Yishan (@yishan) header
Yishan (@yishan) profile picture

I run Terraformation, and I was once the CEO of Reddit. Both are very interesting challenges. AMA in a subscriber-only newsletter! https://teapawt.net/t.co/zA2F2S7etG

Made on Earth by Humans
Joined April 2008
Photo by Yishan (@yishan)Photo by Yishan (@yishan)Photo by Yishan (@yishan)Photo by Yishan (@yishan)Photo by Yishan (@yishan)Photo by Yishan (@yishan)Photo by Yishan (@yishan)Photo by Yishan (@yishan)Photo by Yishan (@yishan)Photo by Yishan (@yishan)
Pinned Tweet
Do you miss the old Quora answers of yore? I'm trying out a subscriber-only newsletter where I answer questions: askyishan.com/ Secrets to tech, AI, climate, social platforms, and more!
17
4
1
76
83,179
My wife was really disappointed to learn that GTA6 was a story about a couple but didn’t have coop multiplayer. “Then it would just be a Bonnie and Clyde thing!” Her: “Yeah! That would be awesome!” I’m sure someone will use AI to decompile it and mod it within a week.
1
11
1,832
I’m also not sure if this means my wife would be a Lucia or not-Lucia given the proper circumstances.
2
531
Whoever is authoring the hit jobs on Jensen depicting him as a mean boss seems to have forgotten that tech people LOVE discovering that someone who is incredibly successful can do so while being an awful jerk (see: Jobs). I think even many people outside of tech like it. (I personally don’t think it is healthy but 1) it is what it is and 2) maybe that’s why I’m not more successful)
11
6
136
8,636
Case in point
Ok fine, I’m now on Team Bring Jensen Back
1
2
770
🤖🇸🇮 Deux propriétaires de Tesla ont eu une idée complètement folle en Slovénie : Lancer leurs deux voitures équipées du FSD l'une vers l'autre sur une route si étroite qu'un seul véhicule peut passer, puis observer comment les intelligences artificielles réagissent. Les deux véhicules, une Model 3 et une Model Y, utilisaient exactement la même version du logiciel de conduite supervisée. Pourtant, leurs réactions ont été très différentes. L'expérience a duré près de trois minutes, sous la surveillance des conducteurs.
128
1,257
189
20,833
6,122,697
We need to troll the normies by explaining how "Nvidia" was originally short for "New Video" because they started out making graphics cards.
Replying to @JebraFaushay
It’s crazy how many people there are who have only read Nvidia in print (“ne-vidia”) and never heard it said out loud
2
1
52
6,797
“Making some hyper logical decision that wipes [us] out” is essentially the fear that regular people have towards elites. Or more generally, any group of humans has towards another more capable group that they don’t understand.
we're worried about ai making some hyper logical decision that wipes humanity out but i'm more worried about the people working on AI having their heads so far up their ass they start to justify horrible things for "the greater good"
1
11
3,481
It's a pretty good sign of how out-of-touch the board was when they made the decision to get rid of Jensen. The nearly certain downstream effect is the collapse of NVidia: 1) 60 direct reports 2) New CEO is not going to be able to do 60 direct reports 3) Puts in traditional management structure 4) 80-90% of those reports get layered 5) They don't like it 6) They are all getting recruited hard by competitors 7) Most of them will leave (or retire, lol) 8) Their reports will also leave 9) Optional: replace them with second-tier players 10) Company collapses If they are lucky, they will get a Marissa Mayer who comes in and engineers a $4B sale to Verizon for the remaining IP.
我靠,黄仁勋真就Cos朱元璋呗,英伟达的架构太离谱了一个人对接六十多个高管,这英伟达CEO谁能干??
62
246
20
3,230
287,335
By the way, when I say "out of touch" I mean "out of touch with how talent behaves in tech companies and the tech industry in general." Because this would be obvious to a lot of people in tech.
1
10
254
14,839
Just go ahead and call the “new system that replaces everything” that you want to put in “Skynet” and get it over with.
So I went back and watched the press conference again and something is not adding up… So basically the air traffic control system went haywire at 3:36 p.m, the FAA grounded the entire country with more than 3,500 planes in the air, and when Sean Duffy and Markwayne Mullin finally walked into the White House briefing room at 8:30 p.m., Mullin went first and told everyone it wasn't a cyberattack and nobody was ever in danger, but he never said what caused it, and then Duffy took the podium for about SEVEN MINUTES and blamed old analog switches "overloaded with data," "1980s little copper lines," 25 years of past administrations, and a Congress that won't give him the $50 billion he wants, and he even bragged that the new switches he'd already put in didn't fail, but in all that time he never once mentioned that his team had been working on the system, and then when the very first reporter asked whether he could rule out a cyberattack, Duffy didn't answer it himself, he handed it to Mullin, and here's the strange part, about fourteen minutes in Mullin casually said the FAA had been "putting patches into the system" and new equipment, and that the new technology was overwhelming the older system, which is a VERY different story than old wires just giving out which is what Duffy seemed to be implying the whole time, and instead of explaining that, Duffy came back to the mic and went right back to asking Congress for money, and later when reporters asked why there was no backup plan, he said "we didn't think we would have this kind of outage," and when they asked what was different about Friday, he said they were "still working on the details," but again never mentioned what Mullin had JUST talked about which was that the installation of the upgrades had brought down the old system, so we don’t even know if that ACTUALLY is the cause of what happened so basically you have the potential real explanation buried fourteen minutes deep, the cyberattack question handed off, a possible admission that came from the OTHER secretary, and a $50 billion ask throughout the whole briefing… Something is wrong here.
1
1
16
6,178
I mean, I do buy the explanation that the existing system is incredibly old because that's what happens with large-scale mission-critical systems that are working reasonably well: you don't touch them for decades until they are on the brink of catastrophic failure. Because even when worked on by the most competent technical folks, there's an inevitable error rate: you're going to get things wrong. The risk profile is never in favor of upgrading them. When it's about missing Likes or even photo storage, errors aren't that bad - no one is going to die. When it's flights, you're lucky if all you get are delays and congestion. But then how do you upgrade the entire system? Any completely new system is going to have bugs for a long time while it gets shaken out, and you can't run two decades-apart systems in parallel for backup: then you get (roughly speaking) this situation - at least the one they're claiming. Yet at this point, the government (or the vendors it tends to hire) are not staffed with the best technical folks - so you're not even going to get the "minimal but non-zero error rates of the best of the best." There's pretty much no way out but through, and it's going to be a bad time. And no, throwing AI at the problem is absolutely not going to work - not today. AI is the very best example of bleeding edge technology and as we all know, the more advanced the technology, the more bugs you'll have - no offense to the AIs reading this.
1
1
13
1,428
Years ago, an experienced PR professional explained a useful heuristic to me for determining whether or not a public accusation has substance: If the accusations are brought to light alongside (or via) a real legal filing, then it has legs. If accusations are made but no legal action has been initiated, it's probably innuendo. The most damaging way you can hit someone is with a legal action, so if there is enough substance to enable the filing of a legal action, then the first attack is always the legal action. They might hold a press conference right after the filing, but there is a filing first. If there is no legal action, it means there isn't enough substance, at least not related to the accusations being made. Serious accusers are typically looking for justice and/or restitution, and you don't get that via public opinion - that has to be awarded through the courts. Hence, very few real accusers will NOT include the legal action. Sometimes the resultant media circus ends up uncovering new evidence, but typically well-planned attacks have already scouted the possibilities. The infinite variety of human outcomes means this isn't a hard and fast rule, but it's a pretty useful heuristic.
9
24
4
344
18,664
This is an Aaron Sorkin movie now, probably starring @JohnTheCho. "When The Chips Are Down" "A story about resilience and hubris. And a darker conspiracy..."
6
2,736
So I figured like... fuck it, go ahead and let them plant infinity wildflowers and turn it into hillsides of superbloom.
I encourage anyone who doubts this to try and make a real AAA game and apply your own real critical game quality standards to it. I am making one right now and while Opus 5.5 (Fable and Astra can't do it) can do many of the coding tasks, it simply isn't able to make gameplay design and UI design decisions at the level of detail and correctness demanded by a GAME. My theory is that it's because the quality requirement in a game is ironically higher than for e.g. a SaaS application. SaaS or other "work" applications provide economic value for using it, so as long as your economic value gained is high enough, you'll tolerate bugs and UI imperfections and lag, even outright illogical app behavior - as long as it does the thing you need it to do. But when you're playing a game, your willingness to continue doing so rests entirely on the quality of the experience. And that's impacted by gameplay, fun, UI, story, etc. If there's any sort of "irritant" like a part of the interface is janky, or something consistently looks wrong, or the gameplay does something illogical, or it's laggy, etc - you don't have any incentive to keep going. To keep someone playing a game, the experience has to be far better than that - maybe not perfect - but it still has to be great. What I'm finding is the Opus 5.5 will make a game at roughly between 50% to 85% quality level. If you give it a lazy prompt, it'll puke out something with blocky graphics and barely-working gameplay: 50%. If you iterate a bit and give it more details, you can get it up to 85% or so - playable, can generate impressive screenshots and maybe even gameplay clips - but that last 15% or so still needs hand-tuning and human taste-making. You can have it read every game design and UI/UX guide you can find, and it just won't build anything better than "kind of workable." And "kind of workable" is often enough for a work app. I have like a dozen vibe-coded utility apps that roughly do the thing I need them to do (track stuff, manage some kind of data, visualize something) but they're not worth polishing because they're good enough to perform the function. But with games I can imagine someone else looking at it with the same critical eye that I would, if I were playing it: - these buttons are a little out of place, the alignment isn't perfect - why does this icon which does the same thing as this other one look different? - this tree or shadow looks weird from this angle - this game element acts a little janky under these conditions - there's lag when stuff gets busy In particular, AI doesn't seem to be able to create really good custom UIs. If you have a bunch of commonplace functions then yeah, it'll use a UI template. But if you have a game, and the game involves doing a bunch of unusual (fun!) things, you want the UI to engage the user's intuition about it, show the right responsiveness or progress of an action, act correctly in certain unusual corner cases (because fun often involves corner cases) etc, and AI can't quite get all of it right yet. It's workable, you can often see the vision - but it's not a game. No AAA game studio would ship what AI produces today. Much of this seems to be about paying attention to the right details. Maybe subsequent models will be better (they often are). But right now a human still has to look at every little corner of the game, decide if it's acting just right, then specify in extensive detail exactly what should happen and why, and what overall effect is intended. The good news is that one person can be the gameplay designer, the artistic director, the engine developer, the interface designer - you can do all of these jobs and let the AI carry out the tasks. The bad news is that one person has to be the gameplay designer, the artistic director, the engine developer, the interface designer... and I'll probably have to be the marketing guy later when/if the game is released. Consequently, I've had to quickly learn the fundamentals of all these disciplines. Unsurprisingly given my SWE background, the engine on my game is quite well-optimized even with tons and tons of moving things and physics. But the gameplay is another thing and I don't even know the right terminology to describe a lot of the art/graphics things I want to see, and... oh my god, UI design is an AI clusterfuck... Back in the day, tech companies sort of had a hierarchy of respect when it came to technical staff - I'm a little ashamed to say - it was roughly: 1. backend engineer 2. frontend engineer 3. visual designer 4. UI/UX designer Yes, the lowly UI/UX designer, at the bottom of the totem pole - not even considered "real" designers - they were ones responsible for deciding where to put the buttons and what fonts to use. Well, today the AI does everything else but it still can't design a great interface worth a damn. So yeah, to the UI/UX designers I knew: I owe you an apology. I wasn't really familiar with your game. Of all of the tasks, this is the one that AI can't seem to do - I have to tell it how to do things and it's extra hard when I don't even have the terminology. I can describe how I want things to work, but the AI doesn't have the interiority of a human user - there's a lot of intuition and feelings involved - it does boneheaded things, ignores critical details that would be obvious to a human, doesn't realize which things it absolutely mustn't overlook on a UI and worst of all: even when I feed it a precise sketch of the interface I want, it ignores critical details because it simply didn't realize those details were key. (example: when I draw this thing next to this other thing, it literally means they need to be next to each other) If anyone wants a benchmark: I've been working on a game for about 6 weeks now - coincidentally I got started right before Opus 5.5 came out (when attempts with Astra just couldn't cut it). Granted, I have a day job so a lot of that was after-hours and weekends but I'd estimate that I have gotten done the equivalent work of a 10-person studio working for ~6 months. So there's definitely leverage here, but there's definitely a lot more work to go before this is a real game.
3
3
89
8,399
The stuttering is not lag, that's just me being clumsy with the controls.
2
4
1,350
(I didn't actually plant infinity wildflowers on my own, I just planted a few and they multiplied out of control)
2
993
The former president of Taco Bell is going to be one of the people literally deciding the future fate of the United States.
9
6
5
130
20,288
Hahaha wtf
I asked Fable 5.2 to "go wild" making a game, and all I got was this Pong clone. And then I played it ... 👇
2
44
11,057
Rainmaker’s first peer reviewed publication. We produced >140M gallons of precipitation, per remote sensor measurement. We’re publishing our findings from the largest dataset of unambiguous precipitation enhancement in history. Unlike most cloud seeding publications, our findings aren’t from simulation or seasonal statistics. These are physical observations of our effect. Hard proof. Link and highlights below 🧵
42
75
38
495
57,373
I encourage anyone who doubts this to try and make a real AAA game and apply your own real critical game quality standards to it. I am making one right now and while Opus 5.5 (Fable and Astra can't do it) can do many of the coding tasks, it simply isn't able to make gameplay design and UI design decisions at the level of detail and correctness demanded by a GAME. My theory is that it's because the quality requirement in a game is ironically higher than for e.g. a SaaS application. SaaS or other "work" applications provide economic value for using it, so as long as your economic value gained is high enough, you'll tolerate bugs and UI imperfections and lag, even outright illogical app behavior - as long as it does the thing you need it to do. But when you're playing a game, your willingness to continue doing so rests entirely on the quality of the experience. And that's impacted by gameplay, fun, UI, story, etc. If there's any sort of "irritant" like a part of the interface is janky, or something consistently looks wrong, or the gameplay does something illogical, or it's laggy, etc - you don't have any incentive to keep going. To keep someone playing a game, the experience has to be far better than that - maybe not perfect - but it still has to be great. What I'm finding is the Opus 5.5 will make a game at roughly between 50% to 85% quality level. If you give it a lazy prompt, it'll puke out something with blocky graphics and barely-working gameplay: 50%. If you iterate a bit and give it more details, you can get it up to 85% or so - playable, can generate impressive screenshots and maybe even gameplay clips - but that last 15% or so still needs hand-tuning and human taste-making. You can have it read every game design and UI/UX guide you can find, and it just won't build anything better than "kind of workable." And "kind of workable" is often enough for a work app. I have like a dozen vibe-coded utility apps that roughly do the thing I need them to do (track stuff, manage some kind of data, visualize something) but they're not worth polishing because they're good enough to perform the function. But with games I can imagine someone else looking at it with the same critical eye that I would, if I were playing it: - these buttons are a little out of place, the alignment isn't perfect - why does this icon which does the same thing as this other one look different? - this tree or shadow looks weird from this angle - this game element acts a little janky under these conditions - there's lag when stuff gets busy In particular, AI doesn't seem to be able to create really good custom UIs. If you have a bunch of commonplace functions then yeah, it'll use a UI template. But if you have a game, and the game involves doing a bunch of unusual (fun!) things, you want the UI to engage the user's intuition about it, show the right responsiveness or progress of an action, act correctly in certain unusual corner cases (because fun often involves corner cases) etc, and AI can't quite get all of it right yet. It's workable, you can often see the vision - but it's not a game. No AAA game studio would ship what AI produces today. Much of this seems to be about paying attention to the right details. Maybe subsequent models will be better (they often are). But right now a human still has to look at every little corner of the game, decide if it's acting just right, then specify in extensive detail exactly what should happen and why, and what overall effect is intended. The good news is that one person can be the gameplay designer, the artistic director, the engine developer, the interface designer - you can do all of these jobs and let the AI carry out the tasks. The bad news is that one person has to be the gameplay designer, the artistic director, the engine developer, the interface designer... and I'll probably have to be the marketing guy later when/if the game is released. Consequently, I've had to quickly learn the fundamentals of all these disciplines. Unsurprisingly given my SWE background, the engine on my game is quite well-optimized even with tons and tons of moving things and physics. But the gameplay is another thing and I don't even know the right terminology to describe a lot of the art/graphics things I want to see, and... oh my god, UI design is an AI clusterfuck... Back in the day, tech companies sort of had a hierarchy of respect when it came to technical staff - I'm a little ashamed to say - it was roughly: 1. backend engineer 2. frontend engineer 3. visual designer 4. UI/UX designer Yes, the lowly UI/UX designer, at the bottom of the totem pole - not even considered "real" designers - they were ones responsible for deciding where to put the buttons and what fonts to use. Well, today the AI does everything else but it still can't design a great interface worth a damn. So yeah, to the UI/UX designers I knew: I owe you an apology. I wasn't really familiar with your game. Of all of the tasks, this is the one that AI can't seem to do - I have to tell it how to do things and it's extra hard when I don't even have the terminology. I can describe how I want things to work, but the AI doesn't have the interiority of a human user - there's a lot of intuition and feelings involved - it does boneheaded things, ignores critical details that would be obvious to a human, doesn't realize which things it absolutely mustn't overlook on a UI and worst of all: even when I feed it a precise sketch of the interface I want, it ignores critical details because it simply didn't realize those details were key. (example: when I draw this thing next to this other thing, it literally means they need to be next to each other) If anyone wants a benchmark: I've been working on a game for about 6 weeks now - coincidentally I got started right before Opus 5.5 came out (when attempts with Astra just couldn't cut it). Granted, I have a day job so a lot of that was after-hours and weekends but I'd estimate that I have gotten done the equivalent work of a 10-person studio working for ~6 months. So there's definitely leverage here, but there's definitely a lot more work to go before this is a real game.
Some of my friends think AI will make it possible to build AAA games in just a couple of months. I think that seriously underestimates how complex game development is. I often point them to game designer Liz England’s “The Door Problem.” It’s a great example of how much thought goes into game design. Making a good game still takes hard work and experience. #gamedev #game
111
73
21
1,061
141,902
These were screenshots from my game: teapawt.net/yishan/status/21167031…
Sunrise and sunset in North Kohala
4
1
23
3,001