Skip to content

Proposed changes to the R Language

7 messages · Angus C Hewitt (DHHS), Jeff Newmiller, Bert Gunter +3 more

Original
Angus C Hewitt (DHHS)
# 
Hi Team,

Please advise if there are any plans in the pipeline for change the R language to "B"  as proposed in the below mentioned statistical series help at Auckland Uni.

https://www.youtube.com/watch?v=88TftllIjaY


Kind Regards,

Angus Hewitt
Senior Analyst | Decision Support
System Design, Planning & Decision Making |  Health & Well Being
Department of Health and Human Services | 19th floor, 50 Lonsdale Street, Melbourne Victoria 3000
t. 9096 5859  | m. 0468 364 744 | e. Angus.Hewitt at dhhs.vic.gov.au
Jeff Newmiller
# 
I have no idea. But I hope not... that sounds like a different tool than R, just as C++ is a different tool than C.
On December 27, 2018 4:36:42 PM PST, "Angus C Hewitt (DHHS)" <Angus.Hewitt at dhhs.vic.gov.au> wrote:

  
    

      
      
    

  


  


      

    

    

      
      

        


  

        

      Bert Gunter
    

    

      
      #
    

  


  

      
I would assume r-devel is where this sort of query should be posted as we
mere users have nothing to say about this.

However, I've seen discussions and talks about better languages for
scientific (but data science?) programming -- Matlab, Julia, Scipy, etc. --
for at least a decade. But with a library of now over 10,000 packages on
CRAN and yet more on Bioconductor and github -- that's a lot of inertia to
overcome.

Cheers,
Bert


Bert Gunter

"The trouble with having an open mind is that people keep coming along and
sticking things into it."
-- Opus (aka Berkeley Breathed in his "Bloom County" comic strip )


On Fri, Dec 28, 2018 at 7:13 AM Jeff Newmiller <jdnewmil at dcn.davis.ca.us>
wrote:

            

      
      
      
    

        

      
      
    

            

      
      
      
    

        

      
      
    

        

  
  


  


      

    

    
1 day later

    

      
      

        


  

        

      jacob bogers
    

    

      
      #
    

  


  

      
It has some good ideas, but R (my personal assesment) is not build for
superspeed, but for superease of use.))


On Fri, Dec 28, 2018 at 3:01 PM Angus C Hewitt (DHHS) <

        
Angus.Hewitt at dhhs.vic.gov.au> wrote:

        

            

      
      
      
    

        

      
      
    

        

  
  


  


      

    

    

      
      

        


  

        

      angus hewitt
    

    

      
      #
    

  


  

      
Hi Jeff,

The suggested language looks like it is shifting to some stricter programming idioms such as defining object's data structure. this video was shoot in 2017 so I'm not sure if these changes have been put on the back burier.

AH

        

      
      
    

  
  


  


      

    

    

      
      

        


  

        

      Jeff Newmiller
    

    

      
      #
    

  


  

      
I don't think you can assume that any of this _research_ was ever on any R-core burner at all. But as Bert said, this would not be where such discussion occurred if it was... go to r-devel to find out, or better yet contact the researchers themselves.

        
On December 29, 2018 9:53:01 PM PST, angus hewitt <angus_hewitt at hotmail.com> wrote:

            

      
      
      
    

        

      
      
    

        

  
    

      
      
    

  


  


      

    

    

      
      

        


  

        

      John C Nash
    

    

      
      #
    

  


  

      
My friend Morven Gentleman who died recently was for some time chair of the computer
faculty at Waterloo and (Fortune nomination!) once said "The response of many computer
scientists to any problem is to invent a new programming language."

Looking at Ross Ihaka's video, I got the impression he wants to preserve R syntax as
much as possible while improving speed and predictability. His example

  x=10
  f=function(){
   if (runif(1) > .5) x=20
   x
  }

where the local/global status of x is unclear underlines just one issue.

The temptation with programming languages is to allow them to expand. That makes
it very much more expensive and difficult to preserve the package ecosystem in a
healthy and verifiable state. A robust dialog between users and language
developers is, I believe, the best way to move to a streamlining of R. If -- and
that is the big question -- Ross' ideas are workable to provide a simplified
R (whatever we call it) that allows a high proportion of existing codes to run
satisfactorily or with minimal/automated changes, the R community would benefit.


John Nash