_plap (@___plap) header
_plap (@___plap) profile picture

ASI onboard quadcopter

NYC
Joined April 2012
Photo by _plap (@___plap)
if you're in defense, pivot to physical AI alignment
3
91
War has stayed primitive because you can't A/B test it. This is now ending
3
45
Claude Code /goal has become unusable. Agents consistently stop after an hour of work despite having clear, verifiable tasks. The only reason to keep using it is for pseudo-frontier tokens
1
57
To boot, each session consistently stops one to two times getting a PR shipped autonomously, despite clear instructions to do so
24
~/.claude/plans/dazzling-strolling-dijkstra.md -- AGI confirmed
155
if you're not using macOS voice-two-text in Claude Code, you're stuck in the permanent underclass already.
168
OAI just logged out all 20 of my chat tabs and is making me type a 2fa auth for EACH TAAAAAAAAAAB aaaaaaaaaaaaa @tszzl pls fix
4
791
powershell is ranked higher than rust, WHAT
Notice the decline path of Haskell. Proves my long-standing points: 1. There was a chance for Haskell in 2010-2012. 2. In 2010-2014, haskellers were trying to convince the whole world that CATEGORY THEORY IS THE ONLY TRUE FP AGREE OR GTFO. Didn't work. Gave Haskell an extremely bad look. 3. In 2014-2015, I started saying we need software engineering practices for Haskell and FP. The answer was "Category Theory and Abstract Algebra is our software engineering, you damn midwit!" I said "no" and started writing posts and articles. In 2016, I started writing the book "Functional Design and Architecture." 4. Around 2012-2014, politics and ideology entered functional communities. In the next 4-6 years, this sparked tons of drama, squeezing pragmatic people and businesses away from the FP languages. The peak was around 2020. 5. Inability of Haskell to deliver, its neglect of engineering practices and unjust division of people into worthy and unworthy depending on ideological criteria ruined the momentum that Haskell was losing anyway. 6. In 2020, I published a post, "What killed Haskell, could kill Rust, too," a transcription of Uncle Bob's talk, "What killed SmallTalk, could kill Ruby too," with small substitutions made. The post perfectly aligns with the decline of Haskell in the picture. Recently, I was blamed for driving businesses away with my skeptical posts and words. That I'm somehow responsible for Haskell's decline. They found a scapegoat and the fact that I wrote three advanced Haskell books doesn't mean anything to them. But the trends were clear, and the reasons were obvious. Haskell lost its chance and sadly will never have another one. I want to ask for forgiveness that I was right all the time.
2
2
402
is it a coincidence that reusable rocketry and AGI seem more and more likely to emerge at the same time?
2
161
let's add fees for honking in manhattan. externalities may meet or exceed congestion.
3
142
when intelligence is free, most software will be bespoke
1
2
138
dreaming of an ai agent that transfers the auth code to fidelity after it forgets "keep me logged in" for the nth time
2
130
i can't wait for sora 6.0 to fully regenerate game of thrones' last three seasons
1
122
is this some mass psychosis? the average age of the founding fathers in 1775 was 41.
Everyone angry about the age of DOGE workers is going to be very surprised when they find out how old the Founding Fathers were.
2
211
this is literally just a myth that originated with Feynman. of the top 30 contributors, the average age was actually 38-39.
Average age of engineers and scientists in the Manhattan Project was 25. Our current gerontocracy is ahistorical. Perhaps one reason startups work so well is they are one of the few places that still let young people exert agency.
1
3
273
how are we supposed to believe simultaneously that deepseek has no gpus yet it has enough to 'mog' openai at serving? or... could it be that o3-mini's release depends more on openai's own cluster build outs?
The fact that DeepSeek mogged OpenAI into giving o3-mini out for free is interesting. Now that agents and operators are out, let me tell you a bit about what's going to happen next. First, people will need to learn to use these tools. But keep in mind that these are just the first iteration of these tools. New best practices will need to be established both on the user side as well as the server side. But now that OpenAI has demonstrated a thing, all the competitors will be scrambling to the duplicate it. Within 6 months there will be several commercial alternatives, as well as at least one or two open source alternatives reaching parity. But this doesn't even begin to underscore the compounding returns we're about to see. Productive will start to climb for everyone, businesses will adopt, scientists and researchers will adopt, and it will just keep going. Remember, everything we've seen up until this point has just been chatbots. We're seeing the first generation of semi autonomous AI being rolled out en masse right now. It seems I have a few videos to catch up on...
1
2
160
what few people realize is that results like this one legitimize Elon’s criticisms of the F35 last week. global superpowers will map the limits of their airspace in near realtime using drones equipped with next-gen event cameras / SPADs. the USAF already maps the oceans like this using sonobuoys deployed via MQ-9B.
After months of work, a Flyby F-11 drone can be turned into a massive 3D scanner. We just digitized our test site with a 12-minute flight.
1
2
343
just saw four amazon drivers drafting on the highway. those drivers are delivering shareholders every last bip, bless their hearts.
3
85