Pile or Rhizome? A Short Pamphlet.

Erez Elul, whose honor it is to have “discovered” or “invented” the data structures I deliberately continue to call rhizomes, originally named his invention pile (or pile system). The relatively few authors (for instance Peter Krieg, Ralf Westphal, Ralf Barkow, Miriam Bedoni as well as others besides them), who both commented and contributed on the invention accepted this naming. I must say I never found the term particularly saying for several reasons.

• In my opinion the most significant characteristic of piles/rhizomes is universal connectivity. Everything can be connected to everything else. Quite differently, Erez Elul’s reasoning apparently focused much more on the fact that relations are “piled up onto each other” - hence the choice of term. In my eyes, this is however not the distinguishing criterion. For example, also in binary trees nodes are “piled” onto each other in a sense. Yet, unlike piles/rhizomes, binary trees follow the traditional node/edge dichotomy. They lack the feature of universal connectivity.
• The term pile is quite close to heap, and both could be confused easily. Runtime environments for programming languages such as the Java Virtual Machine (JVM) internally use a datastructure called heap to hold objects in memory.
• When I first by chance stumbled upon the term rhizome in Gilles Deleuze’s and Félix Guattari’s thinking, which they had borrowed from botanics, I immediately found the term very fitting.

Deleuze and Guattari use the term rhizome throughout their work, especially in their discussion of thought in "[A Thousand Plateaus](http://www.amazon.com/gp/product/0816614024)." They argue that traditional thought is tree-like, in that it follows a linear pattern, branching off at various points. Rhizomes, taken from a kind of [root system](http://en.wikipedia.org/wiki/Rhizome) found in nature, are non-linear, and non-hierarchical.

(Source)
Rhizomes are thus a philosophically well described concept, and a corresponding theory in computer science could possibly profit from earlier work done by these post-structuralists.

Of course there is no ultimate truth in either name choice, and there are good reasons to stick to the customary term pile.