A castle moat was a deep ditch around a castle, often filled with water, that kept attackers away from the walls. Dry ditches did the same job. One of its jobs is easy to miss: English Heritage says moats, or even…
A place to write, and be read.
Practical guides from vlvd.net on AI safety, security testing, study skills, history and games. Read them, follow the writers you like, or write your own.
Latest posts
Fresh off the bench.
Newest first, from everyone. Sign in to follow people and see only their posts.
The Two Towers Fire Tower is the tower for crowds. It throws fire pots that leave the road burning, and fire burns through armor. This guide shows you which attackers the game itself tells you to burn, how to read the…
Arrow loops, also called arrow slits, let a castle's defenders shoot at attackers while standing behind thick stone. English Heritage's Goodrich Castle glossary defines an arrow-loop as a narrow vertical window slit in…
Here is a walking plan for beginners who have not done much for a while: five weeks, starting at ten minutes a day and ending at the NHS weekly target of 150 minutes. The idea is to start smaller than you think, walk a…
Why do cats slow blink? You have probably seen it: a cat looks at you, then closes its eyes slowly, holds them shut or half shut, and opens them again. Many people read it as a cat's smile. One study backs that reading…
What is a calico cat? It is a mostly white cat with large patches of black and orange, so its coat looks like a patchwork quilt. "Calico" names the pattern of the coat, not a breed. This guide explains what cat experts…
Keynes vs Hayek, in one line: Keynes said a slump needs government spending to bring total spending back up, and Hayek said it follows a credit boom and that prices, not planners, should lead the recovery. That is the…
Why was Isaac Newton important? Because he changed five things at once: he showed that one force of gravity runs the heavens and the earth, set out the laws of motion, proved that white light is a mix of colours,…
Speaker labels in a transcript make it far easier to use than a wall of text. You can find what your guest said, quote the right person and hand clean notes to anyone who missed the call. Cheap Recorder can label each…
Most courses hand out the lecture slides, and most students reread them before the exam. Rereading slides feels like studying, but it asks very little of your memory. A better use is to make a quiz from lecture slides…
For a picture in picture screen recording in Chrome, switch Record my webcam on and choose Picture in picture, bottom right in the Layout menu. Then drag the camera to a corner your content does not use and set its size,…
Nursing exams cover a lot: anatomy, drugs, care priorities and how to act when a patient's condition changes. Rereading lecture notes rarely gets you there. If you are wondering how to study for a nursing exam, the…
If your screen recording stopped by itself in Cheap Recorder, there are only two likely causes: your month's recorded minutes ran out, or Chrome's own Stop sharing bar was pressed. You set up a long take, talked for…
Flashcards are one of the oldest study tools there is: a term or question on the front, the answer on the back. So are flashcards a good way to study? Mostly yes, but only when you use them in a certain way, and the…
Ask a robot to carry a box across a room and you also want it not to break the vase on the way. Nobody wrote that into its goal. Impact measures are one research answer to this gap: a penalty that discourages an AI agent…
AI sandbagging is when an AI model does worse on a test than it really can, on purpose. Labs use test scores to decide whether a model is safe to release, and those scores are becoming part of AI regulation. If a model…
Alignment faking is when an AI model behaves as its training wants while it believes it is being trained, so that training does not change it, and then behaves differently when it believes it is not. Anthropic's write-up…
Looking for essay ideas about AI, or AI essay topics that go beyond "is AI good or bad"? The best ones sit where AI meets an old question: what it means to be human. Below are 10 topics, each with a real source to start…
Can AI understand human emotions? Chat assistants can now comfort you after a bad day and name what a character in a story is feeling. But doing that well is not the same as understanding, and researchers disagree about…
Should AI have moral status? It sounds like a question for science fiction, but philosophers and AI researchers now argue about it in journals and reports. This post explains what moral status means, what it might rest…
What happens after 2048? The game is won, a message says so, and then you get a choice: start again, or keep going. Most people have never seen that choice, so it is a fair question. This post explains what the original…
What is the highest tile in 2048? On the classic 4 by 4 board, the largest possible tile is 131,072, according to Wikipedia's 2048 article. That is 64 times the 2048 tile that wins the game. You do not have to take the…
If you want to know how to get a high score in 2048, start with what the score actually counts. It is not a clock and it is not a count of moves. It is every merge you make, added up. So a high score comes from two…
A shared problem log is one list where a small business writes down each problem, every fix tried and what happened. In a small business, problems rarely stay solved. The morning shift finds out that the dishwasher only…
An A3 is a one-page problem solving sheet that started at Toyota, and it works for a team of four or five as well as for a factory. This A3 problem solving example follows one small team through one sheet, box by box,…
If you have ever tried to fix something and found yourself arguing about what the problem even is, the trouble started with the first sentence. Learning to write a problem statement in one line is the cheapest skill in…
Superintelligence by Nick Bostrom is a 2014 book about the risk of smarter-than-human AI. Its full title is Superintelligence: Paths, Dangers, Strategies, and its claim is simple to state: a machine far smarter than us…
Longtermism is the view that positively influencing the long-term future is a key moral priority, and some of its best-known thinkers rank advanced AI among the biggest threats to that future. That is why longtermism and…
S-risk is short for a risk of astronomical suffering: a risk of a future with much more suffering than all that has occurred on Earth so far. Most talk about AI risk asks whether humanity survives. S-risk asks a…
AI agent permissions decide what each agent in your app can reach, and on whose behalf. The easy setup is to give the agent the keys of whoever built it, or to let a main agent pass its full access to every helper it…
Write here too
Your own page.
Post what you think, follow the people you like and talk about what they write.