[Date Prev]
| [Thread Prev]
| [Thread Next]
| [Date Next]
--
[Date Index]
| [Thread Index]
Re: [xml-dev] Incongruous UML data models and XSD data models
- From: Michael Kay <mike@saxonica.com>
- To: Peter Flynn <peter@silmaril.ie>
- Date: Mon, 27 Jun 2016 22:27:30 +0100
The "nested boxes" graphical representation of a hierarchy can sometimes be useful, and it has the advantage of providing a direct visual representation of a "containment" relationship. In dealing with document structures, some people will recognize its use in "clause analysis".
One problem (but that's true of most graphical representations of trees) is that it doesn't scale well - you quickly run out of paper.
However, the tree metaphor for a data structure is very readily understood and very ancient - it goes back at least to the tree of Jesse in Isaiah.
Michael Kay
Saxonica
> In trying to teach [about] XML, I think we all tend at some stage to
> have used the tree as a reference, which is a pity, because the
> classical "family-tree" diagram of XML is upside down: really more of a
> root-system.
>
> Near&Far did at least present its diagram sideways, but I think that may
> have been because it's more usual to have XML documents with greater
> width (siblings, represented vertically) than depth (descendants,
> represented horizontally). If you have to start adding non-element
> nodes, the whole tree-metaphor becomes unreadably dense.
>
> How are most of the people we want to present a graphical model to
> at understanding different-shaped boxes joined by lines anyway?
>
[Date Prev]
| [Thread Prev]
| [Thread Next]
| [Date Next]
--
[Date Index]
| [Thread Index]