Closed kohlhase closed 7 years ago
assigning to @Jazzpirate so he can supervise Marcel, who should do the interface work.
Here is another feature this should eventually have: According to the CICM paper a model is a subgraph of the universal "Physics" graph we want to build, so the we want to have a special "add node" functionality on the MPD viewer. This "add submodel" or "add path", or so - we should ask the WIAS people eventually - should be available on any leaf (in the MPD sense) quantity and should offer all theories that could be connected to the node (i.e. present in the physics graph, but not in the current MPD). The user selects one and that is added to the MPD.
We are currently writing a paper about models, and it would be great if we could have (a first version) of this viewer quite soon.
I suggest that the contents of the nodes are obtained by callback URLs for which Dennis makes special MMT-based web service.
this is done, now it only needs to be integrated again in TGView. See https://github.com/UniFormal/TGView/issues/1
At WIAS we have just invented the notion of a "Model Pathways Diagram" (MPD), which is essentially a special view on the model theory graph. Currently http://gl.kwarc.info/mkohlhase/mathmodels/doc/cicm17/ has a MPD for the van Roesbrook model (vRM; see http://gl.mathhub.info/MitM/models/), and we would like to have a version of Marcel's graph viewer that generates that on the fly. We have added metadata to the vRM theories:
modeltype physical_quantity
for quantity theories --> they should appear as circles with only the notation of the main symbol showingmodeltype physical_law
for laws --> they should appear as normal theory boxes with a name and the main law presented in themProbably (we have to play with the viewer) all other theories should not be shown at all. All the specially treated theories and links should have functionality that gives their full details.