You know it's a good idea when ChatGPT says it is.

Every time I see nondeterministic code it has the same effect on me

If calibrating my speakers to my home office won't increase my productivity, I don't know what will

🚀 New episode of Ship It! 📌 Tars all the way down 🤩 featuring @jonjonsonjr ⚡️ with @rothgar & @autumn808 🗃️ #tar #devops #linux 🎧

context.Context rules everything around me

One thing I’ve learned working on vulnerability workflow tooling at Chainguard: make sure to keep mock data nearby. Vulnerabilities disappear so fast that it wastes time trying to find them in arbitrary build artifacts.

Thanks, @danluhring, and for everyone else, here’s a link to download @arcinternet , the browser he was telling me about! https://t.co/lXLUqt5TYt

Switching from dark mode to light mode in my editor today, AMA

This thing in the sky has totally eclipsed what I was working on this afternoon

I'm finally ready to break my silence. https://t.co/WX1KIM3cdQ

Thankful for the time to focus on my family, and excited now to strap back into the @chainguard_dev rocket ship! 🚀

Cramming AI into your product really cheapens the experience. Understand your users and use AI to get out of their way.

New release! ❄️ Freeze: a CLI tool to generate terminal screenshots of code and ANSI output. Run a single command to generate PNGs and SVGs of terminals with program output or syntax-highlighted code. Use it in READMEs to show off code and CLI output. https://t.co/0jwMhX8F6u

Me to ChatGPT:

Newly discovered vuln in Apple M-series chips lets attackers extract secret keys from Macs. "The flaw—a side channel allowing end-to-end key extractions when Apple chips run...widely used cryptographic protocols—can’t be patched"

What is it called when you run out of ice so you combine espresso with whipped cream and some milk? I just made one of those

I really wish Go had a "zero" keyword for non-pointer return values. Being forced to switch to pointer return types just to have the elegance of a type-agnostic keyword (`nil`) throughout your function body is an awkward tradeoff

Maybe it's the parental leave talking but lexical analysis is really fun

@GitHubCopilot Or perhaps it could just visually annotate when a suggestion it's making has passed or failed validation, to inform your decision of whether to accept it and continue

It'd be cool if @GitHubCopilot (including chat) could run your tests to validate its suggestions first 🤔 Maybe as a toggleable feature. I find sometimes I like the raw brainstorming, but sometimes I'd rather do it myself unless it has a provably good solution