Especially like the coffee filter example graph! 🙂 And for the first time ever, I saw the *real* purpose of # anchor chars in IRIs explained. Even the RDF(S) specifications simply use them without explaining this background. Great!
In the example of Coffee the verb part "born in" could have multiples answers. I ask if would be a better expression for restrict the answer (concept). I saw an example about using Neo4j that the matter was basket ball players and an interesting example: "A PLAYER" "played for" "A team". and the "played for" had an attribute of "contract value" and this attributes lives on the link and not on the node. I had never saw that and would like when it is a case of putting attributes on links and when put on nodes. Does exist some king of normalization?
I love the topic and the presentation seems thorough. However, as someone with background in programming, it feels a bit long winded, and I think that introducing the RDF prefixes earlier would have made the slides easier to read.
Creating IRIs, are point 4 and 5 contradictory? Don't use http(s) unless there is a real webpage on the url. Otherwise use none as the scheme? Or skip the iri grap and simply use the package name thing from Java. At the top level there could be one namespace, so labels are short and human readable.
Very comprehensive. Thank you!
Especially like the coffee filter example graph! 🙂 And for the first time ever, I saw the *real* purpose of # anchor chars in IRIs explained. Even the RDF(S) specifications simply use them without explaining this background. Great!
😃Very interesting lecture
Thanks Professor, very informative lecture
Dankeschön :)
absolutely fantastic
Geile Vorlesung Prof!
In the example of Coffee the verb part "born in" could have multiples answers. I ask if would be a better expression for restrict the answer (concept). I saw an example about using Neo4j that the matter was basket ball players and an interesting example: "A PLAYER" "played for" "A team". and the "played for" had an attribute of "contract value" and this attributes lives on the link and not on the node. I had never saw that and would like when it is a case of putting attributes on links and when put on nodes. Does exist some king of normalization?
I love the topic and the presentation seems thorough. However, as someone with background in programming, it feels a bit long winded, and I think that introducing the RDF prefixes earlier would have made the slides easier to read.
Creating IRIs, are point 4 and 5 contradictory? Don't use http(s) unless there is a real webpage on the url. Otherwise use none as the scheme?
Or skip the iri grap and simply use the package name thing from Java.
At the top level there could be one namespace, so labels are short and human readable.
Brown Frank Walker Charles Robinson Scott