edges do not persist

[IMP1]

Edges are removed and added again. As such their labels are not 'unchanged', and only contain what is in the resultant graph. So the below table, which is nice and consistent, needs rethinking.

	Graph	Rule Match Graph	Rule Result Graph (New)	Rule Result Graph (Existing)	(How to achieve)
empty label	empty	any	empty	unchanged	empty
empty label value	void	any	empty	unchanged	void
empty label markset	unmarked	any	empty	unchanged	unmarked / ¬

Edges in the rule's result graph are never 'existing' edges. They're always new, and so empty labels for them will always be empty.

Is this a problem? That their markset cannot be maintained (because their value can).

[IMP1]

Yeah, so maybe an empty label should always either mean any (in a rule match graph), or otherwise empty.

Edges are removed and added again. As such their labels are not 'unchanged', and only contain what is in the resultant graph. So the below table, which is nice and consistent, needs rethinking.

	Graph	Rule Match Graph	Rule Result Graph
empty label	empty	any	empty
empty label value	void	any	void
empty label markset	unmarked	any	unmarked

[IMP1]

If the user doesn't care what value a label has, and want to leave it alone, it seems a faff to go to the effort of declaring a variable (of type any), just to maintain the edge's value.
On the other hand, given that edges are not maintained at all, it seems confusing to have maintaining a label's value be the default for nodes, but not for edges (since the new edge will have an empty label if unspecified).

[IMP1]

Why edges can't persist

They're not directly referenced to. One edge between nodes 2 and 4 is indistinguishable from another. Consider this rule:

rule r1 
    [ 1, 2 | 1->2, 1->2 ]
 => [ 1, 2, | 1->2(<maintain>) ];

Which edge's label is chosen to carry over? This would require variables for both the original edges' labels. And then there's no way to keep their markset. This seems like a fundamental problem with marksets, or their lack of first-class status.