2019:Libraries/Map making workshop – from Wikidata to interactive off-wiki maps in three steps
 | This is an Accepted submission for the Libraries space at Wikimania 2019. |
- Notes
- This session is a workshop and will take place from 14:00 to 15:30 on Day 3 (August 18) in Tu (B497).
- If you are interested to attend please add yourself here.
- To view the overall program of the Libraries space, please visit 2019:Libraries/Program.
Materials that will be used during the workshop[edit | edit source]
For entire workshop
- Workshop description on Github
- Full workshop outline & notes on Github
- Full workshop slides as pdf on Github, on Wikimedia Commons and on Slideshare
For Module 2
- Examples of embedded maps - https://nl.wikipedia.org/wiki/Gebruiker:OlafJanssen/KladblokMapMakingWorkshopWM2019
For Module 3
- Jupyter Notebook - https://paws-public.wmflabs.org/paws-public/User:OlafJanssen/WikidataMapMakingWorkshop.ipynb (and in raw format)
- Off-Wiki version of final map NetherlandsPublicLibrariesHeatmap.html
Title[edit | edit source]
Map making workshop – from Wikidata to interactive off-wiki maps in three steps
Description[edit | edit source]
In this workshop you will learn how to make both flat, layered, embedded and interactive maps from sets of geo-referenced (P625) items in Wikidata. For instance mapping public libraries in The Netherlands
We will do this in 3 modules of approx. 30 minutes each:
- Module 1: We will start by making various basic flat maps in Wikidata using SPARQL queries. (Libraries example). Next we will make some layered maps, where groups of items can be toggled on/off in the map. (Libraries example)
- Module 2: After having explored maps in the Wikidata query interface, we are now ready to learn how we can embed Wikidata-driven maps in other Wikimedia projects, such as Wikipedia and Commons (see the example on the right). In addition to SPARQL we will look at GeoJson, OpenStreetMap and the Kartographer extentsion.
- Module 3 : Finally, and if times allows, we will have a look at the steps for creating fully interactive Wikidata-driven maps that can be used off-Wiki, ie. in regular HTML pages. In addtition to the above tools & techniques, we will use some basic Python, Jupyter Notebooks (PAWS), Jupyter-Leaflet and Jupyter-widgets.
Intended audiences[edit | edit source]
While this session is tech-focused and will discuss basic Wikidata techniques and programming tools, it is meant to be approachable by beginning Wikidatians and programmers. The workshop leader, by no means an advanced Python programmer nor Wikidatian himself, will be showing and sharing examples and code snippets that can be easily adapted by users with basic SPARQL and Python skills, to make them work for their own datasets.
As the workshop is comprised of three 30 minute modules, participants can decide to skip the modules that they find too advanced (or basic) for their individual knowledge levels.
Required preparations[edit | edit source]
In order to maximize the effective workshop time, participants are requested to do some preparations before coming to the workshop:
- Identify a geo-referenced set of coherent items in Wikidata to work with. Aim for 100-500 items or so.
- Create an account on Wikidata/ Wikimedia, if not yet done so.
- Set up a working PAWS notebook (Help 1, Help 2, Help 3) – only needed if the participant intends to follow Module 3
Relationship to the theme[edit | edit source]
This session will address the conference theme — Wikimedia, Free Knowledge and the Sustainable Development Goals — in the following manner:
Data visualisation is a key skill to learn (SGD 4) for people in today’s data-driven and visual societies. It helps to translate abstract, dry data - typically only understood by experts - into visual end products that can be understood by bigger audiences, thus helping to democratize knowledge and reduce inequalities (SDG 10).
Session outcomes[edit | edit source]
At the end of the session, attendees will:
- Understand the steps to make basic flat and layered maps in Wikidata, based on geo-referenced items and SPARQL queries (Level 1, basic)
- Understand the steps to embed these maps in Wikipedia, Wikimedia Commons and other Wikimedia sites (Level 2, intermediate)
- Understand the steps to create Wikidata-based maps than can be used in other (non-Wiki) websites (Level 3, advanced)
- Will have access to map making resources, SPARQL examples and Python code snippets they can build upon for their own projects after Wikimania.
Session leader(s)[edit | edit source]
- Olaf Janssen
Usernames[edit | edit source]
Affiliation/country[edit | edit source]
- National library of The Netherlands & Wikimedia Netherlands
E-mail contact[edit | edit source]
- olaf.janssen@kb.nl
Session type[edit | edit source]
. The format of this submission is a:
- Computer-based training
- Workshop to identify and try to solve problem
Length of session[edit | edit source]
30/60/90 minutes, duration dependent on
- how much time the Space leaders are willing to provide
- how advanced & interactive participants will want to make the maps (see Levels 1,2 and 3 above, based on 30 minutes per level)
Please note that I proposed the same workshop in the GLAM Space on Friday and Saturday.
Supporting work[edit | edit source]
- Map materials for public libraries in The Netherlands - (work in progress)
- The map-related materials of User:Sic19, see map examples under Sparql-queries and GeoShapes
Requirements[edit | edit source]
The session will work best with these conditions:
- Room with projector for workshop leader. Wifi, seats, tables and wall outlets for participants bringing their laptops.
- Audience: 10-25 people, with basic knowledge about Wikidata, Sparql and Python (Jupyter Notebooks). Knowledge about GeoJson and OpenStreetMap will come in handy.
- Recording: Yes, recording allowed, happy to share knowledge