MathGroup Archive 2004

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: Re: Conveniently Restarting Notebooks

  • To: mathgroup at smc.vnet.net
  • Subject: [mg51233] Re: [mg51150] Re: [mg51114] Conveniently Restarting Notebooks
  • From: "David Park" <djmp at earthlink.net>
  • Date: Sat, 9 Oct 2004 04:19:00 -0400 (EDT)
  • Sender: owner-wri-mathgroup at wolfram.com

Tom,

Various people have their methods of using notebooks and it's not for me to
say that some person's method that works for them is wrong. Still I often
see some practices that appear to me to make things more difficult rather
than easier.

The object is to use notebooks in a way that simplifies one's own work and
facilitates communication with other people. Complicated schemes usually
frustrate this purpose. When using Mathematica notebooks as a means of
communication, you can't expect that readers will understand any complicated
nonstandard scheme no matter how nifty it may seem to you.

The question is: Do you want to get to the subject matter, or do you want to
manage a complicated notebook scheme? (I like to watch opera and ballet and
one of my pet peeves is excessive panning and zooming of the camera. Fred
Astaire used to say that either the camera dances or he dances but not both.
Do you want your discussion and calculations to do the dancing or do you
want the notebook style to do the dancing?)

I think that every notebook should be self contained. It may depend on
packages but shouldn't depend on other notebooks. The only time I have two
notebooks open at once is when I'm copying some material from one notebook
to another, or when I'm changing a package notebook and testing with a
regular notebook.

Use only one each, at most, of the various title cells and put them at the
top of the notebook. Use Sections as the main organizational method. For
discussion and comments use Text cells, not the various types of Section or
Title headers.

I've had cases where new output cells did not overwrite the old output
cells. But it is very rare and I'm not certain what the cause is. I usually
just delete all the output cells and reevaluate and the problem goes away. I
usually delete all output cells before I save a notebook and I think the
problem may occur more often when a notebook is opened with existing output
cells. Maybe other people have experience on this.

I wouldn't use a lot of font colors in input cells, or a lot of color coding
in notebooks. Is the color coding actually adding syntatical information or
is it just decorative? Color schemes are not natural or intuitive in the
sense that certain colors mean certain things - especially technical things.
It's rather a burden to maintain, probably stresses the FrontEnd and parser,
and will probably be lost on readers of your notebooks. It's an example of a
notebook style dancing.

I don't know what to say about your last problem. I must work on simple
things because I never have an Input cell that extends over multiple
screens. Do you actually type in a single multiple screen input cell? It
sounds like it would be more managable if it could be broken into pieces.
Functions of functions or something like that.

David Park
djmp at earthlink.net
http://home.earthlink.net/~djmp/


From: Tom Aldenberg [mailto:Tom.Aldenberg at rivm.nl]
To: mathgroup at smc.vnet.net


Hope this is not off the point too much.

I never understood why the notebook front end allows you to play in
multiple windows with the same kernel, to jump around within each of them,
and between them, yielding a nonlinear sequence of Ins and Outs, and why
the calculation order, nor which (other) notebooks are involved in the
project, is not saved at all, hence cannot be recovered.

The front end is fantastic of course for scientific communication. I have
still some complaints.

1. Once in a while the In-Out combinations go astray, and the corresponding
Out of an In is not overwritten, but pushed down. Don't know when/why that
happens. I can't get rid of it. It spreads through the whole notebook.

2. I started using mixed color font coding within identifiers, say to
highlight word segments like Global or Default. After a while tiny white
space gets between the colors, and the Find (Ctrl + F) utility cannot
retrieve the whole words anymore. Had to stop doing that. Find needs more
options, whole words, formatting, color.

3. I create input cells spreading over multiple screens, extensively using
Maeder's Class mechanism. If you edit inside these inputs, unbalanced
parentheses or brackets cause the text to scroll unpredictably and the
cursor to jump from top to bottom to the middle and so forth. Each time you
have to find back the cursor. Then you type the closing bracket and bang
again. Gone. If you have to do that for hours you go completely nuts. I
like the horizontal scrolling with unmatched parentheses, but would like to
see the thing stabilized in the vertical direction, when  you are just
typing within a line.

Regards,
Tom

Tom Aldenberg
RIVM
PO Box 1
NL-3720 BA Bilthoven
Netherlands
T: +31 30 2743137
F: +31 30 2744413
E: tom.aldenberg at rivm.nl






  • Prev by Date: Re: Re: Problem with Maximize and conditions.
  • Next by Date: Generating r-combination or r-permutations with replacement.
  • Previous by thread: Re: Re: Re: Conveniently Restarting Notebooks
  • Next by thread: Newbie facing $RecursionLimit and Stack Space on OSX