Hi Ralph & Irene, Thank you for a thoughtful discussion. That was a great insight regarding inherited property shapes. I will implement the mapsToClass and mapsToProperty pattern. This should enable me to provide the required user experience and to build some consistency checks.
Also, thank you for the recommendation example. I always struggle with business term vs formal ontology. In this case, I chose an ontology because I sometimes turn the data into triples so I can use SHACL to validate the data, try to answer business questions, etc... It's an on-going experiment as to what to use when. I'd really like to create a wiki-like environment where I can crowd source knowledge around the classes or instances of the ontology but that's another project for another day. Oh, thanks for pointing out the misuse of edg:foreignKey. I'm still learning the EDG ontologies and discovered my error when I went to implement my diagram. Ooops :) Tim On Thu, Apr 16, 2020 at 10:56 PM Ralph TQ [Gmail] <rhodg...@topquadrant.com> wrote: > Tim, > > Having thought some more about how property shapes are inherited by > subclass a mapping would only resolve to the right class if it were a leaf > class. > > I prefer what Irene is proposing, the combination of the class and > property will ensure that a mapping is properly resolved. > > Ralph > > On Apr 16, 2020, at 5:23 PM, Irene Polikoff <ir...@topquadrant.com> wrote: > > There are two properties edg:mapsToProperty and edg:mapsToClass that could > be used. mapsToClass can be used to point from a table or a dataset to a > class. So, between these two properties you would have the contextual > information that is specific to a class. Thus, I do not see a need to map > to the property shape. Further, the property shape for a property that is > broadly used would most likely be declared for some common parent class and > inherited down. It would typically not be advisable, for example, to repeat > a shape for description for every class that may have a description. > > Also note that unlike edg:mapsToTerm, there isn’t any special logic in EDG > for automatically suggesting connections to classes or properties based on > the data profiling and sampling. The suggestions currently work only for > linking data assets to glossary terms. Primarily because most people have > glossaries for business definitions, not formal ontologies. But the > algorithm does use shape definitions. > > For example: > > > -- > You received this message because you are subscribed to the Google Groups > "TopBraid Suite Users" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to topbraid-users+unsubscr...@googlegroups.com. > To view this discussion on the web visit > https://groups.google.com/d/msgid/topbraid-users/6C38CAF1-B7FD-4578-83E5-725C49E48EA5%40topquadrant.com > <https://groups.google.com/d/msgid/topbraid-users/6C38CAF1-B7FD-4578-83E5-725C49E48EA5%40topquadrant.com?utm_medium=email&utm_source=footer> > . > <PastedGraphic-1.pdf> > > Employee ID rule is a shape. It is an instance of edg:edg:DataValueRule > which is a subclass of sh:NodeShape. > > Then you can get a recommendation > > > -- > You received this message because you are subscribed to the Google Groups > "TopBraid Suite Users" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to topbraid-users+unsubscr...@googlegroups.com. > To view this discussion on the web visit > https://groups.google.com/d/msgid/topbraid-users/6C38CAF1-B7FD-4578-83E5-725C49E48EA5%40topquadrant.com > <https://groups.google.com/d/msgid/topbraid-users/6C38CAF1-B7FD-4578-83E5-725C49E48EA5%40topquadrant.com?utm_medium=email&utm_source=footer> > . > <PastedGraphic-2.png> > > Also, in your diagram you used a column as a value of edg:foregnKey > property. Values of this property are instances of edg:ForeignKey class > which is not a column. Foreign keys can be combinations of columns. > > > -- > You received this message because you are subscribed to the Google Groups > "TopBraid Suite Users" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to topbraid-users+unsubscr...@googlegroups.com. > To view this discussion on the web visit > https://groups.google.com/d/msgid/topbraid-users/6C38CAF1-B7FD-4578-83E5-725C49E48EA5%40topquadrant.com > <https://groups.google.com/d/msgid/topbraid-users/6C38CAF1-B7FD-4578-83E5-725C49E48EA5%40topquadrant.com?utm_medium=email&utm_source=footer> > . > <PastedGraphic-3.png> > > > > On Apr 16, 2020, at 4:11 PM, Tim Smith <smith.ts...@gmail.com> wrote: > > Hi Ralph, > > Should edg:mapsToProperty point to the property or the property shape? > For example, in the case of a "generic" property like description that will > be used across many class definitions, such that: > > :Product :description "product description" *mapsTo* Product Table ---> > Description Column for Product (ProdDesc) > > :Material :description "material description" *mapsTo* Material Table > ---> Description Column for Material (MatDesc) > > In this case, I can point both ProdDesc and MatDesc to :description using > edg:mapsToProperty but if I walk from Product to the Product-Description > property shape to the :description property, I won't know which column I > actually need for Product vs Material. > > Thanks, > > Tim > > On Thu, Apr 16, 2020 at 1:48 PM Ralph TQ [Gmail] <rhodg...@topquadrant.com> > wrote: > >> There is a property “mapsToProperty” defined in the SCHEMA EDG Data >> Assets that allows you to make the mappings. >> >> Here is how it is defined: >> >> <Screen Shot 2020-04-16 at 1.45.25 PM.png> >> >> And here is the SHACL property constraint that makes use of it, placing >> the values in a property group for "Traceability": >> >> edg:DataElement-mapsToProperty >> rdf:type sh:PropertyShape ; >> sh:path edg:mapsToProperty ; >> rdfs:isDefinedBy <http://edg.topbraid.solutions/1.0/schema/datamodels> >> ; >> sh:class rdf:Property ; >> sh:group edg:TraceableGroup ; >> sh:not tosh:SystemNamespaceShape ; >> sh:order 90 ; >> . >> >> >> >> On Apr 16, 2020, at 1:39 PM, Tim Smith <smith.ts...@gmail.com> wrote: >> >> <image.png> >> >> >> >> -- >> You received this message because you are subscribed to the Google Groups >> "TopBraid Suite Users" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to topbraid-users+unsubscr...@googlegroups.com. >> To view this discussion on the web visit >> https://groups.google.com/d/msgid/topbraid-users/9C3DCD70-169F-40A3-A452-653506F7CA3C%40topquadrant.com >> <https://groups.google.com/d/msgid/topbraid-users/9C3DCD70-169F-40A3-A452-653506F7CA3C%40topquadrant.com?utm_medium=email&utm_source=footer> >> . >> > > -- > You received this message because you are subscribed to the Google Groups > "TopBraid Suite Users" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to topbraid-users+unsubscr...@googlegroups.com. > To view this discussion on the web visit > https://groups.google.com/d/msgid/topbraid-users/CAF0WbnLAxHcigaHpTtk6h-Jn2k1E%2BHYR0NzgJyBJr_Gficqj_A%40mail.gmail.com > <https://groups.google.com/d/msgid/topbraid-users/CAF0WbnLAxHcigaHpTtk6h-Jn2k1E%2BHYR0NzgJyBJr_Gficqj_A%40mail.gmail.com?utm_medium=email&utm_source=footer> > . > > > > -- > You received this message because you are subscribed to the Google Groups > "TopBraid Suite Users" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to topbraid-users+unsubscr...@googlegroups.com. > To view this discussion on the web visit > https://groups.google.com/d/msgid/topbraid-users/6C38CAF1-B7FD-4578-83E5-725C49E48EA5%40topquadrant.com > <https://groups.google.com/d/msgid/topbraid-users/6C38CAF1-B7FD-4578-83E5-725C49E48EA5%40topquadrant.com?utm_medium=email&utm_source=footer> > . > > > -- > You received this message because you are subscribed to the Google Groups > "TopBraid Suite Users" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to topbraid-users+unsubscr...@googlegroups.com. > To view this discussion on the web visit > https://groups.google.com/d/msgid/topbraid-users/1FCCBA0B-0C5C-4C10-93D9-9C2E78855632%40topquadrant.com > <https://groups.google.com/d/msgid/topbraid-users/1FCCBA0B-0C5C-4C10-93D9-9C2E78855632%40topquadrant.com?utm_medium=email&utm_source=footer> > . > -- You received this message because you are subscribed to the Google Groups "TopBraid Suite Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to topbraid-users+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/topbraid-users/CAF0WbnLYJcwZhqK0cEU2SsRJd7DPN9_Amak7j1s8qgNVC936Tw%40mail.gmail.com.
