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 love how #Deerhoof shared their new track in a number of posts on Craigslist, in their own words "a nice and non-Nazi tech platform".
https://newyork.craigslist.org/brk/crs/d/new-york-singer-available-for-your-party/7835619551.html
'공공장소 흉기 소지' 최고 징역 3년…형법 개정안 국회 통과
공공장소에서 흉기를 꺼내 불특정 다수에게 공포감을 조성하는 경우 징역형으로 처벌할 수 있도록 한 형법 개정안이 국회 문턱을 넘었습니다.오늘 본회…
https://news.sbs.co.kr/news/endPage.do?news_id=N1008028901
@thx상어학산타 
아니 왜 점점 늘어나나요???
田所浩二さんの人権を返せ！
The obsession with 'overdiagnosis' is a much bigger problem than overdiagnosis itself
The discourse around the 'problem' of 'overdiagnosis' doesn't hold up at all when you consider the actual evidence. Pushing it anyway risks harming so many
Archネットブックちゃんは今日も元気です、ってタイプしようとしたらシャツが前後逆なのに気づきました #USBおじさん
Audio Player (f.k.a. Decibels) is out!
Thanks to everyone who contributed, and it's so nice to see the first TypeScript app in GNOME Core! Nice times ahead!
annoying freebsd ipv6 thing:
# ifconfig lo1 create inet6 auto_linklocal -ifdisabled 2001:db8:100::1/128 up
# ping 2001:db8:200::2
(ping uses 2001:db8:100::1 as source address)
# ifconfig lo2 create inet6 auto_linklocal -ifdisabled no_prefer_iface 2001:db8:200::1/128 up
# ping 2001:db8:200::2
(ping now uses 2001:db8:200:1 as source address)
my expectation was that no_prefer_iface would prevent it from choosing addresses on lo2 as source address automatically, but apparently this does not prevent choosing the longest match for the destination :-(
there's 'prefer_source', which seems to be an address-specific attribute, but i don't think i want that since i have both GUA and ULA addresses configured.
there's also 'anycast', but apparently if you configure an address as anycast, you can't bind to it, which seems a bit useless.
#GNOME 48 should release in a few hours, so here is my look at this major new version.
In there, there's HDR, global shortcuts, triple buffering, a new audio player, a new font, digital wellbeing features, and a lot more!
Ebay hat ihre AGB und Datenschutz-Infos aktualisiert. Daten aller Nutzer*innen sollen für KI-Training verwendet werden, solange sie nicht vom Opt-Out Gebrauch machen.
Habe meine Zweifel, ob die Berufung auf berechtigte Interessen zu diesen Zwecken unter der DSGVO rechtmäßig ist. Betroffene sollten bis 21. April widersprechen oder ihre Daten löschen, wenn sie diese nicht im Trainingsset haben möchten:
https://accountsettings.ebay.de/ai-preferences
https://www.ebay.de/help/account/changing-account-settings/deleting-data?id=5365
Is there any other way to keep a FreeBSD workstation free from Poettering-software than building your own packages? Stuff like Pulseaudio seems to find always it’s way to sneak on my computer. So I went to build my own packages with Poudriere. This works well, but is also quite time consuming. Would appreciate any suggestions.
@thx@mustard.blog상어학산타 아무리 선해를 하려고 해도, 정상은 아닌 것 같아요. 뭔가 사달이 나긴 난 것 같습니다.
まだ肌寒いのでカーディガンコパっ
저는 AI에게 감사 인사를 하는 데에도 돈이 든다는 걸 깨달아버려서, 이제는 감사도 표하지 않는 삭막한 인간이 되고야 말았습니다.
뭔가 사단이 생겼다고 생각할 수밖에 없는 행태아닌가?
헌재야... 이거 뭐야???
會爛的感覺…
Making a multiplayer action game in Haskell
https://gitlab.com/-/snippets/4817016
Discussions: https://discu.eu/q/https://gitlab.com/-/snippets/4817016
Bluesky @gssb.bsky.socialGS Pixelfed
@gssb@pixelfed.socialGS Mastodon 
@gssbGS
The Rust crate Askama (used to make jinja rendering) 0.13.0-pre.0 version just got released. Rinja and Askama are now one again.
More explanations on this blog post.
i should do a blog about what the crystalizer plugin in audacious does someday. hint: basically everyone is wrong about how it works
Welcome to today's thread - #CrossChannelRail Day 4 - 20 March - Barcelona - Montpellier - Nîmes - Avignon - Marseille
On the agenda today:
Stations
Montpellier St Roch
Montpellier Sud de France
Nîmes Pont du Gard
Avignon Centre
Avignon TGV
Today's routes on the routes map: https://umap.openstreetmap.fr/en/map/crosschannelrail-2025_1148540#8/42.672/3.766
저는 AI에게 감사 인사를 하는 데에도 돈이 든다는 걸 깨달아버려서, 이제는 감사도 표하지 않는 삭막한 인간이 되고야 말았습니다.
Got an interesting question today about #Fedify's outgoing #queue design!
Some users noticed we create separate queue messages for each recipient inbox rather than queuing a single message and handling the splitting later. There's a good reason for this approach.
In the #fediverse, server response times vary dramatically—some respond quickly, others slowly, and some might be temporarily down. If we processed deliveries in a single task, the entire batch would be held up by the slowest server in the group.
By creating individual queue items for each recipient:
It's a classic trade-off: we generate more queue messages, but gain better resilience and user experience in return.
This is particularly important in federated networks where server behavior is unpredictable and outside our control. We'd rather optimize for making sure your posts reach their destinations as quickly as possible!
What other aspects of Fedify's design would you like to hear about? Let us know!
Coming soon in #Fedify 1.5.0: Smart fan-out for efficient activity delivery!
After getting feedback about our queue design, we're excited to introduce a significant improvement for accounts with large follower counts.
As we discussed in our previous post, Fedify currently creates separate queue messages for each recipient. While this approach offers excellent reliability and individual retry capabilities, it causes performance issues when sending activities to thousands of followers.
Our solution? A new two-stage “fan-out” approach:
+ 
Stefano Marinelli
Marcus Lundblad
NixOS
Justice! Yay!
amos
Juntai Park
洪 民憙 (Hong Minhee) 
Context.sendActivity(), we'll now enqueue just one consolidated message containing your activity payload and recipient listThe benefits are substantial:
Context.sendActivity() returns almost instantly, even for massive follower countsFor developers with specific needs, we're adding a fanout option with three settings:
"auto" (default): Uses fanout for large recipient lists, direct delivery for small ones"skip": Bypasses fanout when you need different payload per recipient"force": Always uses fanout even with few recipients// Example with custom fanout setting
await ctx.sendActivity(
{ identifier: "alice" },
recipients,
activity,
{ fanout: "skip" } // Directly enqueues individual messages
);
This change represents months of performance testing and should make Fedify work beautifully even for extremely popular accounts!
For more details, check out our docs.
What other #performance optimizations would you like to see in future Fedify releases?
I think this is the best writeup I've seen on the (lack of) evidence for the efficacy of nasal sprays as a "layer of protection" against #COVID / #COVID19
There is a lot of misinformation out there about nasal sprays preventing COVID-19. Unfortunately, there are no convincing studies showing that nasal sprays prevent COVID-19. The published studies investigating whether or not nasal sprays prevent COVID-19 each have major issues, which I will detail here.
I have a PhD in biochemistry and one of my PhD projects was on COVID-19. The main takeaway of this post is that there is no sound evidence that nasal sprays prevent COVID-19. Thus, nasal sprays should not be used for COVID-19 prevention in place of effective measures such as high-quality well-fitting respirators, ventilation and air purification.
Got an interesting question today about #Fedify's outgoing #queue design!
Some users noticed we create separate queue messages for each recipient inbox rather than queuing a single message and handling the splitting later. There's a good reason for this approach.
In the #fediverse, server response times vary dramatically—some respond quickly, others slowly, and some might be temporarily down. If we processed deliveries in a single task, the entire batch would be held up by the slowest server in the group.
By creating individual queue items for each recipient:
It's a classic trade-off: we generate more queue messages, but gain better resilience and user experience in return.
This is particularly important in federated networks where server behavior is unpredictable and outside our control. We'd rather optimize for making sure your posts reach their destinations as quickly as possible!
What other aspects of Fedify's design would you like to hear about? Let us know!
Coming soon in #Fedify 1.5.0: Smart fan-out for efficient activity delivery!
After getting feedback about our queue design, we're excited to introduce a significant improvement for accounts with large follower counts.
As we discussed in our previous post, Fedify currently creates separate queue messages for each recipient. While this approach offers excellent reliability and individual retry capabilities, it causes performance issues when sending activities to thousands of followers.
Our solution? A new two-stage “fan-out” approach:
洪 民憙 (Hong Minhee)Context.sendActivity(), we'll now enqueue just one consolidated message containing your activity payload and recipient listThe benefits are substantial:
Context.sendActivity() returns almost instantly, even for massive follower countsFor developers with specific needs, we're adding a fanout option with three settings:
"auto" (default): Uses fanout for large recipient lists, direct delivery for small ones"skip": Bypasses fanout when you need different payload per recipient"force": Always uses fanout even with few recipients// Example with custom fanout setting
await ctx.sendActivity(
{ identifier: "alice" },
recipients,
activity,
{ fanout: "skip" } // Directly enqueues individual messages
);
This change represents months of performance testing and should make Fedify work beautifully even for extremely popular accounts!
For more details, check out our docs.
What other #performance optimizations would you like to see in future Fedify releases?
開始閱讀這一本書：《德語是一座原始森林》
I have a dream. #europe #unitedstatesofeurope
すごいねた
おはよう
めがさめた
Immer dran denken: Einen Feiertag zu streichen ist eine direkte Lohnkürzung.
(Original title: Streichung eines Feiertags könnte Milliarden Euro bringen)
AI 的商业化确实是一个非常大的问题，现在对于大型企业盈利模式基本上是靠强劲的模型持续领先卖 ToB 服务，ToC 这一块跑通的没有几家，少数能做起来的只是在做一些垂类工具，没什么门槛，收益也有限。从港股今天的情况来看，南水对国内公司在这方面投入过多的资金还是比较敏感的，不过也确实要感慨北美的公司在赚钱上确实也要更高效，国内反而加更多的班效率更低...
#MULCIA: 2-3 PhD/Postdoc positions in mathematical logic, Würzburg (Germany). https://tinyurl.com/27783c2l #PhD #Logic #Math
Is there any other way to keep a FreeBSD workstation free from Poettering-software than building your own packages? Stuff like Pulseaudio seems to find always it’s way to sneak on my computer. So I went to build my own packages with Poudriere. This works well, but is also quite time consuming. Would appreciate any suggestions.
Lifetime free updates actually means lifetime free updates.
ジェットコースターが…！（そして迫撃砲をかましまくる
Got an interesting question today about #Fedify's outgoing #queue design!
Some users noticed we create separate queue messages for each recipient inbox rather than queuing a single message and handling the splitting later. There's a good reason for this approach.
In the #fediverse, server response times vary dramatically—some respond quickly, others slowly, and some might be temporarily down. If we processed deliveries in a single task, the entire batch would be held up by the slowest server in the group.
By creating individual queue items for each recipient:
It's a classic trade-off: we generate more queue messages, but gain better resilience and user experience in return.
This is particularly important in federated networks where server behavior is unpredictable and outside our control. We'd rather optimize for making sure your posts reach their destinations as quickly as possible!
What other aspects of Fedify's design would you like to hear about? Let us know!
しかもv1ユーザーは無料
RE: https://misskey.io/notes/a5kk5v3f3z0p06zr
わーー
コニーアイランド遊園地が破壊されていく様子にショックを受けるシンディ（ディビジョン2話は突然に）
헌법재판소가 12·3 비상계엄에 가담한 의혹으로 탄핵소추된 한덕수 국무총리 탄핵 사건의 선고기일을 오는 24일 오전 10시로 지정했습니다.
[속보] 헌재, 한덕수 탄핵사건 24일 오전 10시 선...
Quincy ⁂
しゅいろ
