Note Mathieson and Scally definition is almost a gARG

[jeromekelleher]

We need to note somewhere in the eARG section (or somewhere) that Mathieson and Scally definition is very similar to a gARG, except:

• The don't define how trees are embedded, no mention of breakpoints etc
• The have one node per pedigree individual, not two (e.g. per genome)
• The say they only have nodes where events occur

[jeromekelleher]

I think we're capturing this now by saying that our treatment is just a clarification of theirs, adding some details required for practical purposes.