SNSF SPARK Workshop on Dynamic Data Ingestion

Session 2

05 May 2021

“Introduction by Kaspar Gubler

We are very pleased to have so many interesting projects and engaged participants in the workshop. And more participants have joined for session 2. For example, someone from Hamburg who used to do network analysis with the software Gephi and now wants to try out Nodegoat. A new participant is an archaeologist from the University of Bern who has documented sites in Excel and wants to import and visualise them in Nodegoat. Such data import is not difficult, especially if the data has been entered consistently in Excel. Another new participant plans to visualise cultural heritage with Nodegoat. A good example of how Nodegoat can be used for the presentation of digital, cultural heritage (thus also for art history) is the encyclopaedia on Romantic Nationalism: https://ernie.uva.nl/viewer.p/21/52/types/all/grid

Terminology

Before we start, I would like to remind you of the terminology of Nodegoat, in which we speak of Objects and Sub-Objects as well as Categories. We describe these Objects (column inExcel) in Nodegoat with Object descriptions (like rows in Excel). Object descriptions can be a text, a link, a picture or a link to another Object or a Category (= reference = relation). We can define in our data model the kind of description for each Object description. This gives us the possibility to describe an Object very precisely:

Find a common language

Important: if you want to communicate with another Nodegoat project it is very helpful if you use the terminology mentioned. So the first questions to another project would be: what Objects do you have? And how do you describe your Objects with what kind of Object descriptions? In which Sub-Object do you store your geo references? If you want to get in touch with other projects, you can organise your own zoom meetings on specific questions about Nodegoat. I see many projects that have a lot in common and could certainly benefit from an exchange. I would also like to draw your attention to the Nodegoat Day on 4 June, where you can present your project or your project idea.”

14:00 Welcome and recap of last week’s session

14:15 Object Type ‘Place’ Data Model + Data Entry

14:30 Object Type ‘Place’ Data Import

14:50 Break

15:00 Object Type ‘Person’ Data Import

15:30 Filter + Visualisation

15:50 Break

16:00 Scope & Visual Settings

16:15 Conditions & Export

16:35 Looking forward to next session

16:45 Q&A

  • Difference between Gephi and Nodegoat? → Nodegoat departs from data management + visualization functionality
  • Will there be a possibility to store Nodegoat Data in a data repository like Zenodo? → There are rumors about a Zenodo-Module in Nodegoat coming, currently it’s technically no problem to do it manually
  • How to download a “dump”?  → Via API, export dump of the data + of the model in JSON
  • Is it possible to export a complete project (instead of individual csv sheets)?  → Yes, via API you can export all of the data and the data model in JSON
  • How can I “undo” an import from csv when I notice that some things did not work as intended? Can I mass delete objects? → Yes, cou can mass delete objects via graphical interface, choosing all objects deleting them with the grey multi button, or delete in ‘Model’ the whole of a Object Type with clicking on ‘empty’ or mass delete mass objects via API
  • Can you import by just giving the URL of the Google Doc? → Yes, via API of Nodegoat, check what Google allows you to do via API
  • Can visualisations be downloaded in any way to the desktop? → Yes, Screenshot, or for high resolution use the ‘Capture’ functionality in the visualisation settings
  • Follow-up question to session 1:  can you create an itinerary of a person (object) with just knowing the sequence of the location but not the dates?→ Yes, with storing vague dates in Nodegoat, you make an statement in vague dates (‘Chronology’) like: ‘Studies came after Birth’. Or Yes, use as date: 1, 2, 3 etc. , or use the sequence identifier in a nodegoat date, so if you know a year use ‘1880 1’, ‘1818 2’, ‘1818 3’
  • We can include both a geometry (polygon) AND a precise coordinate in a sub-object? Or as separate subobjects of the same object? → Yes, , yes both options are possible! One geometry can be polygon + point(s) + line(s). Or each in a separate sub-object to be able to add attributes.
  • Are there any example projects that depicts more complex routes? → http://mnn.nodegoat.net/viewer.p/1/47/scenario/30/geo
  • Can you add your own icons to be displayed on the map? → Yes, in SVG format.
  • Nodegoat as Tool to visualise routes or itineraries? →  Yes
  • Is there also a method to show place-specific meta-information on the map instead of the person’s? →  Yes
  • In case of data model refactoring, how should we deal with the already inserted data? For example, if one wants to normalize repetitive data creating a new object type, how can he migrate the actual data to the new data model? Export + Transform + Import is the only way?  →  Yes, but because you now have nodegoat IDs, it’s a matter of a straightforward mapping. Or use an Ingestion process (session 4).
  • is it possible to mark a node with multiple conditions (e.g. one condition for people born in the low countries (orange) + people died in Italy (blue), so objects that fall in both categories marked in two colours)? →  Yes

Links

http://mnn.nodegoat.net/viewer.p/1/47/scenario/30/geo

Slides:

Download Google Sheets as CSV files:

RAG Places small selection: https://docs.google.com/spreadsheets/d/1zvcVj66nr1tm7PAmNJSSf2BI_o5e2rrPCSE2l4PAsHQ/
RAG People small selection:https://docs.google.com/spreadsheets/d/1K2SGF0TkQTVnZ5WQqgMc0MbJdGps1kA_oWVL3Qir6rs/

Guides:https://nodegoat.net/guides/csvfilehttps://nodegoat.net/guides/gazetteer

Another sample data Import:
https://histdata.hypotheses.org/nodegoat-tutorials
Tutorial No 10, to create this map (positions of ships):