[R] R on LinuxPPC

cstrato@EUnet.at cstrato at EUnet.at
Tue Nov 9 22:23:44 CET 1999


Dear Prof Ripley

Thank you and your colleagues for the answers.

I would like to comment once again on the "object explorer":
The nice things about S-Plus are
i, the interactivity, i.e. to get results immediately and to try out things
very easily
ii, the object orientation of the language

In a sense S-Plus for Win95 functions as integrated development environment,
and the nice feature of its object explorer is that you can open all your
results in
table form and check immediately if the results are ok. This does not only work

with test-data used to try my own functions, but also with the real data, which
in
my case are currently data.frames with more than 7,000 rows. I can even change
data in the spreadsheet like form on the fly.
Of course, I could export and re-import the tables, but this is no longer
intuitive.
Besides that, I know currently no spreadsheet program for Linux, which I could
use for these purposes. gnumeric is not able to handle 7,000 rows, and
StarOffice
would use too much memory itself.
The main disadvantage of object explorer, which I see, is that it takes often
minutes
to open S-Plus. (However, recently I managed to bomb the object explorer in a
way
that I had to throw away and completely rebuild the user folder with all my
data.)

Since S-Plus (and R) is an object-oriented language, it would be great to  have
a
class browser similar to other object-oriented languages.

It is clear that R is freeware and that most people do development for fun. I
am
happy, that R does exist, so that I can use my Mac-PowerBook (with Linux)
and need not use Wintel-boxes at home, but as a Mac user I have come to
appreciate a nice user interface, since I can concentrate on the work. Now,
that
Linux has KDE, it would be great to see a version of R which takes advantage
of KDE.

As a final point, I personally think there is no need to go cross-platform, it
would be
enough to stick with Linux and KDE. I don´t care about Wintel, and although
I would have preferred to see S-Plus on the Mac, I can live with the Linux
version of R.

I hope I did not offend anyone too much with my statements.

Best regards
Christian Stratowa, Ph.D.
Vienna, Austria

Prof Brian D Ripley wrote:

> On Sun, 7 Nov 1999 cstrato at EUnet.at wrote:
>
> > > Hi,
> >
> > Currently,  I am working with S-Plus on Win95, but I have now installed R
> > on my PowerBook G3.
> >
> > Now I have the following questions:
> > 1, How can I see large tables?
> > When the table is  a matrix, I can use "data.entry(ma)", however, when the
> > table is a "data.frame"
> > this seems not work.
> >
> > 2, How can I scroll to the end of the table in this "data.entry" window?
> > (e.g. with 10,000 rows)
> > The keys <fn>-page up/down and <fn>-end don´t work (although they do work
> > when I am using
> > e.g. the midnight commander mc)
>
> In both cases I would write out the object and use my favourite
> editor/pager to examine it.
>
> As others have commented, we think the spreadsheet-like data editors will
> be re-worked.  One big issue is look-and-feel: on each GUI platform there
> is a standard way to do such things, and they differ. For Windows, Excel is
> so standard that people are talking about using that if available.
> (Literally using it via automation.)
>
> > 3, How can I see which functions I have defined in R?
>
> > Note: One of the most useful features in S-Plus for Win95 is the "Object
> > Explorer", which shows me
> > all objects, i.e. lists, matrices, etc., but also results of calculations,
> > and also all functions,which I have
> > imported or defined, and which I can open upon double clicking. It would
> > be great, if a similar feature
> > would exist in R.
>
> That `feature' has been a cause of such grief in the Axum-based Windows
> version of S-PLUS that I don't think we with our limited resources will be
> trying a cross-platform solution in a hurry. Notably, MathSoft have tried
> and failed to port it to Unix.  Also, this is the first really positive
> comment about it I have ever heard (and I have heard 100s of negative
> ones). Replacing `most useful' by `least useful' would win quite a few
> votes amongst S-PLUS users of my acquaintance.
>
> Both are things that advanced users seem not to use at all, although the
> lack of cross-platform solutions must be a factor.
>
> --
> Brian D. Ripley,                  ripley at stats.ox.ac.uk
> Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
> University of Oxford,             Tel:  +44 1865 272861 (self)
> 1 South Parks Road,                     +44 1865 272860 (secr)
> Oxford OX1 3TG, UK                Fax:  +44 1865 272595

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help 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-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._



More information about the R-help mailing list