For me, and many others on Twitter, updating the DNS worked (I used 1.1.1.1). Though I'm not sure why.

Same for me, i changes DNS and it worked

Worked for me too!

It says at the bottom:

> When multiple moves have the top score, they are all top moves, even if visual markers (like move arrows) suggest the engine prefers one over the other.

Since all moves that checkmate the opponent will have the same score (M1 or -M1) they'll all be illegal.

At the moment it can recognise the type of statements in the training data set [1] and transpile them to Python, Java or C++ using the mappings defined here [2].

This is very different from how Codex/Autopilot work as it is trained using an NLU framework [3] which is usually used for training chatbots.

[1]: https://github.com/Flex-lang/transpiler/tree/master/transpil...

[2]: https://github.com/Flex-lang/transpiler/tree/master/transpil...

[3]: https://github.com/RasaHQ/rasa

So I can input pseudo code and it will turn that into real code? Is that a proper description? Can you give me a use case? Thank you.

Yes, that's a way to describe it.

One use case could be that a complete beginner to programming can enter some natural commands (e.g. "X is a list of numbers", "read 10 values in X") without worrying about the syntax and convert them to a working program in one of the target languages.

Also, the input doesn't have to be in English as it can be easily trained with samples from other languages. So people can write/speak in their native language and get it translated to working code.

In case you're looking for a VS Code extension to quickly preview, filter and plot data from various file formats you can check out vscode-data-preview [0]

[0]: https://marketplace.visualstudio.com/items?itemName=RandomFr...

Awesome, thanks for the suggestion. Already installed :)

Maybe letting the coin fall on a hard surface, so that it has a chance to bounce/spin randomly before falling flat (or staying on its side!)

It will also save the time and effort currently required for cattle trading by both sellers and buyers.

For instance, Animall (one of the companies mentioned in the article) is working towards building an online marketplace for cattle (like Amazon) rather than just a discovery platform (like eBay or OLX) and will be handling verification and logistics (think "Amazon Fulfilled"). This would solve fraud and also reduce the cost and effort for sellers/buyers in trading cattle.

P.S. I work at Animall

For seeding Big Buck Bunny of course. And Linux distros too. That's all :)

I'm going to build a repl.it clone too this weekend. Seems like a fun project!

Or an "Awesome list" on GitHub.

https://github.com/topics/awesome

Thought of the same, if no one else beats me i'll get started on this in a few hours

To everyone, the list has been created, the rest is easy :D

https://github.com/sofixa/awesome-museums-online

Tomorrow/early next week I'll throw in the boilerplate ( contributing.md, PR template, and see about organisation/generation/TOC/etc.) and then i'll start adding the mentions from this thread and others i know of.

Once it's decent and ready for wide contributions, i'll post on HN, Reddit and co.

Initial version ready https://github.com/sofixa/awesome-museums-online, i'll keep adding museums ( from this thread and elsewhere) and sections

Thanks!

I made this years ago, still update it sometimes.

https://github.com/polm/awesome-digital-collections

"The Short"