[R] Question about <<- assignment

[Duncan Murdoch]

On 7/2/2009 9:46 AM, Peter Dalgaard wrote:
> Duncan Murdoch wrote:
> 
>> <<- doesn't need to find z.  It will replace it if found, or create a 
>> new one if not.  (Personally I would have limited that to the first 
>> case, i.e. it should fail if it doesn't find z.)
>>
> 
> Possibly. It's a holdover from S, where <<- assigns to the global 
> environment unconditionally.  (S doesn't have lexical scope, so there 
> are effectively only two places to choose from: the function evaluation 
> frame and the global frame -- unless you start mucking around with 
> sys.frame, frame 0, frame 1, etc., in which case you'd use assign(...))

I didn't say we made the mistake ;-).

Duncan Murdoch