"KH" == Kurt Hornik <Kurt.Hornik@ci.tuwien.ac.at> writes:
Martin Maechler writes:
>> [CC'ed to R-devel, not R-help]
"Bill" == Bill Simpson <wsimpson@uwinnipeg.ca> writes:
Bill> Thanks very much Douglas for the pointer to nlm. Maybe the
Bill> "Notes on R" maintainer can add at least a mention of nlm in the
>> ======================
Bill> section on nonlinear fitting?
Bill> .....
Bill> (This example could replace the current nonlinear fitting section
Bill> in "Notes on R")
>> You are talking of <CRAN>/doc/Rnotes.ps, or Rnotes.tgz respectively,
>> yes?
>> The first page states both R & R as authors (beside Venables &
>> D.Smith). However, I doubt if they consider themselves as
>> 'maintainers' of these notes (due to lack of time), or do you?
>> Ross? Robert?
>> The problem is that these notes need more work than just ('nls' <->
>> 'nlm').
>> Maybe I should ask again for volunteers, or maybe even a discussion
>> on what could / should be done.
>> The notes have several sections (organized in different *.tex files)
>> in Rnotes.tgz. Could(/should) we have different authors
>> (ie. volunteers from among you) for these sections?
>> About a year ago or so, Kurt Hornik proposed that these notes should
>> become chapter 1 -- n of ``the book'' where the current things in
>> RHOME/doc/manual/Man.tex are just the appendices (they already *are*
>> appendices, currently). R & R instead, originally thought (and have
>> also started ??) of completely rewriting from scratch these chapters
>> (1 -- n).
>> Which should be done (and by whom) ?
>> In the case of "R notes upgrading, it'd make sense if someone first
>> works thru the current notes and makes them (somewhat)
>> ``compatible'' with the Rd.sty that we currently use for the
>> help-manual (i.e. the doc/manual/... things).
KH> We DEFINITELY need to do something about R notes very soon. The
KH> 0.61 tree should be frozen by this weekend, 0.62 needs the bug
KH> fixes that we know of, the re-implementation of factors (asap,
KH> please!) and docs for the new graphics things (dev.xxx), but that's
KH> about it, I would say (we can have the `real' make install things
KH> later).
there are a few more graphic things.. [ par(par()) should work.. ]
KH> Working on Rnotes should at least come right after that.
ok
KH> I am no longer sure about LaTeX and my suggestion from about a year
KH> ago, though. I would more be in favor of having a Texinfo version
Really? I have to admit (my first reaction) that I wouldn't want to drop
all the LaTeX niceties [starting with "math", "real" tables (tabular), ...]
KH> as I could also use this on-line via info, but perhaps everyone
KH> else thinks that I should use now-free Netscape for doing that (my
KH> experience is that the conversion Texinfo -> HTML is trivial
KH> whereas I am never sure about latex2html or hyperlatex).
For me, latex2html has been quite satisfactory, as long as I am willing to
spend some time hand editing the result
(after the hand e
My feeling is that latex2html is really being improved and extended,
(being very configurable) constantly which is not/[much less] true for texinfo.
Also, AUC-tex (with font-latex) as the emacs interface to latex is such a
beauty compared to texinfo mode...
And then there's PDFtex which produces foo.pdf instead of foo.dvi
where everything (ref/label, cite, index, tableofcontents, ...)
is linked with everything [but pdftex is alpha/beta,...].
Maybe, the main point should be:
>>>>>> How are these notes used? <<<<
If ``most of the time: on paper'', I think I'd prefer latex to texinfo
If ``substantially: via Web browser'', maybe texinfo is better.
KH> In any case, I perhaps could try to go through Rnotes sometimes
KH> this weekend to find out what most urgently needs fixing.
ok, thank you in advance.
Ross / Robert: PLEASE we need some statement from you about this!
--------
Martin
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
If ``most of the time: on paper'', I think I'd prefer latex to texinfo
If ``substantially: via Web browser'', maybe texinfo is better.
I think the notes should be intended primarily for paper. There are
things hypertext is good for, but reading large chunks of mathematical
text isn't one of them (even on a 20 inch monitor). HTML/info would be
useful for cut-and-paste and for when the printed copy is lost, stolen or
strayed, but I'd much rather read from paper.
I would be prepared to do some of the revision in June/July, in particular
the model formula/frame/matrix stuff and glms, if it doesn't get done by
someone else first.
How will we handle notes on packages? Terry Therneau has really good
documentation for survival4, which we may be able to include somehow.
Thomas Lumley
-----------------------------------------------------+------
Biostatistics : "Never attribute to malice what :
Uni of Washington : can be adequately explained by :
Box 357232 : incompetence" - Hanlon's Razor :
Seattle WA 98195-7232 : :
------------------------------------------------------------
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
If ``most of the time: on paper'', I think I'd prefer latex to texinfo
If ``substantially: via Web browser'', maybe texinfo is better.
I think the notes should be intended primarily for paper. There are
things hypertext is good for, but reading large chunks of mathematical
text isn't one of them (even on a 20 inch monitor). HTML/info would be
useful for cut-and-paste and for when the printed copy is lost, stolen
or strayed, but I'd much rather read from paper.
Actually, I am very much undecided on this. Of course, we do have all
the help pages for on-line reading, but the question is whether they
contain everything one needs. If someone wants to know what to do with
lists, for example, looking up the help page for list() and then perhaps
going through subscripting etc will be a little complicated.
Or, suppose someone eventually sat down and carefully explained what
expressions, names, ... are.
To put it differently, the on-line thing I am looking for would NOT have
a lot of math in it.
As an aside, there is the additional issue that if someone writes R code
and quickly wants to look up the exact syntax of a function, it may be
great if one could do that without firing up R.
And yes, cut&paste would be great in a structured introduction. But
this could also be in HTML obtained via latex2html/hyperlatex, so no
need for using Texinfo just for this purpose.
I would be prepared to do some of the revision in June/July, in
particular the model formula/frame/matrix stuff and glms, if it
doesn't get done by someone else first.
Sounds great.
How will we handle notes on packages? Terry Therneau has really good
documentation for survival4, which we may be able to include somehow.
We could have a package `doc' subdirectory, and install its contents to
some place in the `doc' tree.
-k
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._