Comment by AdieuToLogic | Hacker News Reader

Comment by AdieuToLogic | Hacker News Reader

Parent Full thread

> Generic methods are a huge win for the language. I've been waiting on these kinds of improvements to the type system ...

It's funny that you and many others have found the introduction of "generics" in Go to be highly valuable, considering one of the motivations for Go's existence was:

  Its designers were primarily motivated by their shared dislike of C++[0]

One of the language features C++ provides, "templates", was explicitly rejected by the language authors as being antithetical to Go philosophy[1]. Thompson put it bluntly:

  DDJ: In the presentation before the awarding of the Japan 
  Prize today, you were quoted on the distinction between 
  reasearch and development. [The former, Thompson stated, 
  was directionless, whereas development had a specific goal 
  in mind.] So in that context, is Go experimental?
  
  KT: Yes. When the three of us [Thompson, Rob Pike, and 
  Robert Griesemer] got started, it was pure research. The 
  three of us got together and decided that we hated C++. 
  [laughter] [2]

And now, years later, generics are "a huge win."

0 - https://en.wikipedia.org/wiki/Go_(programming_language)#Hist...

1 - https://commandcenter.blogspot.com/2012/06/less-is-exponenti...

2 - https://web.archive.org/web/20110521080746/http://drdobbs.co...

Generics were always planned, of course: https://www.youtube.com/watch?v=rKnDgT73v8s&t=3267s

Rejecting templates does not mean rejecting generics.

> Generics were always planned, of course ...

This is provably incorrect.

The position held for many years by the language authors was[0]:

  Generics may well be added at some point. We don't feel an 
  urgency for them, although we understand some programmers 
  do.
  
  Generics are convenient but they come at a cost in 
  complexity in the type system and run-time. We haven't yet 
  found a design that gives value proportionate to the 
  complexity, although we continue to think about it. 
  Meanwhile, Go's built-in maps and slices, plus the ability 
  to use the empty interface to construct containers (with 
  explicit unboxing) mean in many cases it is possible to 
  write code that does what generics would enable, if less 
  smoothly.

Once the community could no longer be held back, the golang FAQ presented a very different position[1]:

  The Go 1.18 release added type parameters to the language. 
  This permits a form of polymorphic or generic programming.

0 - https://web.archive.org/web/20170102202940/http://golang.org...

1 - https://go.dev/doc/faq#generics

Considering that Ian was already working on them before 1.0, and never stopped until a solution was found, we know for certain they were planned by at least one person on the Go team. If you are struggling to say that Go people are not a single monolith then sure. Nobody has ever thought people are a single monolith.

However, the original announcement makes the intent of the project clear: "Not yet", not "never". They were always planned to be accepted into Go. But not before an acceptable solution was found.

Do we know if Go team is working in 2.0 release?

We know they aren't.

They have successfully been able to add major features such as generics without breaking 1.0 compatibility, and the trend is to continue than way. Well there is no major language feature in sight in fact.

Also major figures of the Go team (Russ, Ian, Rob, Ken) are gone.

Oh I somehow completely missed Russ left. I was rather hoping there will be a version 2 from all the lessons they have learned.

Regarding generics (a.k.a. "templates"):

> However, the original announcement makes the intent of the project clear: "Not yet", not "never". They were always planned to be accepted into Go.

This is an exceedingly charitable interpretation of the position held by at least one Go language designer (Rob Pike)[0]:

  Notice that Robert said C was the starting point, not C++.
  I'm not certain but I believe he meant C proper, especially
  because Ken was there. But it's also true that, in the end,
  we didn't really start from C. We built from scratch,
  borrowing only minor things like operators and brace
  brackets and a few common keywords.

  In the end of course it came out quite different from
  either C or C++. More different even than many realize. I
  made a list of significant simplifications in Go over C and
  C++:

  ...

  - no templates

  ...

0 - https://commandcenter.blogspot.com/2012/06/less-is-exponenti...

C++ templates and Go generics are very much not the same thing.

> Generics are convenient but they come at a cost in complexity in the type system and run-time.

By run-time they mean compile-time? There shouldn't be a run-time penalty right?

Also, is there some measure of the additional compilation cost now that generics has been added?

The “runtime” as in the compiler runtime which provides the scheduler, garbage collector, etc.

The runtime also includes reflection, so preserving compatibility of the type system while adding generics was definitely a challenge, so adding generics to Go was definitely a great achievement.

"May well be added at some point" it said and added they were at some point.

> "May well be added at some point" it said and added they were at some point.

That is a very selective quote which does not reflect the context I provided. So I will extract a selective quote which negates the above:

  We haven't yet found a design that gives value
  proportionate to the complexity ...

I don't know how you're reading the quotes you provide, they seem to directly contradict what you're saying.

I definitely recall the days when it was not uncommon to see members of the Go community speak out against the idea of generics and other type improvements for allegedly taking away from the language's simplicity. Thankfully that faction lost that argument or otherwise changed their minds, because I have seen first hand how issues with the type system can hold back larger scale libraries.

A friend of mine years ago told me, "given enough time, every programming language will evolve into a LISP[0]."

I laughed at the thought in the moment and now understand the wisdom shared with me.

0 - https://en.wikipedia.org/wiki/Lisp_(programming_language)

Reply on news.ycombinator.com