This makes me think about the ‘good old days’ of Mini MSFT. We need a new MiniMSFT. Maybe a MacroMSFT? MicroMSFT? Someone to spill the corporate gossip with insights and conjecture.

https://www.geekwire.com/2026/mary-jo-foley-what-the-heck-is-going-on-with-microsoft-lately/

I blogged again:
Tech says it loves you. Check the org's back log
https://www.codevoid.net/ruminations/2026/03/29/tech-says-it-loves-you.html

This has been percolating in noggin for a while now as I think about the (perceived, or real) incentive change since I started in the industry.

If I don’t see another fucking “Why it matters” in a blog post, email, or article ever again I will very happy.

Also. Faux Punchiness. Because it’s annoying af.

Another step towards (good) local models? Combined with the recent reporting (https://simonwillison.net/2026/Mar/18/llm-in-a-flash/) around “memory mapping good actually” for getting chonky boi models on small devices, it feels like we’re getting closer.

https://arstechnica.com/ai/2026/03/google-says-new-turboquant-compression-can-lower-ai-memory-usage-without-sacrificing-quality/