On 09/08/11 16:28, Michael Lachmann wrote:
On 9 Aug 2011, at 4:59PM, Sam Liddicott wrote:
An alternative is to have a separate session for each document. This means that
you can't easily share values between documents. The main times I use this is
when I clean up a document, or sum it up - then I open a second, cleaned
document.
Sharing of sessions between documents scares me. I think because I don't want
to clean up a document. For this reason I generally avoid using shell sessions
- but a shell is probably much more stateful than an R session.
I'm not totally sure what you mean, but I think R sessions are as stateful as
shell sessions, if not more. (You have variables/libraries loaded etc, you have
you current directory, and sessions can be saved/loaded across restarts. They
are saved in the current directory....)
And you are right - it can be a big mess - within a document, while working,
the result of a calculation can easily appear before the calculation, because
you went up, or the calculation could have been erased. A lot of my time is
wasted hunting up and down a document for where something was calculated....
but it also has its benefits, I guess like GOTOs in programming have their
benefits.... hmm...
dear me.. it sounds like one needs to be very disciplined. I can see
that the R1 R2 macros you mention are useful while messing about but the
risks involved and the understanding required (which you clearly have)
to not get in a mess make me think that they should be personal
preferences and not global, and awkwardness of the steps to install
these into the personal preferences file is really a blessing.
But I work with TeXmacs more as a sophisticated terminal for R, not really as a
way to generate documents with R in them.
Yes... what gets me is that sessions do not behave like the texmacs
document. The document is like a program that is executed starting at
the top, and which has a side effect of typesetting a document. Changes
to the texmacs environment made near the top are visible near the middle
and at the end, and so on, but never the other way around.
But it seems that multiple R insertions using the same session can get
into a mess, but that texmacs does nothing to try and preserve the same
pattern of work - so I find sessions a good way to typeset a historical
log of something that was done, but with uncertainty that it is repeatable.
Sam
Michael
_______________________________________________
Texmacs-dev mailing list
Texmacs-dev@gnu.org
https://lists.gnu.org/mailman/listinfo/texmacs-dev
--
[FSF Associate Member #2325]
<http://www.fsf.org/register_form?referrer=2325>
<http://www.openrightsgroup.org/>
_______________________________________________
Texmacs-dev mailing list
Texmacs-dev@gnu.org
https://lists.gnu.org/mailman/listinfo/texmacs-dev
