1 of 27

A worldwide database of cultural heritage institutions on Wikipedia

Wikimedia Sverige�

John Andersson

Executive Director

Wikimedia Sverige

25 April 2015. Nepal is hit by a massive earthquake. Around 9,000 people loses their life and much of Nepal stops working for a long period of time. This was a terrible loss of lives and it was also a terrible loss of cultural heritage.

700 temples, palaces and museums were damaged and for many of them the digital information available was limited, or even nonexistent. This was preventing a fast and efficient response. This lead to more damages, more risks and more problems rebuilding the historical sites and collections.

This might come as a chock for you, I know that it was for me, but as of now we are lacking a global database of where the worlds libraries, archives and museums are located! To change this we have started the initiative FindingGLAMs. Which is a worldwide, multilingual, database of cultural heritage institutions connected to Wikipedia.

https://www.unisdr.org/archive/64971

2 of 27

Wikimedia Sverige

Wikimedia Sverige (founded in 2007) is the organisation that helps Wikipedia grow in Sweden

We specialise in large scale cultural heritage and technical projects

Havängsdösen, Sweden�Niklas Granström

CC BY-SA 4.0

3 of 27

Wikipedia aim to cover all the important information

“Imagine a world in which every single person on the planet is given free access to the sum of all human knowledge. That's what we're doing”��Jimmy Wales

Derawar Fort, Pakistan�Tahsin Shah

CC BY-SA 4.0

5th most popular website in the world�

500 million people read 15 billion Wikipedia articles every month

One of the core platforms we are actively working on is Wikipedia.

With more than 500 million people searching for information on Wikipedia every month, reading 15 billion articles it is crucial that Wikipedia contain as much up-to-date and well sourced information as possible.

Wikipedia is a unique and groundbreaking way of compiling information about the world. Our organizations and the volunteers that contribute to our platforms bring together published expert knowledge from across the world, compile and make it understandable and connected, and then repack it and share it with the world in hundreds of languages.

The articles are filled with references where you can read more and fact check the information.

A lot of people thought that this volunteer based approach would not work, but study after study show that Wikipedia is equivalent, or better, than traditional encyclopedias. So how is this possible?

4 of 27

Wikipedia

5 reasons why it works

Well, first of all we are standing on the shoulders of giants. None of the information on our platforms should be new and unpublished but should be based on verifiable sources. The academic community and experts are crucial for Wikipedia. What we do is making the knowledge accessible for the general public.

Secondly, more people want to do good than to destroy. Isn’t that just amazing? Wikipedia is currently improved by 250,000 volunteers working together every month. They ensure that the quality is high and removes vandalism and mistakes.

Third, we are partnering with great organizations across the world to bring their knowledge to Wikipedia and our other platforms. We are stronger together and by pooling our resources we can deliver amazing value.

Forth, we have very strict rules against self-promotion. Our platforms provide a free space where you can share your knowledge, not promote your organization.�

Fifth, all of the content is freely licensed and can be reused and adapted. People really like to help others and by contributing with freely licensed information they know that their contribution will have the largest possible impact. The free license also allow them to easily build on each other's work and improve on it. This is the idea with crowdsourcing.

5 of 27

Our community

A global community of 250,000 regular volunteers

Affiliate organisations in over 70 countries

Wikimedia Foundation in the US

Less than 700 staff

Mainky financed through donations

Wikimania 2016 - group photo 02

Niccolò Caranti

CC BY-SA 4.0

6 of 27

Wikidata

Wikipedia’s facts database has 77 million subjects

Millions of new subjects in the past year (pink)

Wikidata items map with difference�Adam Shortland

CC BY-SA 4.0

In support of Wikipedia we have started our own database of facts called Wikidata which holds facts on over 77 million subjects, its growing very quickly with 10s of million new subjects added in the past year which are in pink on the map. You can see that not all areas of the world are equally covered.

Wikidata is all about facts and is built in a way that both humans and machines can easily use the information.

Wikidata also contains information that is not shown on Wikipedia as the information on Wikidata is being used in thousands of other products and platforms. For example, when you search on Google, or use Alexa or Siri the information provided to a large extent comes from Wikidata.

Wikidata is closely connected and interlinked with Wikipedia which will ensure that information is easier to keep updated across all of our language versions.

7 of 27

Cultural Heritage on Wikipedia

Wikipedia is the largest cultural heritage resource in the world

150,000 cultural heritage institutions listed on our database, very Western centric

Most of our contributors are from Northern Europe, North America and Japan

Map of GLAM institutions on Wikidata before the project FindingGLAMs started

John Cummings

CC BY-SA 4.0

8 of 27

FindingGLAMs

Takht-e soleymān, Iran�Ebrahim Alipour

CC BY-SA 4.0

9 of 27

FindingGLAMs

GLAM = Galleries, Libraries, Archives and Museums

The first worldwide database of cultural heritage institutions in Wikipedia, in 300+ languages

We are building it through official datasets and crowdsourcing through cultural heritage institutions and the general public

Hardknott Roman Fort, United Kingdom�Markas1370

CC BY-SA 4.0

10 of 27

Starting with an index

Identifying what datasets exist

Collected metadata about datasets about GLAMs

What, where, licence, etc.

Now includes 44 countries + worldwide datasets

meta.wikimedia.org/wiki/FindingGLAMs/GLAM_datasets

Statsbiblioteket læsesalen

Villy Fink Isaksen

CC BY-SA 4.0

As we collect information about the existing datasets, we put them together in a master index.

That way, we have an overview of what data already exists and who owns it. We investigate the size of the dataset and its scope, and most importantly – its copyright status. There’s a lot of data out there that could be very useful for the open knowledge movement, but its copyright status prevents it from being re-used freely. Meaning that there are limitations on how it can be used, which is preventing us from including the data.

Still, we want to know that those datasets exist, so that we or other community members can initiate discussions with the relevant institutions about the benefits of releasing the data.

The index is published on a wiki page, just like all the other documentation from our project. You have the link here on the slide.

Currently we have identified datasets from 44 countries and a number of datasets that are covering the world.

Together, we’re creating a map of GLAM information around the world.

11 of 27

A worldwide database of cultural heritage institutions

We want to make sure all countries have their cultural heritage documented on Wikipedia

A free public resource for citizens, tourists and companies to use in their products

A database suitable for research and other educational uses

Information on location and contact details is crucial to provide support in disasters

A way to find new potential partners

Takht-e Soleymān, Iran �Ebrahim Alipoor

CC BY-SA 4.0

A comprehensive worldwide database of cultural heritage institutions would be valuable and important for many reasons:

Wikipedia’s mission is providing the sum of all human knowledge, not just the knowledge of the Global North. That means we have to actively work to fill the gaps.
This database will be a free public resource for education, for citizens, tourists and it should be available across the web, in many forms, including commercial websites and apps.
A truly worldwide database could be an important research tool for academics. Because the information is structured the same way and connected to other databases where more details can be found researchers will have a valuable resource and new types of big data research are possible, allowing for new findings. Wikidata is a node for linked open data.
We also know that one of the main issues in disasters such as earthquakes is a lack of information on what institutions exists and where they are located. This enhance the problems. Wikipedia is easily available, in multiple languages and can even be downloaded and used offline. The information can easily be extracted and form the backbone for other services.
Furthermore, we can also use this dataset to understand who to work with. We can identify what experts institutions exists regarding a specific topic in a country and work with them to share their knowledge. Without a database such as this, finding new potential partners is very hard and time consuming.

Collecting the basic information about where and how to find and reach the institutions is the first step. In the future we hope to add detailed information about the collections the institutions hold down to specific items. That way we connect the world's historical collections and allow people access to their culture and history, regardless where the physical object is stored.

12 of 27

With the support of the MY-D Foundation�

Source: Fontenelle & Estermann (2019) An International Knowledge Base for All Heritage Institutions

13 of 27

Beyond writing articles...

Wikipedia organizes the largest photography competition in the world

Wiki Loves Monuments

Also Wiki Loves Earth, Wiki Loves Africa and more

Sacra di San Michele, Italy�Elio Pallard

CC BY-SA 4.0

The data collected and incorporated onto the Wikimedia platforms makes it possible for us to organize the largest photography competition in the world. Through the competition we are crowdsourcing the digital documentation of these sites. So we are already using the data to help reduce the gap of information in the world. As soon as we have a site in our database, we can use our community to document it further. The data helps us guide the participants to the objects to photograph.

Wiki Loves Monuments focuses on built cultural heritage and is the largest photo contest with over 2,225,000 images uploaded as part of the contest since 2010.

Wiki Loves Earth focus on protected natural heritage and is also growing every year. Helping to highlight the beauty of the natural world.

In fact, almost all the photos in this presentation are from these photography competitions and are free for everyone to use.��

14 of 27

Our progress

Identified 67 datasets

Imported data about 38,000 institutions

Running a crowdsourcing campaign for more data

Wat Maha That, Thailand�Siripatwongpin

CC BY-SA 4.0

15 of 27

Map of GLAM institutions on Wikidata. Yellow dots represent institutions added as part of the FindingGLAMs project.

John Cummings

CC BY-SA 4.0

16 of 27

How this initiative supports the Sustainable Development Goals

Improve the scientific research (SDG 9.5)

Contributing to partnerships (SDG 17.6, 17.16, 17.17)

These partnerships can help reduce inequality and access to knowledge (SDG 4, 5, 10)

Sheikh Lotfollah Mosque, Iran

Alireza Akhlaghi

CC BY-SA 4.0

17 of 27

Support the initiative

We have the worldwide audience, the tools and the volunteers

We need your help with:

1. Identifying what data exist in your country. The datasets are held by different ministries, regional and local government in different countries and also by universities and NGOs. Not always publicly accessible or online

2. Helping us to get in contact with the right people

3. Introducing us to other organizations to work with

4. Sharing the information through your communication channels

Neptunbrunnen (Berlin-Mitte)

Ajepbah

CC BY-SA 3.0 DE

We would like to work with you to share data about your cultural heritage institutions on Wikipedia.

Through Wikipedia we have a worldwide audience which allow us to educate people about cultural heritage and we have tools and volunteers to write the articles.

�But we have many issues collecting the data, information is spread across a wide range of sources in many, many languages and we don’t yet have the connections or understanding of cultural heritage across the world to collect the data ourselves. To add to this in many instances the data is difficult to access even if you know exactly what you’re looking for.

You can work with us in a number of ways.

We hope that you first of all can help us identify what datasets exists in your country.

Secondly, if the dataset is not open and possible to reuse, we hope you can put us in contact with the right person.

Third, if you know of other organizations in your country that would like to work with us, we encourage you to put us in contact and introduce us.

Forth, we will continue to develop the information we have through crowdsourcing. Could you help us engage the general public to improve the information about the institutions in your country? Could you share the information on social media or other channels?

18 of 27

What is next

Work with data collection is continuing!

Adding more information to the map

Wikimedia Sverige is becoming a thematic support structure

Supporting Wikimedia affiliates and GLAM institutions to work together – by building software, developing documentation and support structures, offering funding and capacity building

Verrucole Castle, Italy�Iris Gonelli

CC BY-SA 4.0

We will continue to add new datasets continuously over the years to come. This database is a key piece of the infrastructure to support our partners with high quality information.

One area is to improve the map data that is available around the institutions so that emergency personnel can reach them faster if an accident happens.

We are working on securing a large agreement that will allow Wikimedia Sverige to become a thematic support structure for the global Wikimedia movement.

The plan is that we will be supporting Wikimedia affiliates and GLAM institutions to work together - by building and maintaining software, by developing documentation and support structures, by offering funding and support with capacity building. By doing so we hope to make things progress a bit faster. So if you are interested in seeing more of your cultural heritage on our platforms, please contact us.

We hope that you will all join us in getting information about the world’s cultural heritage online. It will be a fantastic ride.

19 of 27

Please get in contact!

Talk to us here

Email: john.andersson@wikimedia.se

meta.wikimedia.org/wiki/FindingGLAMs

Let us know of any heritage datasets and lists you’d like to share

Holy Mountains Monastery, Ukraine�Konstantin Brizhnichenko

CC BY-SA 4.0

20 of 27

THANK YOU!

Tack så mycket!

John Andersson

Executive Director

Wikimedia Sverige

john.andersson@wikimedia.se

21 of 27

22 of 27

23 of 27

24 of 27

Why participate

Build relationships

Get technical support

Get experience and skills working with Wikidata

The Royal pavilion in Phraya Nakhon Cave in Khao Sam Roi Yot National Park, Thailand �BerryJ

CC BY-SA 4.0

25 of 27

How this improves Wikipedia

Understand which articles are missing from Wikipedia

Use the data to generate new articles, especially helpful for smaller languages

Use as a list of subjects for Wiki Loves Monuments

A resource for Wikipedia to find new partners to work with

The Royal pavilion in Phraya Nakhon Cave in Khao Sam Roi Yot National Park, Thailand �BerryJ

CC BY-SA 4.0

26 of 27

Wikipedia

The largest online encyclopedia

Available in over 300 languages

Over 50 million articles

Written by volunteers

Available free for everyone to use and improve

Sacra di San Michele, Italy�Elio Pallard

CC BY-SA 4.0

27 of 27

Wikipedia

5th most popular website in the world�

500 million people read 15 billion Wikipedia articles every month

Google, Siri and others use our data to answer questions �

Wikipedia is the largest cultural heritage education resource in the world

Grohman’s Factory, Poland�Marian Naworski

CC BY-SA 4.0