DDLab's user interface allows setting, viewing and amending network parameters by responding to prompts or accepting defaults. The prompts present themselves in a main sequence and also in a number of context dependent prompt windows. It is possible to backtrack to previous prompts in the sequence, and in some cases to skip forward. A flashing cursor indicates the prompt. ``Return'' (or the left mouse button) steps forward through the prompts, ``q'' (or the right mouse button) backtracks, or interrupts a run.
CA rules (or totalistic codes) are set in decimal, hex, as a lookup table bit pattern (using the mouse), at random or loaded from a file. The ``game of life'' and the ``majority`` rule can also be selected. Rules may be changed into their equivalents (by reflection and negative transformations), and transformed into equivalent rules with larger neighbourhoods. The latter is useful to achieve finer mutations of interesting rules. The frequency of canalyzing ``genes'' or inputs in a network can be set to any arbitrary level, useful in modelling gene regulatory networks.
Wiring is set as regular 1d or 2d for the given neighbourhood size, at random, with a ``spread sheet`` to set or alter individual couplings by hand, or a with wiring scheme may be loaded from a file. In most cases regular 2d wiring defines a square grid on the torus, and includes the von Neumann and Moore neighbourhoods of 5 and 9 cells. However the 7 cell regular 2d neighbourhood is wired to define a triangular grid. Non standard wiring can be constrained in various ways, including confining it to a local patch of cells with a set diameter.
The network parameters can be scrutinised and amended in a 1d or 2d graphic display format, or a ``spread sheet`` format, showing each cell's particular wiring and rule, using the arrow keys to move between cells.
An initial network state is required as a seed for a forwards or backward run (to seed a sub-tree or single basin). A basin of attraction field does not require a seed. The seed is set in decimal or hex, as a bit
pattern in 1d or 2d (using the mouse), at random, or loaded from a file. The bit pattern method is a mini drawing program, using the mouse (or keyboard) to set cell values (0,1). It is particularly useful for 2d patterns.
It is possible to create sets of independent sub-networks within the total network*. This is done by saving smaller networks parameters to a file, then loading them at appropriate positions in the total network, tiling a network with sub-networks in the case of 2d. Alternatively a network can be automatically duplicated to create a total network made up of two identical sub-networks* (repeating this results in 4,8,etc). This is useful to see the difference pattern, or damage spread, between two networks from similar initial states. Sub-networks created by either method may be re-wired to influence each other, for instance to create a network of (sparsely connected) sub-networks.