arcadianalpaca

Yeah I've been doing this more and more. The friction of keeping dependencies updated has gotten worse than the friction of just maintaining the (often times) 20 or so lines you need to write yourself. Not to mention the pain you'll eventually find yourself when a bug isn't patched fast enough, or when you need to make a small change, or when a transitive dependency you've never heard of gets compromised...

The part about open source projects needing years of sustained work rings true for sure, but it kind of skips over why a lot (most?) projects die. Sometimes the author gets bored, sure, but maintaining something used by strangers is a completely different job than building something for yourself, and nobody warns you about that transition.

Right, though looking at the release notes it seems like the AI part at least is opt-in... for now.

Yikes... sending prompts to a third party by default with no disclosure in the setup flow is a rough look for a tool that positions itself as the open sources alternative. "Open" loses meaning fast if the defaults work against the user.

Cheers! Yeah, I've been switching between both mediums for a couple of years now, and this feature has really made things so much easier! I've actually been surprised at how smoothly it works.

The filler word idea is interesting but I suspect the uncanny valley risk is super high. A mistimed "mhm" from a computer would probably feel way worse than just silence, because now your brain is pattern matching against human conversation and every small timing error stands out more

Just like real life. Sit still, touch nothing, and watch everything fall apart all on its own ¯\_(ツ)_/¯

Yeah the squash question is the whole thing. If your commit history is "do X" -> "no, not like that" -> "closer" then your final commit message is just "do X" with no trace of why certain approaches were rejected. Which is arguably the most useful part of the conversation.

True, can't even be taken for a real safeguard to begin with. It's there so Meta can say they tried :)

That's basically the problem. If the shortcut produces something passable 95% of the time and nobody is checking, it just looks like you're faster. Journalism just has a more public failure mode than most fields.