@tjparker You can take the operating model from 2023 and compress quarterly to monthly *while* code output goes up 3-6 times for deterministic software.
For probabilistic software, answer is context dependent.
I think the older you get, and the more you build a life you love, the more you realize how much of a scam travel is
There are genuinely very few things that are worth flying 10+ hours, being interrogated, leaving your routine, your loved ones, and messing up your sleep schedule
I think if you’re young and you don’t really have anything tethering you to one place, it doesn’t matter too much where you go and when. It will not cost you much
At this stage in my life, I’ve built a life/home that is more beautiful, more unique than anything else in the world. It’s a sight I will never get tired of
4/ I still really like the concept.
A lot of transactional and task-based queries do feel like they naturally belong in Messages.
Maybe the difference is that I already know texting them isn’t texting a person. I’m also tired of repeating the same context across different models and surfaces.
Tried Instinct. A few reactions:
1/ I used it to lower my Xfinity bill. It didn’t execute the task, but guided me through the steps instead.
Helpful, but the experience often felt like chatting with ChatGPT: long responses and noticeable lag.
3/ At the same time, if ChatGPT or Claude were textable, I’d absolutely use them.
Maybe the difference is that I already know texting them isn’t texting a person. I’m also tired of repeating the same context across different models and surfaces.
Can’t wait until we can say “build me a $1M company, make no mistakes.”
And 3 months later we actually have $1M in our bank account.
The time is coming.