R-beta: Mac version

[Peter Dalgaard BSA]

Colin Farrow <C.Farrow@geology.gla.ac.uk> writes:

> Peter,
> 
> Cautiously sticking my neck above the water line here. I have been
> wondering about the Mac version for a while now. I may be able to get
> an MSc IT student to work on this as a project for about 3 months
> June-Sept. Before putting up the proposal some idea of the scale of the
> task would be useful. Do you know what would be involved? Is there a
> preferred compiler?
> 
> CC'd to R+R for when the power returns:-)
> 
> Colin

I got a letter on the same topic a bit earlier today from John
Bacon-Shone. I'm copying my reply to him below. Basically, I think
that the scale of the task is highly dependent on whether you know
your way around the Mac API to begin with. As for compilers, I have no
knowledge of the Mac platform, but in general freeware is preferred,
if it exists (and works!), because it enlarges the number of potential
bugfixers. (I saw something about the PWB being available freely from
Apple now).

I'll Cc: this one to R-devel too. If anyone there who actually know
what they're talking about (as opposed to me) re. Macs have anything
to add, please do.

---------
John Bacon-Shone <johnbs@hku.hk> writes:

> >Seriously speaking, yes, we really should get around to getting the
> >Mac version up to date. Volunteers?
> 
> What needs to be done?  Create an up-to-date CodeWarrior project, or do coding?
> 
> John
> 
> Dr John Bacon-Shone, Director, Social Sciences Research Centre,
> The University of Hong Kong, Pokfulam Road, Hong Kong
> Tel: 852-28592412  Fax: 852-28584327 E-mail: mailto:johnbs@hku.hk
> Web http://www.ssrc.hku.hk

If only I knew... I suspect that some actual coding changes are needed
for the event loop and the graphics interface, whereas the main body
of the parser and the internals, as well as the interpreted code are
highly portable. That's the situation with the Windows version anyway,
and I can't believe that MacProgramming can be worse that the Win32
API.

One thing that I have been looking at a bit is the possibility of
coding up a Tcl/Tk version of R. This could be made portable between
all three architectures (X/Win32/Mac) and probably evolve into a set
of R/Tk language bindings a.m. STk (Scheme Tk), so we could have menu
interfaces and R functions that are call via a button-click, etc.

[I'm Cc-ing this to the R-devel list. Slightly bad netiquette since
your email was to me privately, but I hope you won't mind.]
-----------

-- 
   O__  ---- Peter Dalgaard             Blegdamsvej 3  
  c/ /'_ --- Dept. of Biostatistics     2200 Cph. N   
 (*) \(*) -- University of Copenhagen   Denmark      Ph: (+45) 35327918
~~~~~~~~~~ - (p.dalgaard@biostat.ku.dk)             FAX: (+45) 35327907

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-devel-request@stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._