[[WM:TECHBLOG]]

The Query Service Tutorial for Wikidata: an interview with Dr. Keren Shatzman from Wikimedia Israel

With the Wikidata Query service tutorial, Wikimedia Israel aims to empower new users to query Wikidata, which currently holds almost 90 million items.

By Dr. Keren Shatzman, Wikimedia Israel and Elisabeth Giesemann, Wikimedia Deutschland

Wikidata is a free and open knowledge base that allows humans and machines to access linked data. To support the international Wikidata community, Wikimedia Israel has developed a tutorial for the Query Service. With the tutorial, the chapter aims to empower new users to query Wikidata, which currently holds almost 90 million items. 


Tell us a little about your background and who you are!

My background is in academia. I did my Ph.D. and studied experimental psychology. While I worked in academia, many institutions were realizing that the business model of commercial publishers was unfair. Publishers get all the data, articles, and the work of the editors for free, and then they charge exorbitant fees from the libraries. This also became a motivation for me, shaking up this “world of paywall.” Most research is already publicly funded, so the public should be able to access and enjoy the fruits of that research. At Wikimedia Israel, I started working as a coordinator for academic projects. Now, I am also the Wikidata Project Coordinator.

What is the Wikimedia community in Israel like? 

The community in Israel is mostly Wikipedia editors. There are very few people who have Wikidata as their focus. As a chapter, this is something we are trying to change. Israel is a small country, but we have a disproportionately large high-tech industry. We are always looking for different ways to get these developers, engineers, and programmers on board. However, they are generally very busy people, who work long hours and often don’t have a lot of extra time to volunteer. So, we try to find scenarios where people gain something from contributing. Very few people volunteer just because they believe in free knowledge; most people want to gain a more immediate reward. 

Why did you decide to develop a tutorial?

I give a lot of talks and workshops to introduce Wikidata. Once participants understand the concept of linked data and see what the query service is capable of, they are very excited. Their eyes light up because the database is so varied; it includes “everything from everything” in all countries and languages and allows you to analyze data in a way that no other information service can do for you. You can research topics that you wouldn’t find in any other database. 

Yet, it is not possible to just use natural language to query; you have to learn SPARQL. So, we figured that teaching SPARQL is the best way to get people to fully grasp Wikidata. Since it is only a query language, it is teachable and limited in scope. People can learn it after a few lessons and get involved. Once they get involved, maybe they will also contribute to the data. If you make a query on French films in the sixties and you see that a third of the films don’t have a producer, you can say “Let’s complete this dataset, let me add what’s missing.” 

Who else do you target with the tutorial? 

We have many different target groups: librarians and information scientists, students, and the technology crowd. Information Scientists and librarians are closer to the content, and easily get how Wikidata is useful for organizing knowledge, but most have no experience with SPARQL.

Another target group is people who are interested in a particular topic, for example, film buffs or fans of a particular sport or team. In the past, we could try to encourage them to write Wikipedia articles about their topic of interest, but now they can also gather data about that topic. For example, a person may collect data about “French film in the sixties” and get it organized in a way that allows an overview or to present a new idea about it.

Why did you choose Star Wars as an example for the tutorial?

This was one of the hardest choices! I wanted something that would be universal. Everybody knows Star Wars.t is not just a regional thing; it’s not that in one part of the world, people think Star Wars is good, and in the other, they think Star Wars is bad. Also, I noticed that many of the instructional materials that Wikimedia produces have their background in GLAM, and often these examples can be quite elitist. They talk about Rembrandt paintings or Mozart pieces, but somebody in Southeast Asia doesn’t necessarily connect to these examples. I thought that if I am doing a tutorial for everyone, I want to use something that many people recognize. 

Wikimedia Israel’s general attitude is that since we have such a small potential audience – just nine million Hebrew speakers – we need to encourage people to make a one-time contribution to Wikimedia projects (Wikipedia, Commons, Wikidata, etc…) on a topic that interests them. We don’t try to turn them into Wikipedians or Wikimedians. Whether they do it on a topic in a school or university project or within other activities, once a person realizes that they are able to contribute, maybe they will do it once more. If it really interests them, they might continue to do that. 

Our aim is to show people that it is not so difficult to make a contribution. Coming from this attitude, we try to make everything very accessible. Star Wars might not be CC-BY-SA, but that was less of a consideration for us than creating an easy entry point for people who don’t know anything about Wikidata. 

What’s the most interesting thing you found out by querying yourself?

One of the things that really surprised me is how, in some Wiki projects, a lot of the information is updated so quickly, while for other projects, it is rather slow. For example, I looked at the Academy Awards, and the information on Wikipedia about them is fantastic. Then I looked at Wikidata and noticed that it was not as complete, which was shocking, because I thought it is so easy to transfer the info into this kind of data. 

It became even clearer to me how much we need to get more people to look at Wikidata – not necessarily because they are data scientists, but because they love football, or they love cinema.ll of these topics attract people. “We need to show that people can do work that means something to them personally.” We just need to show them that they can do meaningful work or work that relates to personal matters. For example, my sister lives in Tenerife. When I visited her and traveled on the island I was told that some Star Wars film was shot on location there. So I ran a Wikidata query for films shot on location in the Canary Islands. Turns out “Solo: A Star Wars Story” was partly shot on Fuerteventura, not on Tenerife.  

How does Wikidata support the multilingual nature of your work?

The core work of Wikimedia Israel is traditionally aimed at Hebrew and Arabic Wikipedia, but Wikidata introduces a whole new concept. It is important to understand that Hebrew Wikipedia is a completely different site than Arabic Wikipedia. In fact, you will find many articles that tell a completely different story. Wikidata is the same data, but you present it in different languages. The information isn’t given in a way that it is already interpreted. No matter what language you speak or what community you belong to, it is the same data. 

When we started developing the tutorial, we first said that we want to do it in Hebrew, but I felt like that it is a lot of investment for a small potential of people, even if we do it also in Arabic. So we did it in English and will translate it afterward. With the technical format that we have in WordPress, it is also easy to translate. It was the first time we developed a tutorial not just for Israel, but for the whole international movement. 

If other chapters are interested in translating it into their language it is very easy to do, you only need to translate the text; the exercises are the same. We are really interested in this kind of international collaboration and seeing if we can get the website in Spanish, French, and German. Also, we want to ask the people in all communities who do Wikidata outreach whether they are missing in the tutorial or want to change something. It is exciting for us to rise above our small chapter and see what we can offer. We want to give something back.

Share your feedback

To share your thoughts about the tutorial with Wikimedia Israel, please write to wdqs@wikimedia.org.il

About this post

Featured image credit: The Wikidata Query Service Tutorial, Dr. Keren Shatzman and Elisabeth Giesemann, CC BY-SA 4.0

Leave a Reply

Your email address will not be published. Required fields are marked *