Rust's model of references confused me when I started out. I'd read that they were faster, so I tried to use them everywhere. You can't return a reference to a new vector though!
Now I see them as primarily useful for signalling 'this parameter is not modified' rather than perf.
miniblog.
@cstanhope A friend also pointed out hotels dot com :). I do think they're rare though.
Reflecting on the dot com hype, there are actually very few sectors where genericnoun dot com has become one of the biggest websites. Seeing https://chess.com/ today made me think how rare it is.
Are the other big examples?
In addition to appliance manufacturers exploring smart/wifi-enabled devices, there are now products that allow you to trigger physical switches wirelessly! https://microbot.is/push/
Reviews are very mixed, but with hindsight this class of device was inevitable.
Activity trackers now exist for animals too! Whistle is a Fitbit-style exercise monitor plus GPS tracker for dogs: https://www.wired.com/review/review-whistle-3-pet-tracker/
Ambient computing is becoming progressively more common.
The Turing Test focuses on distinguishing between humans and computers in a text chat.
There are lots of other domains where it's interesting to compare styles. Do we make different mistakes in speech recognition? How easy is it to spot a chess AI masquerading as human?
Rust's for loop is much less ceremony than a traditional C loop, or even a Python for loop: https://doc.rust-lang.org/1.1.0/book/for-loops.html
Simple abstractions having simple syntax really helps teaching.
Building AI to teach users games versus just play it: https://news.ycombinator.com/item?id=7829713
The efficiency of wasm is really impressive. An optimised build of Stockfish with POPCNT evaluates positions at ~1500kn/s on a single core of machine. By contrast, the wasm build on https://lichess.org/analysis/r4rk1/p5b1/2p2n2/1p2p3/4P2q/P1N1B3/1PP1B1Q1/2KR3n_w#0 can compute ~42kn/s in the browser for the same position!
GitHub will allow you star repo topics, not just individual repos: https://blog.github.com/2019-01-08-topic-starring/
Topics can be useful as a kind of hashtag to say what your project is about. I've not found them very useful for discovery so far. This could change usage patterns and improve matters!
Immutable documentation at Etsy: https://codeascraft.com/2018/10/10/etsys-experiment-with-immutable-documentation/
Basically a list of notes on a topic, collaboratively written on Slack. Notes aren't edited, you just append new ones.
Could work very well, depending on your team's dynamic. Super low friction, and similar to fsbot.
Fabulous article demonstrating how to write a chess AI, including several interactive demos!
https://medium.freecodecamp.org/simple-chess-ai-step-by-step-1d55a9266977
It also shows the importance of a big ecosystem. There are already JS libraries for computing legal moves and showing a chess UI, so you can focus on the AI!
In addition to common things like UI elements, Netflix even A/B tests site performance aspects to see how they affect user behaviour!
https://www.quora.com/What-types-of-things-does-Netflix-A-B-test-aside-from-member-sign-up
"phones, thanks to their centrality in people’s lives as well as the greater likelihood of harm, will always have a faster replacement cycle than PCs"
https://stratechery.com/2019/apples-errors/
(Sales and strategy analysis for Apple.)
One web design style that seems to have become unfashionable: external link icons:
External Page ⍈
I think it's because links are already distracting, and adding icons doesn't help. An email icon is still handy for mailto: links, as they don't open in the browser.
The Stockfish chess engine requires patches to pass a test: it must beat the old version a sufficient proportion of the time.
This introduces an interesting problem: what if a patch set makes it stronger, but applied individually they make it worse?
Which MOOC platforms are the most popular, according to Kaggle users? https://www.kaggle.com/ogakulov/the-mooc-wars-kaggle-s-perspective
Designing online systems so that groups of people are smarter than individuals is not easy. The first collaborative chess games weren't very strong. In Kasparov vs The World, organisers ensured there were several strong players and online discussions.
https://en.wikipedia.org/wiki/Kasparov_versus_the_World
The likelihood of technology problem having a good FOSS solution seems to be roughly
how passionate people get about it x how fun it is to work on x friction when contributing
I'm intrigued to learn that some Stockfish developers think that it would beat AlphaZero under conventional rules for chess engine competitions.
https://en.wikipedia.org/wiki/AlphaZero#Reactions_and_criticisms
Even if that's fair, AlphaZero is still a very impressive demonstration of their ML approach.
Showing 2,301-2,320 of 7,508 posts