Is it because of the type of work you're trying to get Fable to do or because Fable knows you're a medical physicist? I.e. if your CLAUDE.md says "I'm a medical physicist" but you're asking it to write a command line tool to list github prs or something, does it balk at that?
I was talking to a friend recently about those Tupperware parties (she did a different brand). I think maybe those were better for society than modern ads. People had to get out of their comfort zones and talk face to face. It was all so personal, compared to what we have now.
This influencer nonsense described here is gross and fake compared to that. What would a genuine, face-to-face advertising campaign look like nowadays?
Here, I’ll define it: nationalist is the idea that nations, including the concept of borders, are a great way to organize people to prevent situations like what Sweden is going through.
What solution do you have in mind? And is it guaranteed to work? Because deporting the people who are massively over represented in crime is guaranteed to work.
I have a a theory that LLMs are going to be the death knell of big SaaS. It's so much harder to build and maintain an massive SaaS that does 80% of what 80% of your customers want, than it is to build something small and simple that does 100% of what one customer wants.
AIPAC is mostly funded by Zionist Jewish billionaires[1]. Characterizing this as "Americans, doing the American political thing" conflates normal Americans with Zionist Jewish billionaires. They are not the same thing.
I don't mind paying for a subscription, as long as I'm not also paying for the privilege of being locked in to a specific vendor. If I pay for a subscription and then your prices quadruple or something, what are my options? Can I self-host a relay? Do I lose features if I do so?
Hyperbole is a useful tool in presenting an argument. One of its greatest advantages is that someone who is uninterested in engaging in good faith will often nitpick the hyperbole instead of addressing the fundamental argument. This is a great way to tell when someone can't address the argument, or is paid to have a conflicting opinion.
And unfortunately this is our reality--Israel is spending hundreds of millions of dollars on various forms of PR, which means using hyperbole to describe Israel actually helps you filter out the people who are paid to make Israel look good from the people who are paying Israel via taxes.
Comments like yours are what people mean when they say they're switching between attacking and playing the victim. Your response to perceived "antisemitism" is to play the victim, and it is only helping prove what you're complaining about.
The correct way to address this perception is to prove that the attacker (Israel, who is attacking everyone around them) and Jews are not the same. Israel is working very hard to make them seem the same. THAT is what you should be complaining about. Non-Zionist Jews should paint yourselves as the victims of Israel, not the victims of anti-semites. That is the single biggest contribution you can make to this conversation.
I was building a cli tool that showed a graph of git commits, kind of like git log --graph, and deepseek v4 simply could not figure out a specific ui quirk where things weren't lining up correctly. I spent like $0.10 and 30 minutes trying to figure it out on deepseek.
Then I had deepseek summarize the bug, gave it to Opus, and it solved it in $1.12 and five minutes.
This is cool! I assume a lot of this was Fable orchestrating sub-agents with cheaper models, right? Something I noticed with Fable is that if it spun off three sub-agents in the cloud version of claude code, and then hit the 5-hour usage limit, all the work of those sub-agents would be lost (!). Did you run into the same thing?
One time though, I hit the limit when not running a sub-agent, and the agent resumed after the limit expired. Weird.