Before AI: ideas are cheap, execution is everything.
After AI: execution is cheap, so ideas should matter more. But a good idea gets cloned in a weekend โ so ideas are cheap too.
Taste? Copied the moment it's visible. Speed? Everyone ships fast now. So what's the moat?
@unclebobmartin I know you said you don't review diffs any more but do you still read the code? Any concern that if one day you have to read the code yourself you don't understand it any more?
I just finally figured out why I keep seeing weird "explain u-factor of glass" in my Claude.
One of the earliest questions I asked Claude 2 years ago was "explain u factor of glass door". Somehow it went to set my name to "explain u-factor of glass door"๐คฃ
If open models like Kimi are as good as the best closed models like OpenAI and Anthropic's and Amazon and Microsoft can serve them w/o paying royalties, any reason for any one to keep using those expensive closed models?
"Hey Siri" never works when you need it the most. For example, a clock alarm rings while driving with iPhone in my pocket, "Hey Siri stop the alarm" doesn't trigger it, as if iPhone doesn't think it is in a state to receive "Hey Siri".
Both @GeminiApp and @ChatGPTapp failed to generate a runnable web app on iOS.
Kudos to @claudeai but its 3d understanding needs improvement. After 3 iterations of wrong modeling I had to describe the shape in words for it to finally get it.
My fellow Democrats, I have decided not to accept the nomination and to focus all my energies on my duties as President for the remainder of my term. My very first decision as the party nominee in 2020 was to pick Kamala Harris as my Vice President. And itโs been the best decision Iโve made. Today I want to offer my full support and endorsement for Kamala to be the nominee of our party this year. Democrats โ itโs time to come together and beat Trump. Letโs do this.