Purpose of the report

This form is for organizations receiving Annual Plan Grants to report on their results to date. For progress reports, the time period for this report will the first 6 months of each grant (e.g. 1 January - 30 June of the current year). For impact reports, the time period for this report will be the full 12 months of this grant, including the period already reported on in the progress report (e.g. 1 January - 31 December of the current year). This form includes four sections, addressing global metrics, program stories, financial information, and compliance. Please contact APG/FDC staff if you have questions about this form, or concerns submitting it by the deadline. After submitting the form, organizations will also meet with APG staff to discuss their progress.

Background & strategic context

2019-20 represents the first year in the delivery of our new three strategy. This has brought some refinement in our work on underrepresented knowledge, digital literacy, and advocacy for open knowledge. We also started introducing new metrics to help us hone in on measuring our work.

Strategic Framework for 2019–2022

Wikimedia UK believes that open access to knowledge is a fundamental right, and a driver for social, educational and economic development. We work with the Wikimedia Projects such as Wikipedia to enable people and organisations to contribute to a shared understanding of the world through the democratic creation, distribution and consumption of knowledge. We are committed to the ideal of a world in which every single human being can freely share in the sum of all knowledge, and believe that here in the UK, we can play a unique and important role in realising that vision.

Wikimedia UK’s strategy is informed by and supports the strategic direction of the global Wikimedia movement. Our work will focus on the knowledge and communities that have been left out by structures of power and privilege; breaking down the barriers that prevent people and organisations from accessing and contributing to free knowledge, and supporting the development of technical solutions to help eradicate inequality and bias on the Wikimedia projects.

Our vision is of a more tolerant, informed and democratic society

Our mission is to be the platform which enables the long-term sustainable development and use of open knowledge in the UK

Wikimedia UK is working towards the following long-term outcomes:

Our work has significantly increased free, online access to knowledge
The Wikimedia projects reflect our diverse society and are free from systemic bias
Learners in the UK are able to understand and engage with open knowledge
High levels of information literacy have strengthened civil society and democratic processes
Wikimedia UK is recognised as a leading organisation for open knowledge

Our strategic aims for 2019–2022 are to:

Increase the engagement with and representation of marginalised people and subjects
Work with partners to develop digital, data and information literacy through Wikimedia
Create changes in policy and practice that enable open knowledge to flourish
Develop our capacity and profile as a leading organisation for open knowledge

The four goals below correspond to the four programme strands. These represent a refinement rather than an overhaul of our previous programmes, with many projects and partnerships having continued into 2019/20.

Our delivery model is based on the cross-cutting strategic priorities of partnerships, community and technology, which are an essential element of all our activities.

2019/20 Indicators, targets and logic models – background

As part of the process of renewing our strategy for 2019 - 2022, we reflected on and reviewed our success measures. In applying for 2019/20 funding from the Foundation, we defined and set targets for five metrics as required, which are the Grant Metrics and Defined Metrics below. Our aim for this year was to revise and shape our metrics in the light of the new strategy. We spent some time thinking about and developing these throughout the programme year, and are excited to share the initial baseline findings. We are not only hoping to gain a better understanding of our programme’s impact through these new metrics, but also to offer these insights and ideas to other organisations within the Wikimedia movement.

The new metric ideas presented below are inspired by the review of our delivery between 2016 and 2019 and reflections on our Theories of Change, as well as board-level discussions about outcomes, impact and programme excellence, which took place at our away day in September 2019.

In reviewing the logic models for our previous three programme strands, and designing new theories of change for each of our newly defined programmes, we were able to create an overall organisational theory of change, which draws together our vision and our strategic aims - making the link between the different areas of our work:

Wikimedia UK believes that to achieve our vision of a more tolerant, informed and democratic society we need to improve the representation of diverse people in the knowledge ecosystem, increase civic engagement by building digital literacy, and secure policy changes which increase access to open information for all. To effectively achieve these goals we must also work on strengthening our voice and sector recognition.

Without access to knowledge, we can’t build understanding. Without diversity of content, this understanding is limited.

Review of Strategic Delivery 2016 - 2019

Early in 2019 we commissioned an extensive independent report on Wikimedia UK’s delivery during the period 2016 to 2019. The report aimed to summarise progress against our quantitative indicators, pull together the qualitative highlights of our programmes and provide an overall analysis of how our programme activities delivered on our 2016 - 2019 strategic framework and the short to intermediate outcomes identified in our logic models.

The report shows that we were overwhelmingly successful both in meeting our annual quantitative targets and in achieving both our short and intermediate outcomes. It draws on our very wide range of programme activities to illustrate the different approaches that we have taken, the many ways in which volunteers are involved in the delivery of our programme, and the different types of partnerships that have been developed.

The report explores some of the key achievements and innovations facilitated by the chapter during the past three years, such as the appointment of the first ever Wikidata Visiting Scholar, the launch of the first minority languages community conference, the achievement of gender parity in biographies on the Welsh Wikipedia and the creation of a digital literacy framework for Wikimedia. It also highlights the fact that a number of our Wikimedians in Residence - initially, by their nature, fixed term projects - have been so highly valued by their host institutions that the post has been made permanent.

Some of the numbers included in the report which don’t currently form part of our formal quantitative indicators include the 17 project grants awarded to volunteers, 82 community events and meetups, 195 Wikipedia editing events and 31 Wikidata and Wikimedia Commons training sessions, with activities organised in 98 different places across the UK. This extensive offline work with the Wikimedia UK community translated into significant online impact, with 2 million Wikimedia articles added or improved as a direct result of our programme over the three years under review.

Global metrics overview - all programs

For more information and a sample, see Global Metrics.

Overall

Metric	End of the Year Targets	End of the Year Results	Explanation and/or Examples of Activities
Participants GM1	6,000	7,475	People attending our events: Wikipedia in classroom courses, conferences, workshops, training, editathons, and community meetups. The rising popularity of Wikipedia in classroom courses increased our numbers here, particularly via the university-based Wikimedians in Residence.
Newly registered editors GM2	1,000	1,007	Mostly from education courses, plus Wiki Loves Monuments (WLM) contest, our grant to support engaging dentistry students in editing, and new editors trained as part of volunteers trained to help with uploads (e.g. WiciMon project in schools in Wales).
Articles added/improved GM3	250,000	397,202	Almost 120,000 new items on Wikidata on Welsh books, paintings from the National Library of Wales and P.B Abery photographs were added by the Wikidata Scholar at National Library of Wales (NLW) and the resident, 4862 Wikidata items created using metadata for each image - all data added in English and Welsh, items added to the Dictionary of Welsh Biography data, and about 210,000 edits on Wikidata, mainly cleaning data for images and adding image collections from Commons to improve this website: http://zone47.com/crotos/dwynwen/cosmos/ and edits to the data on Wikidata on Dictionary of Welsh Biography by the resident at NLW. Remaining content comes from the rest of our programme activity, e.g. education courses.
Volunteer hours	20,000	24,225	5,400 hours come from Wikipedia in Classroom courses. 584 hours from editors part of the NLW Volunteer Programme creating and/or improving articles on health, the Dictionary of Welsh Biography and translation. 516 hours spent by students in Wales taking part in the WiciMon project in Wales. The rest are made up of wiki training, workshops, editathons, the AGM, meetups and other events (11,000 hours, including 1,600 hours of lead volunteers organising events and interacting with WMUK partner institutions).
Total audience and reach (online or in person engagement with Wikimedia UK)	70,000	64,510	Includes 56,584 social media subscribers, 7,475 event participants plus 451 leading volunteers. We had an issue with our website and blog for a number of months in 2019 which meant that visits and views were not tracked. We believe this explains why the target was not reached. We are still pleased with the size of the audience we reached. In terms of real-life engagement at events, a few significant examples are worth highlighting: 180 people attended a speech by Robin Owain at the Our Voice in the World Festival in Aberystwyth, 120 people attended a workshop on editing Wikibooks at University College London, 500 people attended presentations on Wikidata at the Museums+Tech conference and during the Open Access Week by Martin Poulter and Jason Evans at the British Library in London, 250 people attended Lucy Crompton-Reid's keynote speech - Creating a more tolerant, informed and democratic society through open knowledge - at the CILIP Scotland Annual Conference for librarians and information professionals, 80 people attended Martin Poulter’s presentation in Oxford: "Wikidata: Knowledge Representation the Easy Way" lecture at Digital Humanities Summer School, 60 students attended Aaron Morris’ Skills Week workshop in Ysgol Gyfun Llangefni, Secondary, 100 people attended presentations on the Map of Witches Project at the "Suffer the Witch" symposium and on Mapping the Scottish Reformation with Messy Data seminar by Ewan McAndrew at the Edinburgh Centre for Data, Culture and Society UoE.

Color coding of the results

Achieved	Target has been achieved or exceeded
Opportunity for improvement	Good progress has been made towards achieving the target, but the target has not been reached
Attention required	Little or no progress has been made towards achieving the target

All metrics totals for all programmes

For new metrics, see definitions below.

For metrics disaggregated by programmes, see individual sections per programme below, for example programme 1. Numbering on the left is consistent throughout the report.

	Metric	2019-20 result
1	Total audience and reach (direct engagement) (grant metric)	64,510
2	Digital media reach	56,584
3	Total participants (grant metric)	7,475
4	Number of newly registered users (grant metric)	1,007
5	Number of leading volunteers	451
6	% of leading volunteers who are women	44.68%
7	Estimated number of volunteer hours (grant metric)	24,225
8	Volunteers would recommend WMUK (annual)	88.37%
9	Volunteers feel valued by WMUK (annual)	83.87%
10	Volunteers have developed new skills (annual)	81.82%
11	Images/media added to WM Commons	15,280
12	Images/media added to WM article pages (and %)	2,853 (18.67%)
13	Content pages created or improved, across all WM projects excl. Commons (grant metric)	397,202
14	Number of articles created	145,372
15	Reach of content - image/article views (NEW)	1,219,388,403
16	Content diversity - % of events where the focus is on underrepresented content (NEW)	106 (34.86%)
17	Language diversity - how many languages have we worked across (annual) (NEW)	Articles created: 13 Articles edited: 55 Common uploads 2019/2020 re-used: 53 All Commons uploads re-used: 321
18	Geographical reach - % of events outside of London (NEW)	80.92% (246 out of 305 events)
19	# education courses we work with (annual) (NEW)	20
20	Digital skills - Improved skills and confidence (annual)	Score increased from 3.799 out of 5 to 3.927. Measured by survey; see Programme 2 for further details
21	New inclusion in courses and curricula (annual)	4
22	Responses to consultations (this will become ‘Policy touchpoints’ in 2020)	2
23	Policy change	3

New metrics for the 2019-22 strategic priorities have been approved by the Board during the 2019/20 activity year. The team explored how to interpret, capture and represent them. Below are our initial findings for 2019-20 (totals for the year):

Diversity of leading volunteers: see narrative from volunteer survey - we feel this would be hard to express in a single result KPI. We have included diversity questions in the volunteer survey, however, and can describe the results below.
Reach of content: we primarily considered image views in Wikipedia articles of the content released with support from Wikimedia UK (Commons category). In 2019-20 there were 1.2 billion views (1,219,388,403). We are working on methodology for this and so the results may change in the future. These figures are so big that it may also be interesting in the future to highlight views from particular sections of content - for example, in 2019 photos uploaded from Visit Wales got 1,302,267 views.
- We also considered including views for Wikipedia articles written at initiatives we supported. It is extremely time consuming to provide a number spanning all of our article writing work, so in the future we may want to look at the output of specific events. In any case the articles created in 2019 during WMUK events received 747,979 views (in 13 WP languages).
Programme content diversity: interpreted as the % of events where the focus is on underrepresented content: 35% (106 out of 305 events). This may seem low, but most other events were focused on other strategic priorities including digital literacy and advocacy.
Language diversity: interpreted as the number of languages we have worked across: this could be expressed in different ways, each showing the multilinguality of our work. It may be useful to retain this diversity of metrics rather than narrowing down to one indicator.
1. We created articles in 13 languages
2. We edited articles in 55 languages
3. Our images uploaded to Commons in 2019/2020 were re-used on 60 Wikimedia projects, including 53 languages Wikipedias
4. Our images were used in articles on 321 wikis

Geographical reach: interpreted as the % of events outside of London: 80.92% (246 out of 305 events). We organised events in Wales (69), England(71), Scotland (88) and Ireland (3), as well as sessions online (7), in France, Spain, Germany and Sweden, connected to other Wikimedia chapters’ events. 89.43% of our educational programs were organised outside London, whereas the % of events outside London was just a bit behind, in SG1 (79%), SG3 (70%) and SG4 (73.33%). Scotland was leading for underrepresented content related events, while Wales was leading for education related courses, and England (outside London) for advocacy in SG3.

Breakdown of WMUK events staged outside London
Events outside London per strategic goal

Education courses active: 10 active in Q4 at the University of Edinburgh (3), Queen Mary University London, Middlesex University, University of Stirling, University of Sheffield, University College London. Throughout the year, there were 20 active courses in total, covering the University of Edinburgh (7); Middlesex University (2); University of Stirling (2); University of Sheffield (2); Imperial College London (1); London School of Economics (1); Queen Mary University, London (1) Swansea University (1); University College London (1); University of Derby (1); University of Kent (1).
Digital skills - for the 2019-20 academic year we adopted a new approach to measuring student digital skills: a survey at the start of the course and a survey at the end to establish whether the course had resulted in changes in confidence. The survey indicated that confidence in online collaboration increased, as well as confidence in using data. Giving the results a weighted average, across eight aspects relating to digital skills students increased their ability from 3.799 out of 5 to 3.927. Courses also resulted in greater understanding of open knowledge, open data, and copyright.
Inclusion in courses and curricula - 4 - this activity year there were new courses using Wikipedia at the University of Derby; London School of Economics; University of Kent; and the University of Sheffield in as well as discussions with the London College of Communications and the King’s College London digital humanities department about the inclusion of Wikimedia.

Telling your program stories - all programs

Programme 1: Knowledge Equity

Strategic goal: Increase engagement with and representation of marginalised people and subjects

Background and Theory of Change (TOC)

Long term outcome: Wikimedia reflects our diverse society and is free from systemic bias	Assumptions There are content gaps on Wikimedia projects which makes it biased and not-representative of our society
Theory of change: Wikimedia UK is helping to create more complete information online; by supporting marginalised people to become contributors and community leaders, and by uncovering and sharing knowledge created by and about underrepresented people and subjects.	Where is our intervention We facilitate the transfer of underrepresented knowledge from content holders to diverse audiences. This usually takes place through the Wikimedia projects but can be through other accessible platforms where appropriate. We develop targeted projects and programmes to recruit, train, support and develop volunteer contributors and leaders from all backgrounds.
Strategic objectives: Develop partnerships that address inequality and bias on the Wikimedia projects Deliver projects that increase access to underrepresented cultural heritage Support the development of minority and indigenous language Wikipedias Engage with volunteers and partners across the UK, widening the charity’s geographic reach Diversify content producers by recruiting new editors from under-represented communities Support the development of a more inclusive culture across the Wikimedia projects Ensure that Wikimedia UK’s own policies and practices support diversity and inclusion	Why us We have connections to the existing editor community, experience in running programmes and training sessions, and strong links with UK content holders.
Activities: Partnerships, projects and campaigns focused on Underrepresented cultural heritage Minority languages (with a focus on the “Celtic Knot”) Gender Gap Diverse contributors	Evidence that it’s working + Who are we reaching (Metrics) Content pages created or improved across all Wikimedia projects Images/media added to Wikimedia Commons Number of articles created Newly registered editors Volunteer hours Total number of participants Number of leading volunteers Reach of content - image/article views Content diversity - % of events where the focus is on underrepresented content Language diversity - how many languages have we worked across (annual) Geographical reach - % of events outside of London (annual)

Metrics

(The metrics in the left column in bold are our Grant metrics)

Indicator	2017/18 end year results	2018/19 end year results	2019/20 end year results	Explanations
3. Participants	2686	2,294	1,709	People taking part in our events - Wikipedia/Wikidata training series, open events, editathons, conferences, workshops, meetups, volunteer programs. This result is smaller than in previous years (while we achieved our target across all programmes), which reflects a growing focus on our education programme.
4. Newly registered editors	734	627	553	Mostly registered from editing training sessions, editathons, and volunteer content competitions, or grants. Context as above.
13. Articles added/improved	344,786	1,112,699	394,911	This includes new and modified articles on Wikipedia and files added and edited on Wikidata e.g. our visiting Wikidata Scholars’ mass uploads of Welsh biographies of notable people, paintings and photographs of NLW’s collections, or the WiciHealth project that is translating English articles to Welsh. This number depends heavily on the type of projects we run, and can differ a lot year to year.
7. Volunteer hours	8639.6	8,409	9,257	The amount of time that participants and lead volunteers contributed to the work of Wikimedia UK. This can include attendance at contests, open events, training series, workshops, editathons, conferences, and presentations. As we work on more complex partnerships (needing more discussions, setup etc), the volunteer time increases.
1. Total audience and reach	2819	2,460	1,952	Includes participants as above (GM1) plus leading volunteers (see below). Context as with GM1.
5. Leading volunteers	133	166	243	Leading volunteers who are organising or maintaining GLAM partnerships, contests, workshops, and editathons engaging new communities and institutions, volunteers taking leadership roles in driving the activities of the chapter. Working on more complex GLAM partnerships means more leading people involved.
11. Images/media added to Commons	27,207	41,256	15,280	Wiki Loves Monuments resulted in 10,665 uploaded images, and the rest of the results comes from a diverse set of image and audio files.
12. Images/media added to Wikimedia pages	8456	20,435	2,853	Mostly images added to Wikidata (Listed buildings, Natural landscapes, images of notable people from England and Wales, paintings from the collections of NLW). The Wiki Loves Monuments photos take a while to be ingested into Wikipedia/Wikidata, and so the reuse stands fairly low at this point.
12. % uploaded media used in content pages	31,08%	49.58%	18.67%	% of images that were re-used on WM content pages (incl. Wikidata)
14. New articles added	35,489	66,656	120,334 (incl.112k Wikidata items)	Wikidata uploads and articles created by volunteer programs, participants of educational courses and our editing events.
15. Reach of content - image/article views	N/A - new	N/A	1,227,127,546.00	A new metric, or rather a selection of indicators we are exploring: 1,219,388,403 views of articles using images uploaded to Commons in the Supported by WMUK category+ 747,979 views of articles created in 2019 during WMUK events (views in 13 WP languages)+ views in GLAMORGAN from Wales on categories Natural Resources: 5,688,897 and of “Visit Wales”: 1,302,267 views
16. Content diversity - # and % of events where the focus is on underrepresented content	N/A - new	N/A	106 (34.86%)	Events, training, editathons and workshops with a focus on underrepresented content. This takes all the events we organised this year (over 300) and only reports on the ones focused on underrepresented content. The remainder of events focus primarily on digital literacy or advocacy.
17. Language diversity - how many languages have we worked across (annual)	N/A - new	N/A	Articles created: 13 (eventmetrics) Articles edited: 55 (eventmetrics) Common uploads 2019/2020 re-used: 53 (PetScan) All Commons uploads re-used: 321 (GLAMorgen)	We measured across different tools the number of languages our activities involved, the languages we edited or created articles in, the number of language version Wikipedias where articles re-use images from our uploads in the reporting period and the cumulative no of language version Wikipedias in which images in the supported by Wikimedia UK category were re-used.
18. Geographical reach - % of events outside of London (annual)	N/A - new	N/A	80.92% (246 out of 305 events)	A solid proof we work to make sure not all our events are based in the capital.

Highlights of programme activity

Minority and Indigenous Languages (wit a Focus on the Celtic Knot)

Breakdown of articles created and edited on the English and Welsh Wikipedia's as a result of our work

Breakdown of articles created in a language other than English as a result of our work

The key celebration and knowledge sharing event in this strand of work was our Celtic Knot 2019 conference in July, described in detail in the case study below. Importantly, we focused the event on very small languages and brainstormed support and development opportunities available to these communities. Our ongoing work within the broader Celtic Knot theme included:

We completed a successful year of collaboration with the National Library of Wales and their National Wikimedian (originally a Wikimedian in Residence). The Library delivers specific themed projects funded partly by the Welsh Government, which in 2019 included:
- WiciPobl (Wiki People) project. This resulted in almost 2000 biographies being published on the Welsh Wikipedia, using both human and bot contributions. A translation event was held, together with a focused Wikidata activity.
- WiciLlen (Wiki Literature) project. This initiative focused on openly sharing information about Welsh literature on the Wikimedia projects. The project consists of two main strands. Firstly the National Library shared a huge dataset of all books of Welsh interest ever published in Wales. This dataset contains information about nearly half a million books, their authors and publishers. The data is searchable and reusable in dozens of languages, including Welsh. This will improve access to this important dataset, help improve citations on Wikipedia and provide opportunities for developers and researchers wishing to re-use the data. The second strand of the project is focusing on improving content on the Welsh Wikipedia.
- Wiki Education project, blending content creation with school education - more on this in our Programme 2.
- We’ve also been working on the Peniarth Manuscript collection which is on the UNESCO Memory of the World Register. Images of manuscripts were added to Commons and Wikidata.

We supported the Dumfries Stonecarving Project in Scotland to document and research the stonecarving heritage of Dumfries in collaboration with local photography groups continued throughout 2019. This beautiful element of Scottish heritage is largely overlooked and the project has helped redress this through Wikimedia. It’s also a great example of using digital in cultural preservation.
We already see Scottish heritage as underrepresented, however, within that context, it’s the Highlands & Islands that are often missed out and face more barriers to engagement. For this reason we are slowly building up links for programmes in this area, using existing partners and networks - for example the Highlands Data meetup, University of the Highlands and Islands Learning & Teaching week.

Beyond Celtic Knot languages, we worked with a long standing academic volunteer editing group called LingWiki. We run an ongoing series of events to represent minority languages and cultures on English Wikipedia, with a further aim to improve the general content related to anthropology. The group continued to make edits to minority language and linguistics pages and the task of improving anthropology pages. With our growing expertise in Wikidata, we worked with this group to see how their activity could be boosted by technology. This brings us to the second strand of our Knowledge Equity work:

Technology for Diversity

Increasingly, our work to support minority languages draws on technology-driven solutions, even when this wasn't previously the case. For example, we’ve been collaborating with the editing group LingWiki for several years, but have only recently started discussing the use of Lexemes for testing and improving data models. LingWiki is a social editing group made up of academics and PhD students at the School of Oriental and African Studies (SOAS). Connecting them with a computational linguist has worked particularly well so far and we will be developing this collaboration further.

The power of Wikidata to generate and amplify underrepresented content is significant. The search functionality can also shed light on previously hidden areas of under-representation. See below for our major case study on using data to uncover the stories of Scottish Witchcraft.

In 2019 we grew the number of projects in this area, and can highlight the following:

Science Museum – We met a representative from the Museum at the 2018 GLAMWiki Tel Aviv and discussed the potential use of Wikidata, possibly via Wikibase. The conversation continued, and we are now entering a more detailed collaboration planning stage. They envisage using Wikidata to manage their extensive collections data, with a particular focus on research capabilities and the possibilities offered by linked open data. A particular value we see is interrogating their collections data to uncover areas where the Science Museum has unintentionally underrepresented certain subjects. The technical capacity and expertise in the museum offers the potential for large-scale, well-managed work with linked data, and there is a good level of understanding of the challenges and opportunities of this work within the museum already. As an added dimension to this, there is a key initiative in the GLAM sector in the UK called National collection data aggregator - big programme to set up a unified way and place to capture and share collections data, powered by considerable funding from Arts and Humanities Research Council. Wikidata offers potential solutions to this, and so there is considerable interest in the sector to draw on it to produce solutions to the aggregator challenge. We are in talks with the Science Museum Group to work on this; there is potential for this collaboration to be substantial.
We are discussing a data collaboration with the British Library's Endangered Archives programme. They have a large collection of images of items (photographs, manuscripts, etc.) from archives in the Global South, especially in Africa but also South America, Asia and Eastern Europe. The images are not openly licensed, but there is the possibility of describing the items on Wikidata. They have been using OpenRefine on the archive data so are getting Wikidata identifiers for South American personalities depicted in photographs.
We are planning our involvement with the Hacio’r iaith (tech+language) conference at Aberystwyth University, spring 2020 (moved back from 2019). Usually this conference takes place over one day, but we will be delivering a second day based exclusively on open knowledge and the Wikipedia projects. We will showcase how technology such as Wikidata infoboxes can support the development of a minority language Wikipedia.
Throughout 2019 we’ve been working with Banner Repeater to develop schema for artists' publishing (linked-data compatibility, user-testing and schema development).

Technology is also being used for wider heritage capture and dissemination - for example, in Aberdeen, one volunteer from Code the City is working on capturing plaques and their content through Commons and Wikidata. There are now 132 Aberdeen Commemorative Plaques recorded in Wikidata (and 90 in Glasgow via a process we copied). This now allows for insights such as identifying which plaques lack images, or mapping plaques against a map. It’s possible to generate a timeline, or analyse the gender of plaque subjects. All these are interesting insights for any heritage group, and the volunteer plans to work with this hack group to extend this work.

Gender Gap

University of Edinburgh - the gender gap has long been a focus of activities at Edinburgh, and the university has recently appointed a Women in Red intern, who will work on promoting Wikipedia gender gap events (the next one in their calendar is 34th in series, which shows how embedded this activity is). This work is embedded in the University of Edinburgh's Athena SWAN four year plan to get female role models in Science, technology, engineering, and mathematics (STEM) disciplines online so others can learn all about their lives and contributions and be inspired. The University was Highly Commended for our Women in Red work by the British Society in the History of Science Outreach Ayrton Prize 2019.

We delivered events around the broad Art+Feminism banner - for example with Paul Mellon Centre, in collaboration with ArtUK, and University of Glasgow and Glasgow School of Art.
Women’s Engineering Society (WES) - Improving coverage of female engineers on Wikipedia and Wikidata to mark the WES’ centenary. Building on the first workshop that took place in September, a second workshop with the WES took place in January at the University of Leeds. Over the course of the workshops 26 people were trained to edit Wikipedia. Most of the attendees were women, addressing Wikipedia’s gender gap amongst its contributors, which is a factor in the under-representation of women in its content. In these two events alone, the attendees improved 58 articles, including 13 which were brand new. Since then, they have been read 27,000 times.

Diverse Contributors

We delivered a range of events throughout 2019 that were targeted at specific audiences. Some of our ‘underrepresented content’ work is focused particularly on working with communities linked directly to that content. We had an opportunity to focus on this during Wikipedia’s block in Turkey. We ran a Turkish editathon event with PEN International to expand content on human rights abuses and imprisoned journalists in Turkey. This session was surprisingly productive for new editors, with useful contributions made across a range of articles in both English and Turkish, showing that there is a lot of “low hanging fruit” on the Turkish Wikipedia for motivated people with access to good sources. We ran a follow up event more generally on Turkish culture, creating and expanding content in Turkish.

One dynamic we have noticed is that having trainers used to working on English Wikipedia pushes editors into adding references to other language Wikipedias with a generally more relaxed attitude to citations. An unexpected outcome was Uyghur attendees at one of the events, who created content on the Uyghur Wikipedia during the session and who are unable to return to China (unrelated to their involvement in this event). We are following up to support them, and to see if there’s any room for supporting work on underrepresented languages on Wikidata.

Other examples of working with specific communities include:

an editathon to coincide with Bolton Pride in mid-September
an October editathon with the David Livingston Trust focusing on Black History Month.
We have been discussing a collaboration with the Ahmed Iqbal Ullah Race Relations Resource Centre who are interested in learning how to make some of their archive on Black British history more publicly available. Work such as training on Wikipedia or engagement with WikiJournals could be a path to surfacing some of the archive.
We supported two LGBT+ volunteers to attend Wikipedia for Peace at Europride Vienna 2019, organised by Wikimedia Austria. This included taking and uploading photos for the Wiki Loves Pride competition, an editathon across a range of languages, and participating in Wikimedia’s LGBT+ User Group meeting – advising on future plans, and inputting to the development of criteria for global resource allocation.
We have also been working with West Yorkshire Queer Stories (a project collecting LGBT+ histories in West Yorkshire; the largest UK project of its kind). They are interested in using the content collected to improve Wikipedia coverage of the subject, and we will be exploring this with them going forward.

Innovation

MA students with an interactive map of Scottish witch trials, one of the outputs from processing the witches database.

As mentioned, an increasing amount of our work on language draws on technology. We add this angle to our partnership conversations to enrich and boost our collaborations. LingWiki is one example. With the increased interest in databases from the cultural sector, we are having more conversations around Wikidata and under-representation of knowledge (for example National Gallery, Natural History Museum). For example, the Digital and Data Driven Aberdeen History Group is an emerging group exploring tech for heritage preservation and dissemination, with a lot of interest in Wikidata.
Introducing our new 2019-22 strategy gave us an opportunity to reflect on the metrics we have been tracking, and experiment with new ones. New measurements we have been trialling are connected to impact areas of knowledge equity and digital literacy, and we hope that they will give us new insights into these areas of our work. The process gave us an opportunity to reflect on how under-representation translates to our programmes. For example, analysing the diversification of content producers could be done in a range of ways, potentially looking in depth into diversity statistics tracking or monitoring the number of events we run which have a defined underrepresented group as the target audience.

We have been looking to increase the capacity of our Wikimedians in Residence by recruiting interns, particularly as many of our host organisations, run established internship programmes. The University of Edinburgh had a student internship for the Witchcraft project (described in the detailed case study below), organised through Equate Scotland's scheme to provide paid placements for Women in STEM students. Emma Carroll worked for three months geolocating the places mentioned in the Survey of Scottish Witchcraft database so the accused witches of Scotland could be plotted on a map, making over 47,000 edits to Wikidata in the process.

Changes, Challenges and Learning

Change

Compare to our proposal.

The increasing significance and potential of Wikidata means that we have put a greater focus on this in our partnership work over the past year, and have started exploring more systematically the opportunities that technology offers to our work on underrepresented knowledge. This will be supported by the recent creation of a Technology Advisory Board at Wikimedia UK - made up of staff, trustees and volunteers - who will oversee all aspects of technology for the organisation. Whilst this wasn't highlighted in our proposal, it is a very welcome development.

Challenge

Early in the year we worked through a challenge within our programme in Wales. In 2016 we reached gender parity in biographies on Welsh Wikipedia, which was a first across the movement for Wikipedias with more than 10,000 articles and a very proud achievement. We have been working to try and maintain this balance, by (for example) encouraging editors to create female biographies if they have also written an article about a man. This piecemeal process was somewhat disrupted by the Wici People project delivered by our partner organisation the National Library of Wales, which generated 2000 new biographies. Most of these new articles on the Welsh Wikipedia were about men – such is the nature of the source material within the library – which threatened to skew the site’s hard-won gender parity. To overcome this, volunteer editors worked hard to create a corresponding number of female biographies; however this was a significant effort. We have now agreed with the Library that where possible, future externally-funded projects of this kind will consider aspects such as gender balance and quality, as well as article creation numbers. The situation led to reflections on the complex relationship between targets and programme delivery, and reinforced the importance of working with the Wikimedia community in this delivery.

Another challenge is related to knowledge equity. Working with underrepresented topics poses a constant challenge around the content being rejected on Wikipedia. The Bolton Pride project is an example. We partnered with Bolton Museum and Library in autumn 2019 to hold an editathon coinciding with Bolton Pride. At a drop-in session we had an engaged audience, who enthusiastically set about improving Wikipedia’s coverage of LGBT issues, especially in North West England. One of the outputs, an article on Bolton Pride, was deleted with the reasoning that it relied too heavily on local news sources, but has since been restored and rewritten. The event has given the LGBT community plenty to think about, not least coverage of bisexual history, and they are interested in holding similar events in the future.

Learning

Banner Repeater Wikibase partnership is a new area both for us and to an extent the Wikimedia movement. This means that a lot more exploratory work is needed, and we have had to hold several meetings to discuss the fairly specialised subject of determining the correct schema for the complex and diverse movement of artists’ publishing. We also connected the project to the Royal College of Arts, drawing on their technical and design skills, and run a wider consultation around data schema and linked-data compatibility. This is an emergent area of work - there is a lot of interest about Wikibase in the cultural sector, but the Wikimedia movement does not have a tested model of how to engage institutions or teach it successfully. We are hoping to develop something here that can be reused by others.

On a meta level, we will continue to explore and learn from our new metrics in this programme area; we see them as an important development in being able to track progress towards knowledge equity within our work. We will continue to learn about how to interpret our results, and also explore other ways of capturing this work.

Future

Brexit will have far reaching consequences for the UK’s economy, society and legislative agenda for decades, and may lead to civil unrest as well as a significant economic downturn. It may also lead to the eventual dissolution of the United Kingdom, as there are increasing calls for a second independence referendum in Scotland. The relationship with Northern Ireland may also change. This may affect the shape and role of Wikimedia UK, while potentially giving us an opportunity to work on affiliate subsidiarity in our context.
The ongoing diversification of the UK and our position as one of the most multicultural countries in Europe – as well as the largest Wikimedia chapter for the English Wikipedia – means that there is an ever growing need to include a wider range of narratives and perspectives in our efforts to make knowledge creation more open and inclusive. The drive towards knowledge equity in the global strategic direction resonates with an increasing awareness amongst UK content holders of the need to represent diverse stories and histories. Wikimedia UK has a crucial role to play in opening up and sharing content, and in supporting the decolonisation of cultural collections and education curricula. Equity and decolonisation comes up in most, if not all, of our conversations with content-holding partner organisations, and we will only see this trend grow. As a team we watch this sector conversation closely, so that we can speak to it and offer insights into how Wikimedia can help improve the issues identified.
As mentioned, the UK-wide National Collections project is potentially significant and would draw our focus to working with Wikidata as a key collections data tool.

Case studies

Survey of Scottish Witchcraft, & the University of Edinburgh’s Data Visualisation internship

One of the biggest ScotWiki success stories of 2019 was that of the University of Edinburgh’s Data Visualisation internship, funded by Equate Scotland’s scheme to provide paid placements for women students in STEM subjects. This focussed on the Survey of Scottish Witchcraft, gaining national and international press coverage, and contributing to the ongoing call for a national monument to those killed as witches during the early modern period in Scotland. In the first three months of its launch - https://witches.is.ed.ac.uk/ - the site had over 115,000 users and 200,000 page views. The project, which saw information from the original piece of research uploaded to Wikidata, and then visualised through an interactive map, gained a great deal of press attention (both national and international), saw renewed interest in the dataset. The project gained a great deal of press attention, with the Scotsman, Herald, Guardian, Scottish Daily Mail, Vanity Fair Spain, Evening Express, Edinburgh News, Press & Journal, BBC Radio Scotland, BBC RAdio 4, New York Time,s John O’Groat Journal and Smithsonian.com all covering the story.

What’s been clear from reading the many pieces of coverage (and indeed, the many comments on social media sharing the website) is that the site caught the imagination not just of academics and those with an interest in linked and open data, but of the general public too. Arguably, this is because of the extremely low barrier to entry in terms of understanding the data, and the ability to see the detail of accused witches in your area which helped the user to connect with the data.

- summary by Dr Sara Thomas

A version of this article appeared in Broadsheet, the magazine of the Scottish Council on Archives, no 51, December 2019.

Putting Scotland’s accused witches on the map with Wikidata.

(By Ewan McAndrew, Wikimedia in Residence at the University of Edinburgh.)

A new interactive map (witches.is.ed.ac.uk) tracks more than 3,000 Scots accused of being witches in the 16th and 17th century. Open data experts at the University of Edinburgh’s Information Services directorate have been building profiles using linked open data of all the women and men who were accused of practicing witchcraft as part of an internship project which has breathed new life into an old dataset, caught the public’s attention and helped to change the way the stories of these women and men were being told.

The map is built upon a landmark Edinburgh study that highlighted the plight of the many people –overwhelmingly women (85%) – who were strangled and burned at the stake. Now users can move through a map of Scotland to see where the accused witches lived, as well as the towns and villages where they were detained, punished and executed.

The Survey of Scottish Witchcraft Database Project was led by Professor Julian Goodare, lecturer at the School of History, Classics and Archaeology and Dr Louise Yeoman, formerly a curator at the National Library of Scotland, later a researcher with BBC Scotland. This fabulous resource began life in the 1990s before being realised in 2001-2003. It had as its aim to collect, collate and record all known information about accused witches and witchcraft belief in early modern Scotland (from 1563 to 1736) in a Microsoft Access database and to create a web-based user interface for the database. It’s a data set that has the power to fascinate as an extremely rich historical resource.

Since 2003, the Survey data has remained static in the Access database, and so students on the University of Edinburgh’s Data Science for Design MSc, at the course’s annual “Data Fair” event, were invited to consider what might be done if the data were exported into Wikipedia’s sister project, Wikidata, and linked to other datasets?

Happily, two groups were enthused, both in 2017 and again in 2018, to work on this assignment and busied themselves analysing, modelling and processing the data for import before showing their end-of-project visualisations. The success of this ‘Data Fair’ model, where researchers have three minutes to pitch “data challenges” involving real datasets for the Masters students to work on, in groups of three, over a six week project, prompted questions as to what more could be done with the Survey of Scottish Witchcraft data and whether funds could be leveraged to hire a student with good GIS skills who would take the work done to date on to the next level. The idea being to work over a more extended period to track down all the places mentioned in the database and geolocate them on a map for the first time.

This interactive map was conceived and built by student intern Emma Carroll, who worked for three months collating the historical information and plotting the locations on the map of Scotland, making over 47,000 edits to Wikidata. Emma’s Data & Visualisation internship was organised by Equate Scotland Careerwise – an initiative that arranges paid placements in industry to women working in STEM subjects.

The task demanded a lot of detective work as many of the places recorded do not exist anymore or have changed over time.

Emma worked closely under the direct supervision of the University’s Wikimedian in Residence, Ewan McAndrew, who helps staff and students develop digital skills and information literacy through improving the quality of information online through Wikipedia and its sister projects.

The project was a collaborative one with Emma’s detective work involving colleagues from the National Library of Scotland, the Scottish Studies Archive, the Scottish Place-Name Society, and the website creation itself drew in the creativity and expertise from the university’s Interactive Content team and e-learning developers. The project also involved academics from Edinburgh Futures Institute, as well as University researchers working in the fields of Data Driven Innovation and Digital Humanities.

Emma collaborated with Wikidata experts across the UK, blogged every week of her three month project and produced a new ‘beginners guide’ video tutorial on how to process & reconcile data with the open source software, OpenRefine.

Since the map’s launch in September 2019, the project has gained media coverage across Scotland and the world and shows the potential of engaging with linked open data for students, educators, researchers, data scientists and repositories of all kinds as this resource is a free, open and stable knowledgebase available to everyone. Beyond this, anyone can help connect to and leverage from a variety of other complementary datasets in multiple languages; helping to fuel discovery through exploring the direct and indirect relationships at play in this semantic web of knowledge, enabling insights in a variety of disciplines.

The hope is that this project will both aid students’ understanding of data literacy through the practical application of working with a real-world dataset and help shed new light on a little understood period of Scottish history. This, in turn, may help fuel discoveries by dint of surfacing this data, making it queryable and linking it with other related datasets across the UK, across Europe and beyond. As the Survey of Scottish Witchcraft’s website states itself “Our list of people involved in the prosecution of witchcraft suspects can now be used as the basis for further inquiry and research.“

Impact:

Ewan gave a talk at the Suffer the Witch symposium in Edinburgh, where the Witches project once again contributed to the continued calls for a new memorial to accused witches in Scotland. He also spoke to a new group interested in undertaking a similar kind of project regarding clergy in during the Scottish Reformation. The Map of Accused Witches is now on display on the University Library’s large digital display wall, alongside the Histropedia timeline for Women’s Suffrage in Scotland, crowdsourced from University of Edinburgh Women in Red editors, Commons uploads and structured data in Wikidata.

Furthermore, the Map of Accused Witches project won the Digital Humanities Award in category for best Data Visualisation.

Celtic Knot conference Cornwall 2019

Wikimedia has co-organised the Celtic Knot Wikimedia Language Conference every year since its inception in 2017. It is an important opportunity for Wikimedians working in and with minority and indigenous languages to meet in person in order to share learning, inspire others, invigorate language communities and overcome challenges together. Building on the success of our previous two language conferences, in 2019 we hosted the conference in partnership with Cornwall Council, with the event taking place at the University of Exeter’s Penrhyn campus. This third Celtic Knot Wikipedia Conference aimed to show how Wikimedia can be used in innovative low-cost ways to support small, minority languages. We wanted to help smaller language communities to participate in workshops and sessions with an emphasis on transferring knowledge from larger languages to smaller languages; practical sessions for participants from smaller languages with less expertise; and general networking for language communities to make the most of Wikimedia to support their languages. The groups we focused on experience many challenges that can be hard to appreciate by large, well established languages – such as lack of official backing or disagreements on the written script. A talk by Annas Sedrati from the Moroccan User Group encapsulated these issues perfectly.

The first Celtic Knot conference primarily aimed to support the development of the six Celtic languages, however with this year’s conference we broaded the programme to explicitly include any small, regional language community interested in using Wikimedia and in linking Wikimedia projects between languages. This brought submissions from new communities as well as those we had worked with before, such as Wikidata editors making use of its multilingual applications, and Wikimedians dealing with specific challenges such as integrating oral traditions into a primarily textual website. With the focus on minority languages it was important to match up with the right community – and we were very pleased to see the Cornish group taking the lead on hosting the conference. Cornish is spoken by a relatively small number of people (estimates vary) and is classified as ‘critically endangered’ by UNESCO.

Mark Trevethan working at the Cornwall Council attended our first two Celtic Knot conferences and was keen to host the third. His presence at previous editions of the conference led to discussions with volunteers with the Welsh Wicipedia and researchers at Bangor University. Cornwall Council is aiming to double the number of Cornish speakers, and ensuring there are readily available digital resources is an important aspect of making the language accessible and ensuring its survival. The Celtic Knot has inspired more work on the Cornish language, with interest in using Wikipedia in classroom settings and at heritage sites. As well as bringing together Wikimedians separated by geography but encountering similar challenges, the conference has the opportunity to help the growth of the Cornish language. The conference was filmed so those unable to attend can still benefit from the discussions, and the sessions are available on our YouTube channel and Wikimedia Commons.

One of the aims in coming to Cornwall was to stimulate more Wikimedia activity in Cornwall and in the Cornish language, through exposing participants to the skills of those in more active communities. We acknowledge that there is a very low level of wiki activity in Cornish/Cornwall, but this is partly due to a lack of understanding about how to use Wikimedia to its full potential, for example in education. We therefore also invited people from Bangor University, who are helping to develop the online Cornish dictionary – including use of Wikidata – based on the work we’ve facilitated in Welsh.

Having attended the previous two conferences, Mark was able to see how other mid-sized languages are using Wikimedia and wanted to bring that energy and expertise to Cornwall. A big part of the conference was showing models of how wiki collaborations made things easier for small languages, and it was clear that many ideas could be tried in Cornwall. After the conference Mark reported several encouraging signals of connections as the conference seemed to have a strong impact on the people who attended. The ideas he has been exploring as a result include:

Holding editathons in English and Cornish to develop the Cornish Maritime Churches project on Wikipedia – this is very relevant to Cornish culture.
Developing more articles in Cornish with the help of Robin Owain
Using Wikipedia to provide information on Cornwall’s threatened ecology, particularly 30,000 miles of hedges which are around 4,000 years old in parts, and which provide habitats for local wildlife. This would involve starting to develop our own content, rather than translating.
Using Wikimedia in education, based on the WikiMon model – this will need some work to develop but there is a lot of interest in this. Cornwall is investing in Superfast Broadband, and has an accompanying social inclusion programme which this could potentially fit into.

We believe these are very positive developments, and are working to help the community take some of these ideas forward.

Wikimedia UK and Turkish language editing

Wikimedia UK has long encouraged the creation of content in languages other than those native to the UK, often incorporating this into events with diverse attendees, or through working with universities with foreign students.

For a couple of years we’ve focused slightly more on working with some specific communities such as Turkish and Kurdish speakers in London, and it has led to several successful and highly inclusive events. It takes time to build contacts and partners, but 2019 saw some good engagement with the large Turkish community in London.

One of the key issues for Wikipedia and Turkish language speakers has been Turkey's block of Wikipedia, cutting off access to its largest community of potential readers as well as contributors.

Partnering with freedom of expression charity PEN International - as well as Article 19 and English PEN - we ran an editing event at the Free Word Centre in Farringdon. The focus of the event was journalism and freedom of expression in Turkey, with updates made to articles.

Planning and communications around the event were sensitive, given at the time the Wikimedia Foundation was in the Turkish courts arguing for the ban to be lifted, and the argument that Turkish Wikipedia isn’t an external influence on Turkey but a representation of its own people’s ideas is not necessarily aided by any suggestion of activity happening elsewhere. We therefore worked closely with the legal team at the Foundation to ensure we were positioning the event in a way that wasn't damaging to the legal case being made at the time.

This session was surprisingly productive bearing in mind it was mainly attended by completely new editors, with useful contributions made across a range of articles in both English and Turkish and everybody contributing something to an existing article or creating a new article. This demonstrated that there is a lot of 'low hanging fruit' in Turkish Wikipedia for motivated people with access to good sources.

We ran a follow up event more generally on Turkish culture, creating and expanding content in Turkish (as well as Uyghur at one event).

Some things we noticed, or at least had confirmed for us, were that having trainers used to working on English Wikipedia pushes editors into adding references to another language Wikipedia with a generally more relaxed attitude to citations.

Another is that language communities do intermingle, and often multiple languages are represented even though the subject is one particular culture. But more generally; these are long, low intensity, relationship-building exercises that do result in valuable contributions and improvement in understanding of these resources that are, and should be, available to all.