4 ms·
Promises Are Made to Be Broken: Migrating R to Strict Semantics [pdf]
- kkoncevicius 5y agoRoss Ihaka also mentioned getting rid of lazyness as one of the things he would do if he were re-designing R. Here is his presentation from 2017: https://www.youtube.com/watch?v=88TftllIjaY https://www.youtube.com/watch?v=88TftllIjaY
- landlordlubber 5y agoDid they ever release that new language they talk about?
- kkoncevicius 5y agoRemember I tried looking for it via a quick google search and didn't find anything. But I wasn't too invested in finding it, so didn't spend much time looking. Thinking about it now - he would probably reply to an email - that should be the quickest way of finding out.
- klodolph 5y agoThis has also been identified as something that some people would like to see in Haskell. Basically, an alternative version of Haskell which uses eager evaluation, rather than lazy evaluation. I'd say that the reason behind this is the discovery of what makes Haskell cool. Lazy evaluation seems cool, so you get a language with lazy evaluation and try it out. In such a language, you'd want functional purity, in order to reason about it more easily. Then you discover that functional purity is amazing, and unlocks all sorts of interesting possibilities... but the same is not true of lazy evaluation. It's only logical to want to explore pure functional languages with eager evaluation.
- NoThisIsMe 5y agoYou're likely aware of it already, but just in case, and for others: checkout PureScript
- CornCobs 5y agoI somehow can't download TFA, but I can't imagine strict R being anything similar to the current language. A huge chunk of the language's functionality depends on non-standard evaluation