I don't know why you speak in the past tense, but I consider Go still not being "innovative" enough and also broken by design. We'll talk in about 2 years from now when they'll finally add generics to the language, in a broken way nonetheless, as generics do have to be baked in right from the start to pull it off. And it badly needs generics.
But you see, Go got all this attention just because it is a language released by Google. And in the meantime version 2.0 of a real systems programming language that is innovative and kick-ass is largely going unnoticed.
FWIW I actually agree about the generics, both that Go needs them and that bolting them on afterwards typically doesn't work as well as having them from the start. But in general I wouldn't call it "broken" and view being "not innovative enough" as a feature for a systems language.
Everyone I know who has actually used Go strongly disagrees with this claim, they are very happy with Go's interface system and most of them don't notice the lack of generics at all.
Go is not like other languages, and 'features' are not interchangeable across languages.
> But you see, Go got all this attention just because it is a language released by Google.
Yea, that it was created by Ken Thompson and Rob Pike had absolutely nothing to do with it. /a
how is having classes, interfaces, generics, optinal types, operator overloading and functiontypes a small, solid feature set? Scheme, self that are languages with small feature sets.
I was talking about go, which doesn't have generics, optional typing or operator overloading, doesn't really have "classes" (arguable), and a fairly tight type system which covers structs, interfaces and functions in a minimal way.
I don't really have an informed opinion on dart yet, I'm just noting that 90% of the complaints I'm seeing consist of people with a hobby horse that predated the Dart announcement.
I don't really have an informed opinion on dart yet, I'm just noting that 90% of the complaints I'm seeing consist of people with a hobby horse that predated the Dart announcement.
It's a good thing then that the author of this article falls in your 10% group: someone who actually knows a thing or two about language implemetations, and not a hobby horse rider.
He may know a thing or two about language implementations, but his complaints are predictable: he wants it to look more like perl and less like Java. Additionally, his opinions on the implications of their concurrency model are entirely based on perl's implementation and IMO not necessarily correct in another context (message passing was probably chosen specifically to avoid needing heavyweight threads). I'm not a web programmer but those complaints seem superficial to me, I'd probably spend a little time using it before coming to that conclusion.
1) I have a hobby horse 2) Dart gets released 3) Flog hobby horse
Go had a bunch of similar criticisms, mostly for not being "innovative" enough and favoring a small, solid feature set.