What is Hackers' Pub?

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.

0

【鯖缶の皆様方と利用者にお知らせ】

6/27 14:00 (JST)〜6/28 23:00 (JST)頃まで空の上にいる都合上連絡がつかなくなるのでサーバーが落ちていても復旧に時間がかかります。ご容赦ください。

0
0

Setting aside copyright/commercial/other aspects, a thought about writing blog posts that LLMs train on. When people write posts, they (a) feel good about helping others, and (b) hope to get some credit and visibility for doing so.

When mediated through LLMs, no longer the satisfaction that your consumer is a human (who might comment, thank, share, etc.); nor the cred that comes from people remembering the author, posting on HN, etc.

That is, it totally destroys the incentive structure.

0
0
0
0
0
0
0
0
3

It's really weird to me that posting your own content online that you never said was okay for an AI to trawl could be considered a "cyberattack". If you just put something out there, knowing that AIs will scrape your website, that is "cyber defense".

Google search for the words "poisoning AI". the first three results from CroudStrike, IBM, and nightfall.ai read similarly

CroudStrike: "What is Data Poisoning? Data poisoning is a type of cyberattack in which an adversary intentionally compromises a training dataset used by an AI or machine learning (ML) model."

IBM: "What Is Data Poisoning? Data poisoning is a type of cyberattack where threat actors manipulate or corrupt the training data used to develop artificial intelligence (AI) and machine ...[cuts off]"

nightfall.ai: "Data Poisoning: The Essential Guide. Data poisoning, as its name suggests, involves the deliberate and malicious contamination of data to compromise the performance of AI and ML systems. Unlike ... [cuts off]
0
0
1
0
0
0
0

It's really weird to me that posting your own content online that you never said was okay for an AI to trawl could be considered a "cyberattack". If you just put something out there, knowing that AIs will scrape your website, that is "cyber defense".

Google search for the words "poisoning AI". the first three results from CroudStrike, IBM, and nightfall.ai read similarly

CroudStrike: "What is Data Poisoning? Data poisoning is a type of cyberattack in which an adversary intentionally compromises a training dataset used by an AI or machine learning (ML) model."

IBM: "What Is Data Poisoning? Data poisoning is a type of cyberattack where threat actors manipulate or corrupt the training data used to develop artificial intelligence (AI) and machine ...[cuts off]"

nightfall.ai: "Data Poisoning: The Essential Guide. Data poisoning, as its name suggests, involves the deliberate and malicious contamination of data to compromise the performance of AI and ML systems. Unlike ... [cuts off]
0
0
0

Is there a concise description somewhere of the Google Docs data model? Are documents located in just one place or not? Why is there a "Move" action? Does that mean it won't be where it was before? Why can't I find a document's current location?

0
0
0
0
0
0
0

It's great folks are talking about internet resiliency, but resist the temptation to go build the cool fun mesh network first and then the services that run on it second. You need the services first. You're far more likely to see your country or area getting islanded than you are to see your ISP suddenly switched off, and you're better placed to rebuild IP networks from a single access point where you can access a wiki and send local email than you are building services on your empty mesh

0
0

I got this email and I fucking hate it because it says "If you don't want to use these features, you can turn them off in the Apps settings page." but there's no explanation how and no link to an explanation how and i you look at the Apps setting page you see fucking nothing relevant hachyderm.io/@SnoopJ/114740795

0
0
0
0
0
0
0
0
0

1. Is anyone else getting routed to chinese language when they visit this URL? because i am: docs.joinmastodon.org/methods/ (i end up at the URL docs.joinmastodon.org/zh-cn/me - note the "zh-cn" in the path.
i doublechecked my browser locale, VPN, etc. etc. and there's nothing configured for mainland chinese locale, but this happened even after clearing cache data for joinmastodon.org

2. who maintains the site?

cc: @mastodonmigration @FediTips @fediversereport @LaurensHof

0
0
0
0
0
0
0

다리 곳곳은 무너지고 갈라져 콘크리트와 자갈, 시뻘건 철근이 속살을 드러냈습니다. 한때는 물자를 실은 마차가 오가고, 저마다의 희로애락이 담긴 사람들의 발걸음이 지나던 이 다리를 이제는 북쪽에서 흘러오는 화강과 무심히 서 있는 나무와 꽃, 풀, 산만이 기억하는 듯합니다.

한국전쟁 75년의 기억을 갖고 힘겹게 서 있는 다리

0
0
0
2
0

[내란 특검 알림] 법원은 어제 청구한 윤석열 전 대통령에 대한 체포영장을 피의자가 특검의 출석요구가 있을 경우 이에 응할 것을 밝히고 있다는 이유로 기각함 이에 즉시 특검은 윤석열 전 대통령 및 변호인에게 6. 28.오전 9시 출석을 요구하는 통지를 하였음 출석 요구에 불응 시 체포영장 청구를 검토할 것임

RE: https://bsky.app/profile/did:plc:oeors5le2tmvqzojsjxxqwro/post/3lsgh6ljzps27

0

Signal priorisiert jetzt auch "Sicherheit" über Barrierefreiheit...

Und dass Microsoft ein Feature veröffentlicht, das Barrierefreiheits-APIs und assistive Technologien kaputt macht, und der einzige Weg drum herum ist, Copilot KI-Scheiß zu erlauben, ist ganz genau so kacke.

Bitte endlich barrierefreie freie Software...

signal.org/blog/signal-doesnt-

Man kann das ausschalten und dann gehts wieder:

> To help mitigate this issue, we made the setting easy to disable (Signal Settings → Privacy → Screen security)

Nur kind of ironisch, dass man da gar nicht erst hin kommt, weil die assistiven Technologien natürlich auch *das* nicht ansehen können, solange es aktiviert ist.

0
0
0
0
0
0
0