Patrick H. Lauke

@patrick_h_lauke
1.5K Followers
620 Following
18.1K Posts

WCAG trash panda, web evangelist, ronin, sour kraut. uninvited expert. vague, but exciting. luddite. antifascist. european. #toryScum #a11y #pixelart #pixelfonts #photography #cisgender #TrumpIsAFascistCunt #FuckElon #JailNetanyahu #monocultureWarrior he/him

[Banner image: the top of a Fisher Price baby laptop, with colourful letters of the alphabet, a cartoon dog, and various cutesy images of a pig, duck, house, sun, lion, wagon]

personal site (mostly pixel fonts)https://www.splintered.co.uk
githubhttps://github.com/patrickhlauke
codepenhttps://codepen.io/patrickhlauke
flickrhttps://www.flickr.com/photos/redux
WTF does an AI agent need a birth certificate? WTF is happening? https://dnsid.ai
DNSid™ - The Birth Certificate for AI Agents

DNSid™ is the protocol for verifiable AI agent identity.

Again, if what we're doing now is forcing all the crooks and rapists in Congress to resign one by one as each party tries to maintain or obtain a slim majority in a divided House...

I am so here for that. Let's do it. Please keep exposing each other. Preferably till you are all gone because if you ask me, every one of you is criminally complicit in the rise of fascism and billionaires boiling our kids like soup on a dying planet for five pennies a share.

So I support this. Do more.

shower thoughts: with Shopify and AI/LLM ... should they rebrand to Slopify?

#thoughtLeader

favorite socks 

RE: https://mstdn.social/@NaMi/116442519701376365

Das.
"Und ich habe langsam keine Worte mehr, die auch nur annähernd die Verachtung für diese unmenschliche, asoziale Politik dieser pseudochristlichen Konzernlobbybaggage adäquat zum Ausdruck bringen könnte."
(thread ->)

yes  

Edit: the pic got viral and someone told me the credit should be to this Instagram account https://www.instagram.com/p/DK-RCnsylj8/

I don't have an Instagram to verify myself but someone can correct me if that's wrong, thank you for letting me know!

"A recent 2026 empirical study titled "Beyond Code Snippets: Benchmarking LLMs on Repository-Level Question Answering" (published on arXiv/ResearchGate) explicitly tested LLMs on codebase comprehension. The researchers concluded that high performance often "results from verbatim reproduction of Stack Overflow answers rather than genuine reasoning." " https://www.researchgate.net/publication/403262523_Beyond_Code_Snippets_Benchmarking_LLMs_on_Repository-Level_Question_Answering