Tootfinder

No sign Russia is preparing to restart Zaporizhzhia nuclear plant, Reuters reports: https://benborges.xyz/2025/05/29/no-sign-russia-is-preparing.html

I want to push back on the idea in the world of tech work that a PIP (performance improvement plan) is about getting rid of someone, that they're not intended to be survivable.
This is completely false. (I'm sure there's instances of it, of course, but the mode and vast majority are, in fact about performance improvement. Sometimes they're shadow layoffs, but that is cruel callous behavior that not everyone will exhibit.)
Now _most people do not survive the PIP process_. This is to be expected: if someone is in fact not performing, and more gentle remedies haven't worked, it's not looking good.
But here's where I get a bit spicy: most performance problems are constitutional problems with management and management style, not individual performance problems. However, since managers are as a class 'in power' somewhat, the individual contributor takes the fall for this structurally.
The intent of a PIP is not to get rid of people. It's to right performance.
However, as a system, PIPs do largely get rid of people who are constitutionally misaligned with management. Even when it's a management problem (and it usually is)

Nordic countries are serious about emergency preparations. We should be too.
https://apnews.com/video/how-norwegians-are-preparing-for-emergencies-1fb35749bdfa4905be8caef5feabb5d2

How Norwegians are preparing for emergencies
Local authorities in Norway have dusted off Cold War-era bomb shelters, installed a new satellite communications system and are working with the military on plans to help a deployment of Western forces in case of conflict.

The deadline for FProper (SIGPLAN Workshop on Functional Programming for Productivity and Performance) full-paper abstracts has been extended to June 14th. (Which is also when the papers themselves are due.)
Send us your paper!
https://conf.researchr.org/home/icfp-s

Talk about an unholy alliance. #react and #windows. 💀
https://…

Windows 11 Start Menu Revealed as Resource-Heavy React Native App, Sparks Performance Concerns
Recent observations from users on the social platform X have uncovered performance issues tied to the Windows 11 Start Menu, revealing that the component

Berlin-based Buena, which is building AI-powered software to automate property management tasks, raised $58M, including a Series A led by GV (Cate Lawrence/Tech.eu)
https://tech.eu/2025/07/23/gv-and-20vc

GV and 20VC back Buena’s vision to reinvent property management through AI-powered rollups
Berlin proptech Buena has raised a total of $58 million in funding, including a Series A funding round led by GV, with participation from 20VC, Stride, and Capnamic.  Property management is incredibly fragmented, with over 30,000 firms in Germany employing an average of just five people each… Meanwhile, the industry is notorious for talent shortage and record-low customer NPS scores, with 96.3 per cent of German property managers.  Buena is changing the industry in two ways: first, by bui…

Ich bin kein #Prepper, jedenfalls bisher nicht. Aber diesen Ansatz zu sanftem #Preppen finde ich in Zeiten der tiefen Umbrüche bedenkenswert:
#SanftesPreppen
"Es ist Preppen, aber n…

Gentle Prepperism: A Roundup of Resources + Tips for Navigating the Next Era
In a world that feels increasingly uncertain, it’s easy to ping-pong between panic and paralysis. But there’s another way: Gentle Prepperism.

Watching the first Question Period of the 45th Parliament of Canada!
#CanPoli #CdnPoli

Learning to Align Human Code Preferences
Xin Yin, Chao Ni, Liushan Chen, Xiaohu Yang
https://arxiv.org/abs/2507.20109 https://arxiv.org/pdf/2507.20109

Learning to Align Human Code Preferences
Large Language Models (LLMs) have demonstrated remarkable potential in automating software development tasks. While recent advances leverage Supervised Fine-Tuning (SFT) and Direct Preference Optimization (DPO) to align models with human preferences, the optimal training strategy remains unclear across diverse code preference scenarios. This paper systematically investigates the roles of SFT and DPO in aligning LLMs with different code preferences. Through both theoretical analysis and empirica…