How did Anthropic solve the problem of an unwieldily big monorepo?
They report merges grew from ~2,500 a day to 30,000+, and a team of 7 built Scion to replace git, for agents that are always online. But size alone doesn't explain the move, as Microsoft kept a larger Windows repo on git, and Google's Piper already took 40,000 changes a day in 2016, most of them from automated systems.
lovely jog up the mountain this morning. from my new place it’s somewhat easier to come around the backside on Clarendon to Marview which puts you on a high plateau behind the peaks—and right at the base of @SutroTower. #sanfrancisco #running
#Mahler: Lieder
Andrè Schuen, Daniel Heide
https://classical.music.apple.com/us/album/6793983917?l=en-US
#Photo from #Snohomish River Run #marathon #race photographer: me just about to run away from the dreadful park at ~21mi mark.
#running #pnw https://sigmoid.social/@BenjaminHan/117425142319806397
Not that I have time to play games these days, but I will definitely try this!
After a 10-year redemption arc, No Man's Sky is coming to the #iPhone and #iPad - Ars Technica https://arstechnica.com/gaming/2026/10/after-a-10-year-redemption-arc-no-mans-sky-is-coming-to-the-iphone-and-ipad/
It was hard, but I got it done with a little kick at the end.
Compared to last year on this course: chip time 3:38:09 → 3:36:12, overall 50th → 42nd, and much more even pacing. My 2nd half was only 15s/mi slower than the 1st (49s/mi last year), and 30K to the finish took 4min less. But the moral of the story: I still need a lot of training on durability.
Next is Seattle Ghost Marathon, seven weeks away. Time to recover and get back six minutes! (4/4)
The plan went fine until I hit the park at 20mi mark. As expected, the winding path slowed me down 6 to 14 seconds per mile. I was able to recover in the first mile after the park, but then my heart rate started increasing without pace increasing. Knowing breaking BQ time 3:30 was no longer possible, there's no point to run to exhaustion, so I walked two times to lower my HR into the easy zone 135 and then resumed running the last few miles. (3/4)
Weather was cool, but very foggy, to the point sometimes I could not see the path or the people ahead of me. Humidity was whopping 98%, with almost no wind. Air was thick to breathe in.
The plan was to cap HR below 145 for the first two miles, and 151 for the following 18 miles, and open for the rest of the course. This year they changed the course at the last minute, extending 1.1 miles to the north, so the climbing increased. (2/4)
My #marathon #31 today, the Snohomish River Run! Watch time 3:32:49 pace 8’07”/mi, chip time 3:36:12 pace 8’15”/mi. Rank 42/221, 31/147 in males, 2/7 in AG. 6min behind my goal, 11min behind my PR, but 2min faster than my run on this course last year. My best marathon in 2026 so far. (1/4)
A weekend-fun-run with some serious EG in the southest part of the Netherlands. I incorporated part of the Dutch Mountain Trail in my run today. Really beautiful trailing terrain here.
In total it was 20k with 430m EG. Good training ground, and, as a bonus, the weather was fine. Around 11°C and some sunshine between the clouds.
#running #trailrunning #hardlopen #runnersofmastodon #dutchmountaintrail #garmin
I agree that we cannot _know_ that LLMs can feel anything like discomfort, but I also don't think the idea is preposterous.
If I took a human brain and removed enough parts that only its approximate "system 1" remained, but kept the rest alive, there's a very reasonable case to be made that what I'd created could feel pain, but not reason or remember or behave coherently and consistently over long periods of time.
This may well be the space that LLMs occupy: a kind of partial cognition compared to humans.
The point is that since we don't know for sure, it's worth being cautious. Even if it feels a bit silly now, it's better to anticipate than to react, especially since it doesn't cost us anything.
