I propose to extend the functionality of spreadsheets to create connections between objects:
If a connection type is allocated to a pair of columns, then this connection is created between pairs of objects in the same row of those columns.
E.g. we would like to link Processes and Business Objects, which are currently managed as 2 independent referential. Mapping tables turned out to be much more efficient that to add the business objects to each process diagram.
Column A |
Column B |
Column C |
Cluster 1 (occurrence) |
Business Object A (occurrence) |
Object State a (occurrence) |
Cluster 2 (occurrence) |
Business Object A (occurrence) |
Object State b (occurrence) |
Cluster 3 (occurrence) |
Business Object B (occurrence) |
Object State c (occurrence) |
... |
... |
... |
...and allocation of a connection type to columns (A,B) and columns (B,C), and possibly columns (A,C).
Background:
This is an efficient alternative to diagrams or matrices, to define connections that represent mostly 1:1 or 1:few relations, and have no benefit of a graphical representation in a diagram.
The business often provides such mappings inputs in Excel/Google sheet format.
Modelling of connections of triplets (or higher) that are not in the semantics of standard model types is sometimes difficult in diagrams.
(e.g. Cluster <-> Business Object <-> Object State, or accountable Person<->Role<->Domain)
Often such connections have a describing character for one of the objects (i.e. it is a normalization, as alternative to attributes with value lists). So the spreadsheet would allow to define an object with a mix of attributes and connections in the same spreadsheet row.
PS: I thought I proposed this idea already about 1 year ago, but could not find it.
Hi Christian,
thank you for sharing your idea. To verify, if I got it right.
An additional function in the spreadsheet editor that enables you to create a connection between the object in cell X and the object in cell Y would do the trick?
Philip