๐ค I wonder if coding agents (e.g. Claude Code) might yield a better experience for folks with ADD/ADHD.
Personally Iโm finding it pretty boring and tedious to give direction and oversight to the agent and then wait until it next needs input/feedback/oversight/directions. I guess Iโm supposed to switch away and work on something else in the meantime? But thatโs *way* too much context switching. I find it cognitively taxing.
Weโre trying out Claude Code at work. I just gave it a pretty sophisticated debugging problem, and it is *flailing*. It reads like a bunch of smart people in a room with a wide range of skill and expertise working on the problem, but theyโre all pretty drunk and keep grabbing the keyboard from each other to try something different. Bizarre.
Iโm returning the Instant Pot I purchase recently (and have not yet used).
https://www.eater.com/news/886762/trump-maga-instant-pot
Gross.
Iโve been seeing Fisker Oceans more often recently. Seems odd.
Itโs time for me to either get rid of my Leaf or spend $1K on a CCS to CHAdeMO adapter.
Leaning towards the former. I should probably just get a CCS car.
(I sold my final Tesla last week. Good riddance. Great car, reprehensible brand.)
This is a great way to explain Jevonโs Paradox:
@mathowie thanks for posting about Visor! I have a similar relationship to the car market, and I agree, itโs fantastic!
๐จ Great sci-fi alert! ๐จ
These Burning Stars by Bethany Jacobs is *excellent*
https://app.thestorygraph.com/books/2153ea5c-318b-40d7-a1f3-757b6f8225e1
My review of Devil in the Stack: A Code Odyssey by Andrew Smith:
> Iโm a software engineer and I was impressed by how well this book captures the *feels* to work on software today. Not only that, but thereโs plenty of explanations of how computers work, all of which I knew already, but I was never tempted to skip those parts. They are refreshingly clear, straightforward, and accurate. The prose is clear, cogent, brisk, and wry. This all added up to a very enjoyable book.
https://app.thestorygraph.com/books/a8baf746-d4af-425b-ae7c-39fcd294b451
Iโm trying Claude Code (CC) after reading [AI Changes Everything][1] by @mitsuhiko and my initial reaction is that Iโm a bit baffled.
It keeps suggesting small, individual changes (diffs) to specific lines of specific files, and asking me to approve them or not. Most of them are OK, some of them are not. Whatever. But I have to sit here twiddling my thumbs between each interaction. Itโs tedious. I could do this work faster on my own, and Iโd have higher confidence that itโd actually work. (Some of the changes will need verification โ especially because Iโm running CC on a โquick and dirtyโ โtemporaryโ project that has no automated tests.)
Iโm guessing maybe Ronacher has CC configured to not prompt about each individual change, but to just go ahead and work through an entire change โ in other words, the review/update/review cycles are larger, coarser-grained. So maybe Iโll try that.
