I suppose if the new 'venture' is for-profit it might give credence to the view that Altman and the board diverged because of his desire to turn OpenAI toward exactly that.
There are many, many life lessons I have drawn from my youth spent playing poker semi-professionally. Perhaps the most consistently useful one was that, as mentioned in the article, one's performance should not be assessed purely on the outcome.
In poker, you can play perfectly and still lose. That is true in almost every field of life. Or, to give it more nuance, you can play exceptionally well and only win a little, whereas someone has had greater luck can play relatively poorly and win a lot.
"In poker, you can play perfectly and still lose."
Only true over the short term: all you have to do is play GTO poker over an infinite number of hands in a rake free game and you should at least break even I think!
That said, this fact is the main reason that the game of poker really "works". If playing perfectly (or at least at a much higher level than your opponents) allowed you to always best your opponents, then you'd quickly find yourself out of opponents. There needs to be enough luck involved to keep losing players coming back to the table (the same reason most casino games have a thin house edge).
Poker can be an absolutely soul crushing game to play professionally. "Running bad" where you lose money over 10's or 100's of thousands of hands even while playing well can make you seriously question your sanity. It can also cause you to add subtle (or not so subtle) errors into your game (e.g. playing too cautiously or aggressively) which can turn your once winning game into a losing game, making it tough to decide whether you are currently a winning player running bad, or actually a losing player.
I was a professional poker player in another life and I agree; learning not to be result-orientated is one of the biggest life-lessons it thought me. Deciding to flip a coin for money where you get paid 3x your bet if you win, as long as you can afford to lose half of the time, is correct, regardless of outcome.
The other one is not being afraid of "tough" decisions - often, decisions are tough because all options are close in value. The closer their value is, the less it matters to choose the "correct" path, keeping in mind the future is unknowable and we always have imperfect information.
Yeah, that's a really fundamental and important insight.
The way I like to think about it is that whenever you face
uncertainty, your actions might seemingly be proven wrong in retrospect, but may still have been correct based on your past knowledge.
That's undoubtedly correct, but I find it sometimes difficult to apply, since there's always the possibility that you could have overlooked important information, or that you are using that logic as an excuse. In poker, different from real life, you always know exactly what you know and what you don't know.
In one sense, if you have overlooked important information, then you don't have it. And looking for overlooked information is itself an action, with costs and benefits that you need to evaluate with your current information.
I like this comment and I learned the same lessons from golf and poker. Pay attention to the process and what you can control. As long as you make emotionless decisions and execute the process you shouldn't worry about the outcome. In fact the pressure of competing comes from the expectations you put on yourself by worrying about the outcome before you finish executing the process.
The problem here is, in professional poker, the long run might never come.
In online poker, every hand you play is tracked and can be analysed and people have "run bad" over hundreds of thousands of hands; more than a casino-player will ever see in their life at 20 hands per hour.
It should be dependent on your edge no? With tiny edge you can run bad way longer than you can with a big one. And in live-play you should be able to find a table with a huge edge.
You're basically the best player. You'll play maybe 20 hands an hour, live - most of which you'll instantly throw away and make $0. Let's say you play 50% of your hands like a (somehow winning) maniac. That's 10 hands an hour you might make money on. Even the best players' "won money in a hand they put money in" stat is way under 50% - lower, the higher %age of hands they decide play. So you'd expect get 4 winning hands per hour, or 40 hands a day if you have the stamina (and the bad opponents to play with!). Now, most of your winning hands will result in relatively small wins (your opponents also try to win and don't light money on fire) and guess what - the bigger wins and losses stem from the more volatile situations (e.g. going to showdown) where luck plays an even bigger role than in poker in general: It's totally normal to do everything correctly and having to root for a 32% outcome in a big hand.
That long run might never arrive for you, even though you're the best.
I'm not "against poker" by any means, I would not be where I am today without it. Live poker is a (nice) gamble though, not a job.
There is also an interesting life analogy that I like : During a tournament, an often underestimated skill is choosing your table (and thus your opponents).
This led me to look up whether the British Indian Ocean Territory (Chagos Islands), and therefore the British Government, received any money as a result of the .io domain.
It's amazing to me that British seem to get the full PR whack of that attrocity when it was all under the auspices of the US who paid the Brits to kick them out.
Kind of like how we all collectively forget that the Allies dropped the bomb on Japan and not just the "Americans".
TIL that the Americans consulted with the British before dropping the bomb on Hiroshima (but declined their request for British participation in the mission). The Soviets were apparently not notified in advance (although I could be wrong about that) and the British had veto power over what would be revealed to the Soviets about the bombing.
I didn’t know any one else was consulted in any serious manner about the Japanese bombings. From the other poster it appears the UK was heavily involved? It’s def more than the US, but if only them, then not all of the “Allies” unless we’re only considering the US and UK the “Allies”. In the end though, both cases had both countries. Not to mention other situations in the world like the Middle East.
Not sure if America had anything to do with India. Probably not.
I'm not sure this is really reflected in reality, perhaps aside from the 65+ generation. I grew up in Scotland, and it was very much a beer drinking culture.
This is addressed in the beginning of the article,
>The alcohol belts refer to the traditional beverages of countries rather than what is most commonly drunk by the populace today, as in terms of drinking habits beer has become the most popular alcoholic drink in the whole world, including various parts of the wine and vodka belts.
I have a feeling they got that leverage because these vessels are still on Earth, and the corporations are still bound by earthly limits. Space is a whole different game.
The point about 'getting used' to poor features, or indeed even bugs, applies not just to video games but any type of software you work on for a long time (especially personal projects, where you don't have a QA team).
This will no doubt be familiar to many of us on HN.
That's my understanding of what "Hallway usability testing" from the classic 'Joel Test' meant. Having fresh eyes on a thing vs. the same people 'testing' it forever.
There's a similar phenomenon when looking at an electronic todo list. You get used to reading it top-to-bottom and skipping over certain uncompleted items every time. It's nice if there is a feature to randomise the order of the list, to give you a jolt into properly considering each item individually again.
There is a free McDonald’s restaurant in the corporate headquarters, but there’s also a deli counter which has fresh sandwiches, as well as a counter for other hot meals (chilli, chicken curry (UK) etc.
