The Sage notebook is run by typing
sage: notebook()
$HOME/.sage/sage_notebook.
Other options include:
sage: notebook("directory")
$HOME/.sage/sage_notebook. This can be useful if you
want to have a collection of worksheets associated with a specific
project, or run several separate notebook servers at the same time.
When you start the notebook, it first creates the following files in
$HOME/.sage/sage_notebook:
nb.sobj (the notebook SAGE object file) objects/ (a directory containing SAGE objects) worksheets/ (a directory containing SAGE worksheets).
After creating the above files, the notebook starts a web server.
A ``notebook'' is a collection of user accounts, each of which
can have any number of worksheets. When you create a new worksheet,
the data that defines it is stored in the worksheets/username/number
directories. In each such directory there is a plain text
file worksheet.txt - if anything ever happens to your worksheets,
or Sage, or whatever, that human-readable file contains everything needed to
reconstruct your worksheet.
From within Sage, type notebook? for much
more about how to start a notebook server.
The following diagram illustrates the architecture of the Sage Notebook:
----------------------
| |
| |
| firefox/safari |
| |
| javascript |
| program |
| |
| |
----------------------
| ^
| AJAX |
V |
----------------------
| |
| sage | SAGE process 1
| web | ------------> SAGE process 2 (Python processes)
| server | pexpect SAGE process 3
| | .
| | .
---------------------- .
For help on a Sage command, cmd, in the notebook
browser box, type cmd? and now hit <esc> (not
<shift-enter>).
See About this document... for information on suggesting changes.