃음이 지어진 것이, 커뮤니티에서 이상한 명령어나 기준을 배워 와서는 직무에서 사달을 내는 것은 인간도 마찬가지가 아닌가 하는 생각이 들었다.
가끔 인간은 인간도 딱히 잘 못 하는 걸 AI라면 해낼 거라고 기대하는 것 같다.
Whatever the output gains promised by LLMs, their initial productivity surge is erased over time, and replaced by heavier workloads—and that leads to workers experiencing “cognitive fatigue, burnout, and weakened decision-making.”
LLMs have no model of correctness, only typicality. So:
“How much does it matter if it’s wrong?”
It’s astonishing how frequently both providers and users of LLM-based services fail to ask this basic question — which I think has a fairly obvious answer in this case, one that the research bears out.
(Repliers, NB: Research that confirms the seemingly obvious is useful and important, and “I already knew that” is not information that anyone is interested in except you.)
Everyone's going to keep complaining about discord but no one is going to stop using it because nothing exists that could take it's place, nothing is even close, and to delete discord is to forfeit your online social life. This is why we should have had alternatives to discord like a decade ago.
추가로 공개된 문서에서 미성년자 성착취범 제프리 엡스틴과의 관계가 알려진 것보다 장기간·다층적이었다는 게 드러난 하워드 러트닉 미 상무장관에게 초당적인 사퇴 요구가 쏟아지고 있습니다. 러트닉 장관은 2005년 이후 관계를 끊었다고 주장해왔지만, 2018년까지도 교류를 지속해온 정황이 담긴 문건들이 속속 드러나고 있습니다.
Thinking about how a white woman recently told me she was specifically trying to get her Instagram algorithm to never show her anything political. And isn’t that just the whole problem summed up right there
the challenge of 2026, for me, is going to be getting my unfinished projects and mandatory administrative paperwork to annoy me as much as people strutting around being confidently wrong about the possibilities of AI