Ajit retweeted
⭐ New talk! andymatuschak.org/tat
Coding agents might help us finally break out of two cages: the app model, which traps computing in one-size-fits-all silos; and programming as a specialization, which has crowded out cultures of imagination and domain insight.
Ajit retweeted
Frame of preference
A history of Mac settings, 1984–2004
…と言いつつ前身のNeXTも入ってる
aresluna.org/frame-of-prefer…
Ajit retweeted
whoah seeing these industrial design style copic marker sketches really drives home the skeumorphic approach of ui design of that era
Ajit retweeted
What’s a functional prototype you’re proud of building?
Me: I once built a prototype of a clinical workflow using vue.js and populated realistic dummy data in Firebase. You could switch different profiles to have scenarios of edge cases and high data fidelity. – at San Francisco, CA
Ajit retweeted
Read about the full project here:
theturnsignalblog.com/the-su…
Ajit retweeted
I'm telling you: there is a "Claude Code Parent" demographic there. It's not super visible, but I know more than one dad who is reclaiming parent time right now :)
Ajit retweeted
Replying to @mitchellh @NickGideo
(And we closed the deal. I don't always wear a suit, but you have to know your customer.)
Ajit retweeted
“Sometimes magic is just someone spending more time on something than anyone else might reasonably expect.” ― Teller.
Ajit retweeted
This is, not true.
The machines in Apple's model studio are very different from the production machines used by Catcher and Foxconn for mass-scale production.
Fun Story: The iPhone 6 bending debacle forced Apple to quickly re-evaluate using a 7000 series of aluminum in place of their standard 6061. They had (of course) experimented with this long ago, but cost and finishing concerns - 7075 is harder to anodize, and finishing is the *hardest* thing Apple dos - drove them to use a slightly tweaked version of industry standard 6061.
Catcher and Foxconn both thought this would be a nice opportunity to milk some more margin, so they told Apple they could do 7075, but the cycle time was almost 2x, so the price would go up substantially.
One of the machinists thought this sounded like bullshit, so Apple called the US distributor for the machines that Catcher and Foxconn use, pulled the identical tools they were running in prototype (in new BT30 holders), and ran back to back tests to prove out the cycle time and tool life on the same machines their vendors were using. Until this point, they did all their own prototyping and process development, but it was up to the vendors to use the competitive fight for more orders to figure out how to get the cycle down; Apple never got very specific as to the process - just the quality and price.
Well, lo and behold - 7075 had the same cycle time as 6061. The vendors were bullshitting them to the tune of tens of millions of dollars. Apple showed them the results of their testing and told them to pound sand with the price increase. Ever since, Apple has run a full-scale production lab (separate from Ive's model shop) to validate production methods and guide Apple's design and procurement decisions.
Apple's prototype shops run very high-end, hyper-accurate multi-axis machines that are ideal for rapidly turning concept sketches into machined parts. If you know CNC stuff - they are a big on Hermle and Willemin-Macodel shop, but they have whatever gear they want. Other machine shops at Apple support engineering efforts running similar equipment. The production validation lab however, uses the same gear that is used in Asia, and evaluates precisely what Apple can expect their suppliers to be doing when they move to full-scale production. These are similar, but very different machines than the model/engineering shops run - lower cost, faster, more fussy to set up, but designed to absolutely bang-out lots of parts very very rapidly.
Ajit retweeted
Spotted these while watching "Learn Great Copywriting in 76 Minutes" ft. @harrydry where he talks more about techniques like 1 Mississippi 2 Mississippi, his 3 rules, and a lot more with examples.
Go ahead and watch the video.
youtu.be/TUMjnmfsPeM?si=IjzD…
Ajit retweeted
this is a 1996 guide on how to help someone use a computer. it’s strikingly resonant with ‘how to be a parent’, or really ‘how to help anyone with anything’. a nice example of “the universal within the particular”
Ajit retweeted
No wonder nothing I have made for the web feels alive. I have applied 3-4 of the 12 principles of animation and the interfaces I use every day have applied 1-2: vimeo.com/93206523
Ajit retweeted
Thrilled to share my latest creation: a Telugu-inspired animated wedding invite! 💌🎨 Drawing inspiration from the legendary Bapu's intricate art, I've embarked on a journey to celebrate my culture and my sister's big day!
Ajit retweeted
Replying to @nakhreywaali
Just wait a bit. When you're a fossil, I'm sure some dashing paleontologist hunk will carbon date you.
Ajit retweeted
Design a lot. Look at design a lot. Study what makes a design successful. Be nice to the people you work with. Be nice to yourself. Be nice to everyone. Remember that there is more to life than design. Go outside. Travel. Take care of your health. Get a dog.
Ajit retweeted
Shame Apple didn’t adopt the mallet keyboard from @normalvr normalvr.com/blog/an-open-so…





































