Summary of mails concerning R help system

Friedrich Leisch
Thu, 29 Jan 1998 10:17:29 +0100


here's a short summary of all the mails I got concerning the R help
system during the last days ... I really want to have all open
questions in this list, so if I forgot your most important topic
please forgive (and remind me again) ... it's probably only because of
lack of time such that I had to compile the list in a hurry ...

Ok, here we go:

1) It would be useful to have a button at the bottom and/or top of each page
which returns to the index. If you jump around a few "see also's" it is hard to
get back to the index ==> fairly easy, will be in the final version

2) dealing with packages not located in $RHOME/library ... that's the
most complicated part ... ideally the package list would look
different for every user, such that everybody sees the system-wide
installed pkgs together with his private pkg's

two possible solution's come to mind:

 - building the pkg list at run-time (via help.start())

 - building a list for every user, i.e. everytime one issues R INSTALL 
   a file in the user's home directory is modified

both have one major drawback which I don't know how to handle: I have
to jump between a static file located in $RHOME/docs/html and a
user-specific file, say $HOME/.R/pkg-list.html

but there seems to be no tilde expansion or $HOME variable in html,
i.e., an URL like "file:~/.R/pkg-list.html" works neither in netscape
nor lynx.

3) Short file names for windoze 3.1 systems ... there seems to be an
easy solution that I missed so far (simply replace the filenames by
something meaningless, because I use hash arrays to match aliases and
coprresponding files anyway), so this could be finished by
monday (if it really works like I think it should).

4) Rdconv -t rtf  to support the windoze help format ... I definetely
don't have time or any motivation to do that, so if somebody wants to
do that, he's welcome.

5) That's actually a wish/suggestion from me: Maybe somebody wants to
help the R project but doesn't know too much about writing code
etc. ... we need a lot more of and better documentation, e.g., short
tutorials on various topics to get people started. These should
explain the basic concept of something and accompany it with some
examples that can be copied & pasted ...

The current demos or the appendix of the R notes could get you
started.  Topics could include those as the the non-existing links
under ``Data Analysis with R'' (Basic Statistics, Exploratory Data
Analysis, Linear Models, ...)

So if you think you could write two or three pages on how to use R for
any topic you could simply write it in LaTeX ideally in such a way
that it can be translated to HTML using hyperlatex ...

r-devel mailing list -- Read
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: