SNSF SPARK Workshop on Dynamic Data Ingestion

Session 1

28 April 2021

Introduction by Kaspar Gubler

“Welcome to the Nodegoat SPARK workshop ‘Dynamic Data ingestion’ We are very happy about the 140 participants from all over the world. On the map, which was of course created with Nodegoat, we can see the places of origin of the participants. They come from very different disciplines: history, literary history, German studies, English studies, legal history, historical geography, art history, musicology, theatre studies, film studies, African studies, Islamic studies, sociology, digital humanities and also from archives and libraries. This impressively shows how Nodegoat has established itself in the last ten years as an international research infrastructure for the humanities, an interdisciplinary research infrastructure that helps digital research to gain new insights and more visibility – and facilitates the collaboration of projects, especially beyond one’s own subject boundaries. Pim van Bree and Geert Kessels began developing Nodegoat as part of a project at the University of Amsterdam in 2011. Pim van Bree has a Master’s degree in Media Studies, Geert Kessels a Master’s degree in History. Both are also accomplished software developers. Their particular strength is that they know both worlds, the world of humanities and the world of programming. They combine these two worlds in their workshops, which they conduct at educational institutions worldwide. With their deep knowledge of methods, sources and questions in the humanities, they can create fitting and working data models for the different disciplines in Nodegoat to extract new scientific information and knowledge from the data.”

Programme

14:00 Welcome by Kaspar Gubler

14:15 General introduction to nodegoat

14:40 Login and set up your nodegoat project

14:50 Break

15:00 Object Type ‘Person’ Data Model

15:15 Object Type ‘Person’ Data Entry

15:35 Classification ‘Capacity’ Data Model + Data Entry

15:50 Break

16:00 Object Type ‘Book’ Data Model

16:20 Object Type ‘Book’ Data Entry

16:35 Looking forward to next session

16:45 Q&A

  • I don’t have access to the ‘Model’ section in nodegoat → Check Page Clearence. Each nodegoat project has 1 administrator at the beginning, who can set up additional users. Management > Users > add User > add the user  and activate the ‘Model’ in the settings  for the page clearance (tab)
  • How does the Scope work? For Visualizations? → In the Scope you define which of your databasefields you want to use for the visualization, so you activate the field that contains the georeference wiht the coordinates
  • Are there facilities (planned?) helping to prepare a RDF rendition of the database? →Yes: on one’s own nodegoat installation, you can configure a translation module to translate the data model to some RDF vocabulary
  • Is it possible to export, as static or dynamic representation, a computed spatio-temporal / network analysis ? →Yes
  • Würden Sie sagen, dass sich Nodegoat grundsätzlich auch als Bilddatenbank (mit Zusatzbeschreibungen und Querverweisen) eignet? →Yes, absolutely, see the links below.
  • Regarding custom gazetteers and prosopographies: Are there size limitations? → Not in general, size limit of CSV import is set to 60’000 rows at a time
  • Is it better to store a region like Germany as JSON or via Reference >city > autofill option Germany? →It depends on what you want to show on your map, if you are more interested in areas it’s Geo JSON, if you are working in general with dots on your maps, it’s maybe better to store it as a point like your other data
  • Can we import polygon data from an existing map with territorial circumscription so that we don’t have to draw them by hand? → Yes
  • Can we specify a schema or other constraints to ensure consistency of the data (e.g. Birthdate < Deathdate, no overlapping residence periods etc.)? → You can use visualisation to do some error checking, but there can be no hard enforcement of such constraints at the moment. Or you can filter specifically on Birthdate < Deathdate, more advanced.
  • Does database harmonise the different sources of location, e.g. if I put “Roma (IT)” from “City” and also point from “Geometry” which is actually Rome, will database understand it is the same? On the map it looks the same, but how is it in database? Is it linked as the same entity? → More on this next session!
  • Is Arabic script supported or more generally, are scripts running from right to left supported? → Yes: Everything Unicode
  • Can we geo-visualize more than one object type, say author’s places and book’s publishing houses’ places? → Yes: using the Scope (https://nodegoat.net/guides/visualisationsettings)
  • Is there anything in the guides as regards Nodegoat and RDF? → Yes, it’s work in progress, see this sample nodegoat project working with a Subject-Predicate-Object (RDF Format) data model: https://www.manto-myth.org/blog/a-half-dozen-ways-to-die-mythically
  • If I uncheck „Fixed field“ in Object „book“ it throws: The data Model does not have a configuration that can be used to generate Object names, please check your settings.“ Why? → Tick/check one or more of the Object Description ‘name’ (‘use object description for name’) checkboxes
  • Can nodegoat handle localised object descriptions? E.g. book reviews in different languages? → Yes
  • Can nodegoat handle uncertain dates / data? → Yes, see the following blogposts:

https://nodegoat.net/blog.s/45/how-to-store-uncertain-data-in-nodegoat-ambiguous-identities

https://nodegoat.net/blog.s/44/how-to-store-uncertain-data-in-nodegoat-conflicting-information

https://nodegoat.net/blog.s/43/how-to-store-uncertain-data-in-nodegoat-incomplete-source-material

https://nodegoat.net/blog.s/42/how-to-store-uncertain-data-in-nodegoat

      Slides