4 ms·

Motoko: A programming language for building directly on the internet

6 points by davnicwil 6y ago

nudpiedo 6y ago
ok, nice claims... but where is the language page/framework/api documentation or demos? Perhaps this one here? https://sdk.dfinity.org/docs/index.html https://sdk.dfinity.org/docs/index.html
hardwaregeek 6y ago
It's a little lame having a new language post without any examples. I'm also skeptical that a new language is worthwhile. If I were assessing a new tech stack, I'm not sure I'd be fine with a new platform and a new language at the same time. That's too much unproven territory. If the platform turns out to be a bust, I'd have a bunch of code in Motoko that I'd have to deal with.
anhner 6y ago
does this really need to be reposted every day? [1] [1] https://hn.algolia.com/?dateRange=pastWeek&page=0&prefix=false&query=motoko&sort=byPopularity&type=story https://hn.algolia.com/?dateRange=pastWeek&page=0&prefix=fal...
threatofrain 6y ago
https://news.ycombinator.com/item?id=24262448 https://news.ycombinator.com/item?id=24262448
mkl 6y ago
Big discussion from 4 days ago: https://news.ycombinator.com/item?id=24262448 https://news.ycombinator.com/item?id=24262448 A few other instances recently and 10 months ago: https://hn.algolia.com/?q=motoko https://hn.algolia.com/?q=motoko
setzer22 6y ago
Worth mentioning that, apparently, this is the programming language that will power DFINITY, an ethereum-like network for distributed computing.
choeger 6y ago
The type system description is exactly one paragraph with a lot of strong claims: > Motoko has been designed to be intuitive to those familiar with Javascript and other popular languages, but offers modern features such as sound structural types, generics, variant types, and checked pattern matching. (There are also some more paragraphs that broadly describe what a type system is, which is not a bad idea to have, but does not tell you anything about the actual type system). I did not find anything about the dynamic semantics of the language. Guys, when you design a programming language, the most important things are type system and dynamic semantics. If you do not publish the gory details, it makes you look like amateurs. If you think you can easily hack a type system with all the features you like, you are wrong. If you think you can rule out undefined behavior without a very detailed understanding of both, you are wrong. And finally, if you think you need to reinvent this stuff to get your whole "computation in the internet" system going, you are wrong, too. You can just as easily extend an existing language like Rust, ML, Haskell, or Java to deliver the same benefits. To be completely honest, this looks like you got a compiler-construction phd student in your team that told you what a type system looks like and now wants to do their own implementation. This is ok, but please tell them to publish a more detailed spec or you might end up with something utterly broken.
- lindig 6y ago
  You might want to check the publication record of the author, Andreas Rossberg: https://dblp.org/pid/90/3425.html https://dblp.org/pid/90/3425.html
  - choeger 6y ago
    Intriguing. Thanks for the hint. Now I wonder even more about the lack of detail.

6 points by davnicwil 6y ago

ok, nice claims... but where is the language page/framework/api documentation or demos? Perhaps this one here? https://sdk.dfinity.org/docs/index.html https://sdk.dfinity.org/docs/index.html

hardwaregeek 6y ago

It's a little lame having a new language post without any examples. I'm also skeptical that a new language is worthwhile. If I were assessing a new tech stack, I'm not sure I'd be fine with a new platform and a new language at the same time. That's too much unproven territory. If the platform turns out to be a bust, I'd have a bunch of code in Motoko that I'd have to deal with.

anhner 6y ago

does this really need to be reposted every day? [1] [1] https://hn.algolia.com/?dateRange=pastWeek&page=0&prefix=false&query=motoko&sort=byPopularity&type=story https://hn.algolia.com/?dateRange=pastWeek&page=0&prefix=fal...

threatofrain 6y ago

https://news.ycombinator.com/item?id=24262448 https://news.ycombinator.com/item?id=24262448

mkl 6y ago

Big discussion from 4 days ago: https://news.ycombinator.com/item?id=24262448 https://news.ycombinator.com/item?id=24262448 A few other instances recently and 10 months ago: https://hn.algolia.com/?q=motoko https://hn.algolia.com/?q=motoko

setzer22 6y ago

Worth mentioning that, apparently, this is the programming language that will power DFINITY, an ethereum-like network for distributed computing.

choeger 6y ago

The type system description is exactly one paragraph with a lot of strong claims: > Motoko has been designed to be intuitive to those familiar with Javascript and other popular languages, but offers modern features such as sound structural types, generics, variant types, and checked pattern matching. (There are also some more paragraphs that broadly describe what a type system is, which is not a bad idea to have, but does not tell you anything about the actual type system). I did not find anything about the dynamic semantics of the language. Guys, when you design a programming language, the most important things are type system and dynamic semantics. If you do not publish the gory details, it makes you look like amateurs. If you think you can easily hack a type system with all the features you like, you are wrong. If you think you can rule out undefined behavior without a very detailed understanding of both, you are wrong. And finally, if you think you need to reinvent this stuff to get your whole "computation in the internet" system going, you are wrong, too. You can just as easily extend an existing language like Rust, ML, Haskell, or Java to deliver the same benefits. To be completely honest, this looks like you got a compiler-construction phd student in your team that told you what a type system looks like and now wants to do their own implementation. This is ok, but please tell them to publish a more detailed spec or you might end up with something utterly broken.

lindig 6y ago
You might want to check the publication record of the author, Andreas Rossberg: https://dblp.org/pid/90/3425.html https://dblp.org/pid/90/3425.html
- choeger 6y ago
  Intriguing. Thanks for the hint. Now I wonder even more about the lack of detail.