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

Re: "Anti-Comments"?

• To: mathgroup at smc.vnet.net
• Subject: [mg69193] Re: "Anti-Comments"?
• From: AES <siegman at stanford.edu>
• Date: Fri, 1 Sep 2006 06:41:35 -0400 (EDT)
• Organization: Stanford University
• References: <ed68rf$jg7$1@smc.vnet.net>
• Sender: owner-wri-mathgroup at wolfram.com

In article <ed68rf$jg7$1 at smc.vnet.net>,
 "David Park" <djmp at earthlink.net> wrote:

> I want to respond to this but I'm not quite certain of the best response! So
> this will be more in the line of musing.  What I
> think you are really getting after is a tighter integration of textual
> material and active definitions and commands. 

Yes, exactly right.  

And more broadly, rather than doing calculations and preparing graphs in 
Mathematica, then copying results from it over into a separate word processing 
document, in many (in fact, most) cases I'd like to have just one master 
Mathematica notebook for some project or problem or topic, which would contain 
pretty literally **everything**, from initial analyses through numerical 
calculations and graphing, and on to transmitting results as reports and 
even slide presentations to co-workers, colleagues, sponsors, and other 
audiences.

Such a notebook, with all cells fully open, would contain and display 
pretty much *everything* I did on the problem.  

A lot of useful project-specific utility modules for calculation and 
display would like be defined at the start of the notebook, then hidden 
from view (cells closed) so I don't have to scroll over them as I do 
further work, or other readers don't have to wade down through them.

[This could of course be done with Packages -- but I want everything 
*right in the notebook*, rather than creating, re-creating, and 
navigating among other files every time I decide to modify a module a 
little bit.]

As final forms for equations or display modules were developed there 
might be test cases that took a long time to run.  Once these tests came 
out OK, the test results could be deleted but the test routines hidden, 
so that they would be there if I needed to go back to them.

Included at the start of some sections might be initial analyses and 
algebraic messing around that I did to get formulas and equations into 
forms that I liked; these would be hidden once completed, but still 
there if I needed them.

As the project develops I'm very likely to be inserting new and more or 
less free-standing sections in the notebook which are pretty much 
independent of definitions and calculations in earlier sections.  I'd 
like to be able to disable those earlier sections so I can just use 
cmd-A, Enter to run the emerging section repeatedly as I edit it, with 
no selections or mouse clicks.

Some "final" results (graphs with many points, or long iterative 
calculations) might take a long time to run.  Once they ran 
successfully, I'd freeze the result and disable the calculation routine 
-- unless I needed to run it again, perhaps for new variable values.

Depending on the intended audience I will likely want to print the 
notebook (or maybe just a few sections of it) to PDF, with varying 
levels of content included and suppressed -- e.g., headings, text, major 
equations, figures and tables for a report, but just figures and tables 
(one per page) for presentation slides.  

Some of the figures will get Exported to PostScript or PDF; TeX or LaTeX 
or Word will get fired up; and I'll start Copy and Pasting or 
re-keyboarding only if the target audience is a journal ms or technical 
meeting submission that requires those formats and is to cover only 
selected portions of the material in the notebook.

And in all of this, I'd like the notebook to stay as "readable" -- as 
"ordinary notebook looking" in its structure as possible -- but also 
have the variations in displayed or active sections be controlled by 
flags set in the opening few cells of the notebook.

These general considerations have been the source of a lot of my 
questions to this group over time.