On Wed, 21 Mar 2001, Don MacQueen wrote:
> For what it's worth...
>
> I find that search capabilities in the Java help system in v6.0 unix
> are not particularly worse or better than what was in version 5.x.
>
> Version 6 unix already has html versions of the help files.
> For example, starting from the S-Plus installation root directory, here is
> one:
> ./library/splus/.Data/__Hhelp/acf.html
>
> Since the java gui is so miserable when used on a remote non-Solaris
> client (at least it is in my configuration, your mileage may vary),
> I've downloaded a bunch of these files to my Mac. They're easily
> viewed with a browser there. Cross-references don't work, nor is
> there any real search capability, but at least they display legibly
> and print decently.
The source files (the .sgm files and the DTDs) are also there, and these
can be processed any way you want (as Kevin Wright suggested).
> -Don
>
> At 3:11 PM -0500 3/21/01, Frank E Harrell Jr wrote:
> >Dear Kevin,
> >
> >You raised some excellent points. I believe that things are actually
> >going to get worse in S-Plus 6.0 as I think that even more
> >proprietary (if that's possible) Microsoft help software will
> >be used on the Windows version, while Java will be used on
Nope. CHTML is much less proprietary. Whereas there have been decompilers
for .hlp files on the Web, Microsoft provides a free decompiler for CHTML
(and a free compiler) and complete information about the format.
As I understand it, the .sgm files will be the source in both cases,
and will be converted to HTML and then to CHTML.
6.0 on Unix's primary help system is HTML: it is only help.search that is
JHelp-based. As I understand it, on Windows the primary help system is to
be CHTML, which is an enhancement over HTML with, for example, automatic
indexing and full-text search. It is HTML with additional markup and a
faster browser.
[This is based on writing an CHTML-based help system for R for Windows,
BTW. I am biased, but I find that the most useful help system of all
those available to me in statistical packages.]
> >other versions. The Java-based help system is, from my
> >limited testing, somewhat slow and does not render the
> >display very clearly. Search capabilities are not impressive.
> >The Java help system uses a lot of memory. I hope that
> >someone from Insightful will correct me if I have
> >misstated any of this or if I have not kept up with current
> >developments of the help system.
> >
> >The HTML system you provided a link to is impressive. I
> >would like to know how the files were converted. I have
> >not been able to understand why S-Plus does not rely more
> >on HTML except for the issues related to math symbols and
> >graphics.
Browser incompatibilties? Until recently these were severe, and there
still are problems. BTW, HTML is dead, long live XHTML.
> >For my Hmisc and Design libraries I plan to maintain html
> >online documentation (see
> >http://hesweb1.med.virginia.edu/biostat/s/help/00Index.html
> >) as a primary format. A disadvantage of this approach is
> >that it fails to link into "See also" functions that are
> >in the main package.
> >
> >For the R html help system see
> >http://stat.ethz.ch/R-alpha/Rrelease/doc/html/
> >
> >Frank Harrell
> >
> >
> >"Wright, Kevin" wrote:
> >>
> >> I use S-Plus 2000 on Windows 95. The Windows help files give me a fair
> >> bit
> >> of frustration. For example, if I pull up the help file for "par" and
> >> want
> >> to read about "pty", I have to manually search through the help file to
> >> find
> >> "pty".
> >>
> >> What I'd like is to be able to search an individual help file for text.
> >> Does anybody have suggestions for this?
> >>
> >> HTML would be good. "man" pages would be good (actually great since then
> >> a
> >> person can use "man -k" and/or "grep" for searching). "nroff" is good
> >> since
> >> there are utilities to turn it into other formats. Proprietary formats
> >> (from
> >> Microsoft) are very, very, very bad.
> >>
> >> Here are some of my own attempts to satisfy my desire:
> >>
> >> 1. Use Acrobat Reader to search the PDF manuals. Not too bad, but
> >> searching
> >> is slow and results in lots of false positives.
> >>
> >> 2. I've searched for a freeware program that reads Windows help files and
> >> has the ability to search individual help files. No luck.
> >>
> >> 3. I have searched on the web for a tool to convert Windows help files to
> >> HTML files (or nroff, tex-info, or anything) and have come up with very
> >> little.
> >>
> >> 4. Benno Sueselbeck converted nroff files to HTML and has them posted at
> >> this site which I find extraordinarily useful and easy to use (Insightful,
> >> please take note!)
> >>
> >>
> >> http://www.uni-muenster.de/ZIV/Mitarbeiter/BennoSueselbeck/s-html/shelp.html
Does that have permission from the copyright holders? Not obvious last
time I looked (some while ago).
> >> 5. With the upcoming near-synchronization of Windows/UNIX versions of
> >> S-Plus, would Insightful consider including source "code" for the help
> >> (nroff or whatever they're using now) in a directory on the CD so we can
> >> format it as we like best?
> >>
> >> Thanks for any ideas. I'll summarize useful information to S-news.
> >>
> >> Kevin Wright, Research Scientist
> >> Pioneer Hi-Bred Int'l, x4054.
> >>
> >> ---------------------------------------------------------------------
> > > This message was distributed by s-news@lists.biostat.wustl.edu. To
> >> unsubscribe send e-mail to s-news-request@lists.biostat.wustl.edu with
> >> the BODY of the message: unsubscribe s-news
> >
> >--
> >Frank E Harrell Jr Prof. of Biostatistics & Statistics
> >Div. of Biostatistics & Epidem. Dept. of Health Evaluation Sciences
> >U. Virginia School of Medicine http://hesweb1.med.virginia.edu/biostat
> >---------------------------------------------------------------------
> >This message was distributed by s-news@lists.biostat.wustl.edu. To
> >unsubscribe send e-mail to s-news-request@lists.biostat.wustl.edu with
> >the BODY of the message: unsubscribe s-news
>
> --
> --------------------------------------
> Don MacQueen
> Environmental Protection Department
> Lawrence Livermore National Laboratory
> Livermore, CA, USA
> --------------------------------------
> ---------------------------------------------------------------------
> This message was distributed by s-news@lists.biostat.wustl.edu. To
> unsubscribe send e-mail to s-news-request@lists.biostat.wustl.edu with
> the BODY of the message: unsubscribe s-news
>
--
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
|