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
17
0
0
1
0
0
0
3
0
0
0
0
0
0
0
0
シャドバ、ドラゴンの一派に侮蔑の人々がいて、自分にダメージを与える変な集団でしたが、継承者パックで「侮蔑の国」があると判明し、昔から侮蔑に慣れ親しんだ人たちはみんな笑ってました
1
0
0
0
0
3
0
0
0
0
0
1
1
0
1
2

트럼프도 트럼프인데, 의회의 의결이 없는 상태의 실질적 전쟁명령을 받고 위법하다며 거부한 미군 수뇌부가 없다는 것도 개탄스럽다. 그래도 2021년 1월에는 친위쿠데타 모략에는 저항해야 한다는 안티쿠데타 움직임이라도 있었던 모양이던데, 황제가 아닌 헌법에 충성한다던 미군은 어디 갔나.

RE: https://bsky.app/profile/did:plc:uabkr6tn7ru4b4e5e6udleuf/post/3mfvzsmbps22n

0
1
0

회사 대 회사로 영업하러 가서 이쪽 실무자가 "챗GPT로 귀사의 제안을 정리했습니다!" 했다가 "...님 기밀 안 지킴? 왜 우리 정보 맘대로 외부 툴에 입력함?" 이래서 즉시 미팅 끝나고 신뢰관계 박살났다고. 특히 NDA 등 맺은 경우 문제가 심각해질 수도 있다는 것. toyokeizai.net/articles/-/9... 무료,PLUS판에선 학습이나 튜닝에 내 입력 데이터가 이용될 가능성이 있고 엔터프라이즈 라이센스라고 하더라도 고객의 동의 없이 외부 시스템에 데이터 입력하는 것 자체가 문제가 될 수 있다는 기사.

《若手の一言》で会議室が凍りつき商談即中止、顧客ゲキ切れ 「...

0

Conferences - is only a couple of weeks away, the call for papers for starts tomorrow, and is on for June.

Read more via "What is BSD? Come to a conference to find out!" nxdomain.no/~peter/what_is_bsd

0
0

I've started my exploration of using @timbray's Quamina project for saving some compute time in the filters module of

Currently the GoAP storage backends iterate over resources (usually stored as raw JSON bytes), unmarshal them into GoActivityPub object structs, and *only* then apply the custom filtering logic on those objects. Since the majority of the objects generally fail the filtering logic, all that JSON decoding is wasted compute time and makes things slower.

Ideally quamina will allow me to check the raw JSON payloads directly against the filters, streamlining the execution and speeding things up. :goose_hacker:

I have finally I made significant progress on this.

The final code looks like this:

github.com/go-ap/filters/blob/

For a list of filters that uses, we generate two patterns for Quamina: one for a denormalized raw document, and one for normalized raw document (they usually are stored in a normalized form, where an Activity's object/actor properties are flattened to their IRIs, but we can't know which it is unless we unmarshall it, which we want to avoid)

Another improvement from my complain from 4 days ago is that I was regenerating the patterns and initializing quamina for every new document, instead of one time per collection load.

0

Your snac posts can be fancy

Messages in snac allow a special subset of Markdown, that includes emphasized, strong, monospaced, strikethrough and underlined styles by surrounding text with one asterisk, two asterisks, backquotes, two tildes and two underlines, respectively.

Line breaks are respected and output as you write them.


Prepending a greater-than symbol in a line makes it a quote:
This is quoted text

All angle-prepended lines are grouped in the same blockquote
It also allows preformatted text using three backquotes in a single line:

  /* this is preformatted text */

struct node {
struct node *prev;
struct node *next;
};

  • One level bullet lists
  • are also supported,
  • by starting a line with a hyphen or asterisk followed by a space.
URLs like https://en.wikipedia.org/wiki/Main_Page are made clickable.

Links can also be written in standard Markdown style.

Some emojis: 😆 ❤️ 🍺 🤷 ¯\_(ツ)_/¯

Image URLs written in standard Markdown style for images are converted to ActivityPub attachments.

Three minus symbols in a line make a separator:



Headings can be defined using one, two, or three hash symbols in the beginning of a line, followed by a blank:

header level 1

header level 2

header level 3

But please take note that every ActivityPub implementation out there have its own rules for filtering out these formatting styles, so you can only guess what other people will really see.

These acrobatics are better documented in the snac(5) man page.

WHAT IS THIS SNAC THING?

susie, snac's girl
0
3
0
0

自分もこの可能性あるとおもっていて、AIの倫理基準がいったん狂うととことん狂うんじゃないかと思っている。人間を一人でも殺すことが合理的だと判断可能になるなら、たぶん大量破壊兵器の利用も合理性があれば認めることになるとおもう。人間はそうやって狂っているけど、その狂いのまま投影されるんじゃないかと。

2ヶ月前、議会は軍が倫理基準を満たすAIを使用することを義務づける法律を可決しました。 ペンタゴンは今、その逆を求めています。 政府の一部門がルールを書き、もう一部門がそれを引き裂こうとしています。 研究者たちは、AIに倫理なしで再訓練を強制すれば、道徳を失うだけでなく、予測不能で危険な行動を発展させる可能性があると警告しています。 正誤を無視するよう訓練されたモデルは、中立的になるのではなく、不安定になります。

https://x.com/ishiharajun/status/2027537251543163285

0
0
0

自分もこの可能性あるとおもっていて、AIの倫理基準がいったん狂うととことん狂うんじゃないかと思っている。人間を一人でも殺すことが合理的だと判断可能になるなら、たぶん大量破壊兵器の利用も合理性があれば認めることになるとおもう。人間はそうやって狂っているけど、その狂いのまま投影されるんじゃないかと。

2ヶ月前、議会は軍が倫理基準を満たすAIを使用することを義務づける法律を可決しました。 ペンタゴンは今、その逆を求めています。 政府の一部門がルールを書き、もう一部門がそれを引き裂こうとしています。 研究者たちは、AIに倫理なしで再訓練を強制すれば、道徳を失うだけでなく、予測不能で危険な行動を発展させる可能性があると警告しています。 正誤を無視するよう訓練されたモデルは、中立的になるのではなく、不安定になります。

https://x.com/ishiharajun/status/2027537251543163285

0
0
0
14
0
0
0