My experience is quite a bit different than what the author is citing. I live just outside of Minneapolis, and have a bunch of friends who live in the city. I have never heard anyone express concern about what ICE is doing. The biggest worries are around commute disruptions due to protesters. It might be because this is a self-selected circle of white collar types, but I can say the report in this story is not universal.
As someone who lives in St Paul, I'll say the blog post sounds like my experience here.
As someone with a white collar software job, the description of people fearing for their children at school or not leaving their house is playing out at my workplace.
Cover of today's StarTribune (Greenland article at bottom), from the top down:
• Our community is in crisis
• Rumors, anxiety spread beyond the Twin Cities
• Citizens sue, seek ruling to rein in agents' tactics
• DHS' Minnesota presence dwarfs largest metro police forces
• Review: No widespread illegal voting by migrants
• Trump cheers protesters in Iran, but not Minnesota
"I have never heard anyone express concern about what ICE is doing."
"heard" and "express concern" commonly do a lot of heavy lifting in anecdata:
• Include or exclude social media comments from friends?
• Only include people who have spoken to you directly?
• Speak often or infrequently to others?
• Include or exclude content that you have seen that you trust?
• Are part of a large/small circle of family/friends?
• Concern or only experience causing concern?
• General opinion or on specific available information?
• etc., etc.
Without being more specific, "heard" and "express concern" don't mean much, and with specific qualification, are easily manipulated.
I also live just outside of Minneapolis. I have white-appearance family members whose heritage traces literally back to the Mayflower that have been harrassed while working from home, and literally dozens of others expressing concern due to those in their circle being harassed, and many others expressing concern due to viewing media that they trust.
In thinking just now, it's hard to think of anyone not expressing concern, on "both sides", partly because I am hearing expressions on the topic from a tremendous number of persons.
Have you considered using the on-device Apple foundation models as a fallback/alternative to the remote APIs? There is of course Apple's SpeechTranscriber (which is likely faster and more power-efficient than Whisper), and also the permissiveContentTransformations guardrail which is specifically designed for your use case.
Unfortunately, IDE integration like this tends to be very prefill intensive (more math than memory). That puts Apple Silicon at a disadvantage without the feature that we’re talking about. Presumably the upcoming M5 will also have dedicated matmul acceleration in the GPU. This could potentially change everything in favor of local AI, particularly on mobile devices like laptops.
Cline has a new "compact" prompt enabled for their LM Studio integration which greatly alleviates the long system prompt prefill problem, especially for Macs which suffer from low compute (though it disables MCP server usage, presumably the lost part of the prompt is what made that work well).
It seems to work better for me when I tested it and Cline's supposedly adding it to the Ollama integration. I suspect that type of alternate local configuration will proliferate into the adjacent projects like Roo, Kilo, Continue, etc.
Apple adding hardware to speed it up will be even better, the next time I buy a new computer.
