Conversation

Notices

1. Claes Wallin (韋嘉誠) (clackemovedtoheldscalla@quitter.se)'s status on Wednesday, 21-Sep-2016 17:49:25 UTC Claes Wallin (韋嘉誠)
   To read: The argument against classes in JavaScript: https://github.com/joshburgess/not-awesome-es6-classes/blob/master/README.md
   • Claes Wallin (韋嘉誠) (clackemovedtoheldscalla@quitter.se)'s status on Wednesday, 21-Sep-2016 17:51:04 UTC Claes Wallin (韋嘉誠)

     in reply to
     I already know that I agree with the sentiment, but some of these look like they could be enlightening.
   • Claes Wallin (韋嘉誠) (clackemovedtoheldscalla@quitter.se)'s status on Wednesday, 21-Sep-2016 17:52:56 UTC Claes Wallin (韋嘉誠)

     in reply to
     I first saw a class notion in JavaScript in YUI. I suppose if frameworks do that, a standard makes sense. i always favored functionalish JS.
   • Mike Gerwitz (mikegerwitz@social.mikegerwitz.com)'s status on Thursday, 22-Sep-2016 01:13:13 UTC Mike Gerwitz

     in reply to
     JavaScript's "classes" are syntactic sugar atop of the original prototype model. If someone's interested in Classical OOP as is commonly understood, they won't like them. If someone is interested in functional programming, they won't like them. If someone is interested in a Scala-like marrying of the two, they won't like them.
     
     I prefer functional programming, but the OO paradigm in itself isn't the problem. Since we use OOP exclusively at work, I wrote a Classical OOP implementation for ES3. Today it supports Scala's concept of stackable traits as well.
     
     https://gnu.org/s/easejs
     
     And for the record: I believe that classical inheritence (a class extending a class) is a violation of principle: it tightly couples, often breaks encapsulation for subtypes, and is just an all around mess. Given composition through objects and traits, OOP doesn't have to be bad. It gets a bad name because people (and languages) do terrible things. So do educational institutions.
     Claes Wallin (韋嘉誠) likes this.
     Claes Wallin (韋嘉誠) repeated this.