A robot arm dips a pipette into a dish and transfers a tiny amount of bright liquid into one of many receptacles sitting in front of another machine. When all the samples are ready, the second machine tests their optical properties, and the results are fed to a computer that controls the arm. Software analyzes the results of these experiments, formulates a few hypotheses, and then starts the process over again. Humans are barely required.
The setup, developed by a startup called Kebotix, hints at how machine learning and robotic automation may be poised to revolutionize materials science in coming years. The company believes it may find new compounds that could, among other things, absorb pollution, combat drug-resistant fungal infections, and serve as more efficient optoelectronic components.
This sounds like a perfect application for machine learning. It’s a great combination of smart design and brute force effort.
For a long time I thought I didn’t like carbonara because the few I tried were too creamy and eggy. Recently I gave it another try because the picture in the menu looked delicious. It was excellent, and I decided I wanted to learn how to make it myself. I needed some practice before I felt comfortable with the timings, but overall it’s a pretty simple recipe that wasn’t hard to learn.
I was consistently able to have lucid dreams when I was in high school. Partway through a dream, I would realize that I was dreaming and could then control the rest of the dream to a small extent. As I’ve grown older though, this has become less and less common. Maybe age has changed something about my brain.
Naturally, I’ve been interested in lucid dreaming and have read a lot about it. Tim Ferriss has a fascinating beginner’s guide on his blog. Different people seem to have different approaches and theories, so I thought I would write about my own experiences.
I wonder sometimes about what a better society than ours would be. What should our goals be as we try to improve?
As our society improves and we get more efficient, I like to imagine a future where we don’t have to work as much. That seems like a nice thing to aim for. Universal basic income (UBI) seems like one possible solution. If we could afford to give everybody a fixed minimal income, people could work less and still have a basic amount of wealth needed to survive. But I’m not yet convinced whether UBI is the correct answer to describe an ideal future.
I had been playing World of Warcraft on and off for about 10 years, but I decided to stop playing again about a year ago. I quit because of a series of changes that were made to the game, and it motivated me to think about why I reacted that way as a paying customer.
A series of recent changes made to WoW were designed to motivate players to play by ensuring they always gain something and grow a little every time they play. Unfortunately, a side effect of these changes was an increased pressure to spend many hours playing. This made me feel like my limited time spent in the game was less valuable, and is what ultimately led to me quitting the game.
Picular is a site that lets you search for words like “water” or “summer” and see a list of related colors. It seems to work by running a Google image search and extracting the primary color of that image.
This is the kind of project I love. It’s simple and easy to understand, but fun to play with.
Play Pac-Man online against other players. 1 player plays as Pac-Man and 4 other players play as the ghosts. Pac-Man’s goal is to collect as many dots as possible, and the ghosts’ goal is to work together to catch Pac-Man. Players receive points at the end of the round based on performance, and roles are rotated for the next round.
In terms of strategy, the ghosts will need to work together in order to win. Pac-Man will move faster than the ghosts, so the ghosts’ main strategy will be to trap Pac-Man in a corner. To encourage teamwork, all ghosts will receive the same amount of points at the end of round regardless of who actually caught Pac-Man.
I’ve been thinking a lot recently about what benefits there are to building a new product on a blockchain. Blockchains allow an app to be decentralized, and the biggest value they bring is a solid way to handle trust and consensus in an environment with no central authority. I was having a hard time thinking of products that would really benefit from being decentralized, but I came across an idea about decentralized platforms that makes a very strong case for using blockchains.
The idea, introduced to me by this post by Alex Tabarrok, is that a blockchain-based platform is uniquely able to align its needs with the needs of the users. If a traditional platform wants to create economic value for itself, it often needs to do things at the expense of its users. It can’t maximize value for both the users and the platform at the same time. But a blockchain-based platform can, through its tokens, increase its value directly by increasing value for its users.
When I started using Facebook in 2005, it was exclusive to college students and was by far the best way to communicate with my classmates. It was the first time I could learn what my friends were up to without asking each one individually. Everything I wanted to know about my friends was in one easily accessible place.
All recent updates were shown as one long list, but Facebook eventually introduced the News Feed. This was an effort to filter everyone’s updates to show just the most important ones. Facebook made educated guesses about whose updates were most important to you and hid the ones you wouldn’t care about. I think they also kept an option to see all updates in chronological order if you didn’t want to miss out on anything.