i'm seeing the same thing
Posts by Vincent Pickering
#Design used to be filled with conversation about fixing human problems in humanistic ways and listening to peoples issues and needs so that you could achieve human focused outcomes through collaboration.
Now: "Wanna buy a bridge? It comes with a bottle of snake oil for a limited time only!"
Spacing isn't consistent either on the UI. Buttons are Jackson Pollocked all over the shop.
The Green dragon Inn where we had a cup of Amber Ale
The mill and working water wheel
Finished up at the Green Dragon for a drink. The pub is beautiful inside with a real thatched roof
Fire burning in the main sitting area
Bunkbeds for little hobbits
Study with desk, chair and seed collection
Kitchen stove, wonder whats cooking
We got to go inside one of the hobbit holes. They are quite big inside (scaled up slightly for humans). The detail is insane.
Samwise dads hobbit hole!
Bilbos pipe and book on a bench outside his home
Bagend. Bilbos home!
Close up of Bagend front door
Shire notice board
Hobbit hole with a red door. All the veg is real (including that pumpkin)
Signpost to east and west farthing
This hobbit hole was smoking fish with a real fire
All the different hobbit holes are unique and themed around a specific family with clues as to which family they are
The hobbiton sign
A hobbit hole with a blue door, chair and table outside
A hobbit hole with a yellow door and rocking chair outside
Hobbiton was great yesterday, the Shire is amazing and so detailed. Some highlights!
Hobbiton Movie Set leaflet
Time for Hobbitses
Raglan Beach. Empty!
Beach was pretty empty today. Son was boogie boarding all day in amongst the surfers ๐๐๐โโ๏ธ๐โโ๏ธ
Ah Like The Ladykillers? I think the Coens did a remake a while back? Not sure on the quality though ๐ค
Lock Stock and Two Smoking Barrels maybe? Snatch probably also counts!
Werner Herzog's Barbie Project
Ecocide *and* epistemicide wrapped up in one. What's not to like?
oh wow, so SHOCKING, "AI" doesn't exist actually (Apple paper proves LLMs can't math or think or understand a problem)
nitter.poast.org/heynavtoor/s...
Screenshot of the article on Inc. titled People Who Love Corporate BS Are Bad at Their Jobs, New Cornell Research Confirms. The psychologist behind a new study confirming the link between dumb jargon and dumb decisions explains how to fight corporate BS.
People Who Love Corporate BS Are Bad at Their Jobs, New Cornell Research Confirms
Ever had to work with someone who spewed corporate bullshit all day long, yup, that person. Science proved what we all knew all along: they are bad at practical decision making and analytical thinking.
A big punnet of strawberries
Strawberry?
reminder that anthropic ran (and is still running) an ENTIRE AD CAMPAIGN around "Claude code is written with claude code" and after the source was leaked that has got to be the funniest self-own in the history of advertising because OH BOY IT SHOWS. it's hard to get across in microblogging format just how big of a dumpster fire this thing is, because what it "looks like" is "everything is done a dozen times in a dozen different ways, and everything is just sort of jammed in anywhere. to the degree there is any kind of coherent structure like 'tools' and 'agents' and whatnot, it's entirely undercut by how the entire rest of the code might have written in some special condition that completely changes how any such thing might work." I have read a lot of unrefined, straight from the LLM code, and Claude code is a masterclass in exactly what you get when you do that - an incomprehensible mess.
and what if i told you that if it passes a page range to its pdf reader, it first extracts those pages to separate images and then calls this function in a loop on each of the pages. so you have the privilege of compressing n_pages images n_pages * 13 times. this function is used 13 times: in the file reader, in the mcp result handler, in the bash tool, and in the clipboard handler - each of which has their entire own surrounding image handling routines that are each hundreds of lines of similar but still very different fallback code to do exactly the same thing. so that's where all the five hundred thousand lines come from - fallback conditions and then more fallback conditions to compensate for the variable output of all the other fallback conditions. thirteen butts pooping, back and forth, forever.
holy shit there's another entire fallback tree before this one, that's actually an astounding twenty two times it's possible to compress an image across nine independent conditional legs of code in a single api call. i can't even screenshot this, the spaghetti is too powerful
alrighty so that's one of 43 tools read, the tools directory being 38494 source lines out of 390592 source lines, 513221 total lines. I need to go to bed. This is the most fabulously, flamboyantly bad code i have ever encountered. Worth noting I was reading the file reading tool because i thought it would be the simplest possible thing one could do because it basically shouldn't be doing anything except preparing and sending strings or bytes to the backend. I expected to get some sense of "ok what is the format of the data as it's passed around within the program, surely text strings are a basic unit of currency. No dice. Fewer than no dice. Negative dice somehow.
This is amazing. A dev on Mastodon is going through the Claude code leak and appraising the code.
To say the code is bad would be an understatement and shows just how much of a grift all this AI BS really is.
neuromatch.social/@jonny
Some highlights attached!
guys this is so nice :') I love that we're all online watching a thing that's good instead of bad :')
Amaze Amaze Amaze #artemis
Supernote are good and everything pretty much can be easily replaced if broken or updated. You do pay a little extra for it though
supernote.com
Forever and always, trans rights are human rights.
When I meet designers with no empathy, I struggle.
Empathy is the foundation of design, it permeates everything we do. Without it you're designing with your eyes and ears closed.
A GitHub setting labelled โAllow GitHub to use my data for AI model training", set to disabled
Quick reminder, especially if youโre a freelancer or developer using Free/Pro/Pro+ plans for client work: Opt out of GitHub using your data for AI model training before April 24 (seriously, wtf that this isnโt opt-in!).
github.com/settings/cop...
Absolutely, it feels like there is a great cost/benefit waiting to be evidenced!