[Rd] R Newsletter: 1st Call for Articles

Prof Brian D Ripley ripley@stats.ox.ac.uk
Thu, 21 Dec 2000 08:06:35 +0000 (GMT)

On Thu, 21 Dec 2000 Bill.Venables@CMIS.CSIRO.AU wrote:

> With regard to the burgeoning side-discussion on whether or not to accept
> WORD submissions, I plead for a quick, decisive and final ruling in the
> negative.
> Apart from the major practical considerations to do with the amount of work
> putting together a document from sources of both kinds and achieving an
> acceptable result, there is the relatively minor issue of principle: R is
> open source software and should do what it can to promote the use of other
> open source products.
> I fully realise that this will create problems for people who are not
> familiar with LaTeX.  I hope for people like Yves it is the start of a
> liberating journey of discovery...but for many it will not be.  It would
> seem to me the best option for such people is simply to accept submissions
> in neat ASCII, (with some diffidence, of course, as such submissions will
> require someone to do a lot of work before they can be incorporated into a
> document).

One option for Word users is to save in RTF, which is at least a documented
format (and shared between word processors).  Then get a copy of rtf2latex  
or (better) rtf2latex2e from CTAN and check out the conversion.
Peter D mentioned MiKTeX.  I would recommend fptex to Windows users
instead: it is closer to the Unix/Linux standard system (teTeX) and is
more complete and up-to-date.  (The latter is particularly important for    
pdftex users: for example at one time fptex would process the R refman.pdf, 
and MikTeX would not.)  fptex used to be harder to instll, but no longer,
and it is the preferred system on the TeXLive5 distribution, I'm told.

> When it comes to submitting graphics, though, I have no easy solution for
> those unfortunates crippled and imprisoned by the MicroSoft arrogation of
> the universe.  It would be simple if R had a pdf.graph device: is any such
> development in prospect?  For that matter, how should Unix/Linux people
> submit graphics?

Yes, there is one already (take a closer look at bitmap, whose help says

     Note: despite the name of the functions they can produce PDF via
     `type = "pdfwrite"', and the PDF produced is not bitmapped.

).  A native PDF device is planned, and was one of the reasons for the
changes to postscript() in 1.2.0 (to enable code to be shared).
For some time my entry on developer.r-project.org has said

   A native PDF graphics driver.

and that site is a good place to look for future plans (but some team
members keep their secret).

I don't see the problem: R can produce PostScript everywhere, and it
can even produce EPS these days.  Windows users can produce other figures
in WMF if they want, but that can be converted to EPS successfully by a
number of routes.  If people want to annotate figures, R can produce .fig,
and xfig runs under Unix/Linux and Windows ....   So I suggest figures be
produced in PostScript.

Brian D. Ripley,                  ripley@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-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