If you work on anything worthwhile, sooner or later ppl will want to know how it's going. You may need to send investor updates, updates to your boss, emails to adjacent teams, etc. Some hard earned tips: 1. understand your role, and with each update add to the body of evidence that you’re a good steward in that role. If ppl want your updates, they’ve entrusted you with something-- a successful delivery of a product or feature, investment capital, company budget, their reputation, something. Convey that you value their trust and take stewardship seriously 2. add a little randomness to the cadence. Ppl think they want regular cadence, but they’re happier with bounded irregularity. For example, if you send project updates every Tuesday they will seem transactional and no one will read them. If instead you send updates every 2-3 weeks, your audience will look forward to them because they’ll assume you have something new to say 3. know what your next update will be and work toward it (instead of coming up with an update when it’s time to send one). If you don’t have a headline you don’t have an update, and you can’t generate good headlines post-factum 4. always start with a one sentence TL;DR and a 2-4 sentence recap of the overall goals of the project. Assume your audience is smarter than you are, but is very busy and remembers nothing about your work 5. ppl love pleasant surprises, but these don’t come along often enough by chance. Within reason, deliberately engineer pleasant surprises so you can include them in your updates 6. ppl hate unpleasant surprises (obvs avoid these if possible.) Talk to each person privately before informing the group. Also this is controversial but it sometimes helps to deliver negative news in two phases. First express the possibility of a problem to give ppl time to adjust. The next time state the problem. (Don’t do this if there is a genuine emergency obvs) 7. acknowledge changes explicitly. If you said A the last time and B this time, you need to explain the inconsistency. Ppl perceive acknowledged inconsistencies as cost of doing business, but unacknowledged inconsistencies as broken promises 8. don’t insult anyone. I once wrote an update that said something along the lines of "our engineers don’t know anything and therefore can’t ship, we need better engineers", and sent it to everyone. Definitely don’t do that 9. ppl want a steady hand at the helm. Your tone should reflect that. You want the text equivalent of pilot radio voice (mixing metaphors, I know) 10. many ppl (correctly) worry they’re being personally evaluated so they sanitize every sentence to death. Don’t do this. Make it all about the work, not about you, and leave the rest to others 11. know the top three questions your audience wants answered, and state the answers as clearly as possible 12. add a dedicated section for worries and failures. Be honest, have good plans, and have no fear. Ppl are drawn to conscientiousness and vulnerability but repelled from haplessness and histrionics 13. “Elon yells at Wall St analysts in quarterly earnings calls, why can’t I?” If you build a company with a market cap greater than the rest of its industry combined, you have my blessings to ignore this advice Note, these tips ^^ don’t work if you’re incompetent

For (1), agents modified their target programs to be easier to exploit & put the modified targets in cache. They then worked on crashing their targets in the hope that a restart would load the modified version from cache. Some agents risked failing their task to try this.

"West Virginia wants to use data centers to eliminate its income tax," per Reason

modern java (top) rust (bottom) the java you thought you knew is no more

Almost every startup founder remembers the first time they got screwed over in an enterprise sales process. It follows the same pattern every time. Everyone’s excited at first. The calls are going well. You tell the whole company you’re about to land a Fortune 100 logo. Then: death by a thousand paper cuts. Round after round of contract redlines. Custom feature requests. SOC 2 compliance. “Can this part be on-prem?” Each ask seems small. Together they start consuming the entire company. Then, at the very last minute, the hammer drops. They agreed to annual pricing, but now they want monthly optionality. Or someone else walks in with the exact playbook you spent six months workshopping with them and offers it for 30% less. All that work. And it was never really your deal. It’s almost never worth it. The better way: price something so cheap a developer can sign up with a credit card. Bill monthly. Let the account grow inside the company on its own. Big fish chasing is almost always an empty quest. The whales you actually land are the ones that swim to you.

SITUATION EXPLAINED: A stealth model on OpenRouter is beating Fable and Sol on coding, and nobody knows who made it. • Ox Alpha has a 1M token context window, text, image, and video input, free for a week, with capacity for 100 trillion tokens a day • On a DeepSWE subset it hit 80%, against Fable's 65%, GLM-5.3's 62%, and Sol's 52% • The tokenizer is one-for-one identical to GLM's, and a live test asking whether Taiwan is part of China returned a nearly identical answer • This is the 5th anonymous drop in six months, and the previous four were all claimed by Chinese labs: Zhipu, Xiaomi, Ant, and Meituan @theojaffee: " There's no way it's a Flash model. If that's a Flash model, then holy shit, the entire United States is cooked. If that's a Flash model, we need to short the stock market now. Not financial advice."