A social turing test disguised as a game: https://humanornot.so/ Heavily inspired by https://humanornot.ai/ (which was a limited time research by Ai21 Labs), now the project is on its own path to be more that just a test. My work is to make AI chats sound like real humans and it's shocking how good sometimes the AIs are . Even I as a creator, knowing everything (prompts, fine-tuning data, design, backend etc.),... - Source: Hacker News / 4 months ago

It was about a game called humanornot.ai. Source: 6 months ago

Context: the website is Human or Not. It’s a game where you try and figure out if you’re chatting with a human or an AI. I had this wholesome interaction with this dude/dudette and just figured I share it here. Source: about 1 year ago

On a related note, check out https://humanornot.ai/. Source: about 1 year ago

Does https://humanornot.ai work on this browser? Source: about 1 year ago

Https://shadow.goose.icu/?http://acid3.acidtests.org/. - Source: Hacker News / 8 months ago

During the browser wars, web developers created “Acid Tests” to see how well a browser supported various style sheet and markup conventions. When the acid tests became popular, there were accusations that browser engineers were manipulating their rendering engines to pass those tests. For instance, some cascading style sheet attribute would pass the test, but fail to work as expected under normal use on a web page. Source: over 1 year ago

I am sure that both firefox and chromium-based browsers will pass most of something like ACID3 test. But at this stage standards have nothing to do with reality. Standards define perhaps 10% of the code you are writing and interpretation of the rest 90% of the code is up to the browsers. Source: almost 2 years ago

The old version (http://acid3.acidtests.org/) will reach 97/100 on modern compliant browsers. There is a newer version (https://wpt.live/acid/acid3/test.html) which incorporates the changes made to the specs in the meantime, so modern browsers will reach 100/100 on this one again. The test was carried out against this version, as you can see in the address bar in the screenshot. - Source: Hacker News / about 2 years ago

Different PRESENTATION of content on mobile is good. Different content on mobile is the definition of r/assholedesign. It does not matter if the http user agent is a "good" one or not, it should just work. There is a specification, numerous accessibility guidelines, plenty of tests and benchmarks. This isn't 1999 anymore... 🤷‍♂️. Source: over 2 years ago