"disables various anti-features"
look inside
"disables Autofill/Autocomplete in the URL Bar"

​​

score: 98
│​​​​​​​​​​​​​​​​​​​​│
│​​​​​​​​​​​​​​​​​​​​│
│​​​​​​​​​​​​​​​​​​​​│
│​​​​​​​​​​​​​​​​​​​​│
│​​​​​​​​​​​​​​​​​​​​│
│​​​​​​​​​​​​​​​​​​​​│
│​​​​​​​​​​​​​​​​​​​​│
│​​​​​​​​​​​​​​​​​​​​│
│​​​​​​​​​​​​​​​​​​​​│
│​​​​​​​​​​​​​​​​​​​​│

During discourse, do not insist on civility, but good faith.

It should be acceptable for a person who has suffered a grave injustice to shout in rage, and it should not be acceptable for a charlatan to mislead with eloquence.

imo the real solution to the bot flag situation is to implement a new entityType property (which could also replace isCat).

not only does that provide a clear path for people to express themselves, but it could also cover heaps of other entity types beyond bots and cats. dogs! birds! mice! buses!!!

and software could (if it desired) provide way more fun and expression through this. tails, wings, whatever!!! then the bot field can be used for its intended purpose of "this is an automated account and might be spammy, treat accordingly" and everyone's happy. (well a few people might go "grrr no fun/expression allowed" but they can find somewhere else to be miserable.)

I realise this would require at least some work but it just seems way better as a solution to me

• Dune, page 106;

Building a website in 2005: "here are a list of tips to help crawlers index your website"

2025: 'here are a list of tips to help avoid crawlers indexing your website"

problem: there's a leaking tap

FE dev solution: remove the tap and break the sink, nobody will notice

Alt...

🧹 Discord is finally "fixing" memory leaks! /s The client is testing a feature that monitors its own RAM usage and performs an intentional restart to clear resources if it reaches 4 GB while you're AFK. No restarts occur if you're currently in a call.

I partly got into coding because computers were determinate and did exactly what they were instructed to reliably and without variance. If there was a problem, it was because your code was wrong.

But, congrats everyone, because by shoving AI everywhere we’ve made computers indeterminate.

WHY DOES MY CODE USE 12.4 TERABYTES OF VIRTUAL MEMORY

Migrating #Dillo from GitHub to our self-hosted server.

I wrote a post about the current situation with #GitHub and how we ended up self-hosting our own infrastructure to be robust against data loss. We now store all important data (including issues) in git repositories which are replicated accross #Codeberg and #Sourcehut.

https://dillo-browser.org/news/migration-from-github/

A thing that has been in my head for years:

An intern assigned to a project to improve caching in a hot path system declared one day that the problem was that we had been thinking about it all wrong: we didn't want a better cache eviction policy, we wanted a better cache admission policy. He then set about implementing a really simple thing that he only added stuff to the cache after it had been used more than N times in a sliding window.

Cache instantly performed like 50% better. There's a lesson here.

I think my favourite point so far in the progression of AI was when Microsoft launched the new Bing Chat (Sydney) which was really quite horrifically misaligned, manipulative, and frankly just completely evil.

Quick little micro post about it: https://vale.rocks/micros/20250417-1200

#AI #LLM #Bing #Microsoft #ChatGPT #OpenAI