sundry stuff: https://nitter.cf/t.co/2iQ0IcO2PW
Paris, France
Joined September 2008
- Tweets5.3K
- Following447
- Followers259
- Likes3.3K
Ulysse Sabbagh retweeted
Raycast turning into a huge, sluggish, awful piece of garbage is all of the vindication I needed for being too old and crotchety to move away from Alfred all those years ago.
Ulysse Sabbagh retweeted
One pattern I find useful for working with LLMs is a nice long ramble session. Sometimes the LLM needs more bits to understand what you're trying to achieve, but you're too lazy to type them. In these cases I like to lean back, switch to /voice and just ramble for like 10 minutes, total mess, anything goes, full stream of consciousness. Sometimes I declare it up top, something like "switching to speech recognition sorry for any typos...". Sometimes I turn it into a small interview of a few turns. But I find that the LLMs are somehow very good at reconstructing long incoherent rambles and often their echo of your own tangle of thoughts comes out quite a bit cleaner than what you started with. The result is that you improve the mind meld and have to correct things less from that point on.
I have been reading Om Malik since I was a teenager and was always struck by his ever-original insight and smooth way with words and photography. RIP @om and thank you for your teachings
Even if they didn't have "the model" that they want to capture the market would still be true
I want you to notice something about Anthropic.
They seem to be dropping all these new products in their suite super fast.
you are asking why?
because I am 100% convinced they have created the end game AI model.
whatever they call it, whether that is "Opus 6" etc.
They have the model, that they know once they release it.
People will be able to build real apps, in days.
they want to grab as much of the market as possible before they do.
Must be bait: Gmail, Instagram, Slack, Google Earth (many other Google products) started as side projects. The culture creates great products. In a way it's the ultimate strategy. Good people don't need to be herded towards corporate, abstract goals
Feels like it's written out with an AI, which makes it tasteless. What about other cultures than Europe to factor in discussion about taste, like Japan? Maybe read In Praise of Shadows. What about the anthropological needs for differentiation, which implies that the discriminator will survive?
One year since vibe coding was coined!
There's a new kind of coding I call "vibe coding", where you fully give in to the vibes, embrace exponentials, and forget that the code even exists. It's possible because the LLMs (e.g. Cursor Composer w Sonnet) are getting too good. Also I just talk to Composer with SuperWhisper so I barely even touch the keyboard. I ask for the dumbest things like "decrease the padding on the sidebar by half" because I'm too lazy to find it. I "Accept All" always, I don't read the diffs anymore. When I get error messages I just copy paste them in with no comment, usually that fixes it. The code grows beyond my usual comprehension, I'd have to really read through it for a while. Sometimes the LLMs can't fix a bug so I just work around it or ask for random changes until it goes away. It's not too bad for throwaway weekend projects, but still quite amusing. I'm building a project or webapp, but it's not really coding - I just see stuff, say stuff, run stuff, and copy paste stuff, and it mostly works.
Ulysse Sabbagh retweeted
Another entry in “everything had to be invented”, some of the UX improvements from early word processor WordPerfect:
Losing status could be a proxy for getting kicked out of the group, which is not good
This quoted post is unavailable.
The thing is that salt with added iodine often contains anti-caking agents. The taste is altered. Many chefs will use kosher salt or sea salt. You can get your iodine needs met with seafood. Use additive-free salt
The secret is that the better you are at backend stuff, the more contempt you have for frontend, as you should
Ulysse Sabbagh retweeted
the fact that important numbers like e and pi are irrational feels like proof our number system is fucked up. we made some mistake somewhere and that's why everything is shit
Ulysse Sabbagh retweeted
RHLF (training for answering human questions perfectly) is antithetical to creativity (answering human questions imperfectly)
Bring it back
people forget how influential Palm webOS was. lots of modern iOS uses similar patterns.
palm was basically the OpenAI of their time. such talent density that Steve Jobs had to threaten them to stop poaching.
• horizontal app card multitasking w/ flick up to quit
• edge swipe gesture nav for back nav and quick app switching
• swipe away banner notifications
• swipe down system tray (like Control Center)
• "Just type..." global search bar
• threaded messaging
• magnetic charging puck called Touchstone (like MagSafe)
Anyone using AI for more things than just work will have felt a new resistance to thinking. Outsourcing thinking to a machine makes the thinking muscle to weaken. I don't think this is surprising
everyone’s sharing this thread on "cognitive debt" as if it was anything but garbage
of course it sounds deep with the "BREAKING," "let that sink in," "AI is making us dumber"
but the paper's methods are sloppy, the experimental design doesn't make sense and it's only goal is to fearmonger while not saying anything useful
wait, academics have interests in telling people that AI bad?! what?!
shocked_pickachu_face.jpg