[Grace-core] How many method application brackets does one language need?

[James Noble]

On 13/03/2016, at 10:49am, Andrew P Black <black at cs.pdx.edu> wrote:

>  I have no energy for reaping this and figuring out all the details of yet another solution

Certainly in the interests of progress, I'm happy to keep them for now.
I presume you need to teach with them; I want to get a spec out SOON. 


>  Probably because I don’t see the problem with lineups.

I think a few problems have been identified - the question is how serious they are.

 - caller side confusion on which syntax to use writing "list (1,2) vs list [1,2]"
 - syntactic ambiguity of "[]" for generics (better than "<>" but still there with lineups)
 - minimal interface of "[]" versus full interface of "seq []" 
 - we can't find a good strong name for "lineups", it may be hard to communicate why we have them 
 - we still have something collection-like baked into the core of the language
 - additional complexity of another feature 

I don't remember discussing the option of 
 - doing away with varargs
 - relying on overriding by arity to get the same effect, just with messier declarations in collections
in Pomona or elsewhere. 

Now, I just wish that we discussed that before fixing on "lineups" 

James