Webel ADTs include inheritance (and aren’t actually formal ADTs)

The term Abstract Data Type (ADT) has been "hijacked" here from its original meaning as popularised for use in Mathematica by Roman Maeder, and the implementation is (heavily) adapted from one proposed by Roman Maeder.

Dr Darren says:

I might just have to live with people thinking I don't know what an Abstract Data Type (ADT) actually is. When I started working on this the term had some tradition in Mathematica, and it was a convenient way to distinguish ADT "pseudo classes" from the MTools classes I was otherwise using. And 'A', 'D', and 'T' are close together on a QWERTY keyboard.

Whether a particular Webel ADT actually counts as a formal ADT depends primarily on how the operations are implemented (and whether they realise formal ADT constraints).

In any case, where it appears here prefixed as Webel ADT it does not mean a formal Abstract Data Type (ADT), it just means a stateless pseudo class or strong type implemented using TagSetDelayed in the Wolfram Language.

Some departures from formal Abstract Data Type (ADT) usage are indicated by custom stereotypes with icons in the Webel SysML4Mathematica (v1) modelling:

Visit also: Mathematica: Webel ADT pseudo classes: The SysMLv1 models are supported by custom stereotypes with icons. Some of these indicate OO-similar use of ADT-Methods and departures from formal Abstract Data Type (ADT) definitions.

There are some external links below for those curious about the formal computer science and maths definitions of Abstract Data Type (ADT).

Up next

Webel ADT hierarchy views as tree, graph, and SysML

Notes

[EXPLANATION]{INFORMATIVE} Mathematica: Webel ADT: The term 'Abstract Data Types (ADT)' is used informally (the Webel ADTs do not always adhere to strict definitions of ADTs). Please just think of it here as meaning a strong type or stateless "pseudo class".

[ASSERTION]{STRONG} When domain modelling with SysML, Generalization, inheritance, and even multiple inheritance are not "evil" and can be used relatively freely. We are not doing code-oriented Decorator patterns - an implementation choice - we are doing domain modelling!

[TIP]{STRONG} Mathematica: Webel ADT pseudo classes: TIP: Prefer the Decorator Pattern over multiple inheritance of ADT-Method UpValues! (But not because "inheritance is evil" per se, it isn't.) The base ADT libraries make heavy (and very useful) use in inheritance.

Snippets (quotes/extracts)

Visit also

Mathematica: POLICY NOTES for the Webel Abstract Data Type (ADT) pseudo classes with inheritance for the Wolfram Language

Dr Darren of Webel IT Australia makes the case for better object-orientation for Mathematica using pseudo class ADTs without losing the super-functional power of the Wolfram Language