Celtic Knot Conference 2020/Submissions/Panel discussion: COVID-19 and the Wikimedia community/Discussions
Heading The below content is an archive of the note-taking pad that was used for this session. |
Welcome to the note-taking pad of one of the Celtic Knot Conference 2020 sessions! This space is dedicated to collaborative note-taking, comments and questions to the speaker(s). You can edit this document directly, and use the chat feature in the bottom-side corner.
✨⏯️ Session details
[edit]- Name: Panel discussion: COVID-19 and the Wikimedia community
- Speaker:
- Iñaki LL (Basque Wikipedia)
- Kimberli Mäkäräinen (Northern Sami Wikipedia)
- Netha Hussain (Malayalam Wikipedia)
- Subhashish Panigrahi (Odia Wikimedians User Group)
- Link to the video/replay: https://www.youtube.com/watch?v=bjuCz5GuTkM
- More details: Celtic Knot Conference 2020/Submissions/Panel discussion: COVID-19 and the Wikimedia community
- See also: Wikidata helpdesk Celtic Knot Conference 2020/Wikidata helpdesk
💬❓ Questions
[edit]Feel free to add questions here, while or after watching the session. Please add your (user)name in bracket after the question. The host of the session will pick a few questions to ask them during the livestream. The speaker or other participants will answer on this pad (asynchronously: the answer may come in a few hours or days).
Should there be some retro charge to the various national health authorities regarding all the work ...
Even if it's just to get funding for the local wikimedia group....
- It's definitely worth using "look at all this work we did" as a message in future funding requests
- This is a great point.
🖊️🔗 Collaborative note-taking
[edit]Feel free to take notes about the session here, add some useful links, etc.
- [Introduction missed]
- Each panellist will have some time to present their own topic, but we will also have a fair amount of time at the end for a question and answer session
Subhashish Panigrahi
- As Daria has introduced, this is about the WIkimedia communities around the world and how we are coping with COvid-19 in relation to our respective Wikimedia projects
- Iñaki LL is from the Basque-language WIkipedia
- Kimberli is from the Northern Sámi language Wikipedia
Subhashish
- Netha Hussain from the Malayalam language Wikipedia in India
- Pictures from meetup in Stockholm (Wikimania 2019), community has grow up since then, with several of us looking at minority- and marginalised-language Wikimedia projects
- Kimberli proposed the idea for this panel discussion to represent the linguistic diversity of the Wikipedia
Iñaki
- I am from the Basque Wikimedians User Group. I'll be talking about the circumstances arising after the state of alarm was announced in Spain and the state of confinement in both Spain and France.
- For context, the language community lives in 3 different administrative units across France and Spain. Most of the messaging was presented in French and Spanish.
- There was confusion over the rules for confinement as, despite almost everyone understanding French or Spanish, we needed messages coming also in Basque — for the education community, older people and more generally.
- Since it was an emergency situation, all the information was coming in French or Spanish only, even in the Basque Autonomous Community, where Basque is recognised as an official languge.
- The language status is unequal. [Slide shows Basque-speaking density and official status]. The official status is variable and is not even implemented that much,.
- We relied a lot on free Commons licensing, talking advantage of summary agreements arranged at the end of 2019 and early 2020. We agreed with media organisations that their files and text could be stored in a Commons licence, so we could use them to feed Wikipedia and to share those messages also in Basque.
- This informaiton came from the only Basque-language newspaper, Berria [URL?], the Basque govt and local media, such as the weekly magazine Kronika [URL?]
- We translated informaiton here; we can see [slide] a woman giving guidelines on how to protect against Covid-19. This was from a local weekly media outlet and so we could take this information onto Commons and share it in Basque on social media. This network of local media are called "Tokikom" [URL?]
- We saw an opporutnity as well from a project started by Kimberli, to colloborate and translate graphics into Basque. We prepared this page WikiProject COVID-19 Graphics [link?] and could translate into Basque.
- Issues and opportunities
- Downsides
- Non-SVG files were difficult to edit and translate, so we relied mainly on SVGs.
- Licensing
- Need for remote volunteering
- Web based; we need betterb automatisation of SVG tasks
- Bright sides
- Networking opportunities
- derived from using volunteers in many minority languages across Wikimedia
- Expansion of Commons
- Licensing to local press, making up for the deficiencies in information and coverage of the pandemic
- Networking opportunities
- Downsides
Kimberli Mäkäräinen
- Wikimedia Finland and Wikimedia Norway; a translation & technologist.
- Hack the Crisis Finland, [Hackathon weekend in March: ]
- https://w.wiki/5fC — map of Unesco endandgered languages. Languages that are not orange (safe languages) or pink dots (extinct languages) are languages that are not high-priority for government agencies
- March: Susanna Ånäs and Kimberli realised the need and urgency for material in multiple minority languages. When they started, material was created in Swedish and Finnish, the official languages of FInland, plus English.
- [Something about] a spin off in New Zealand, ensuring information was presented in Māori
- For Finland, we created 2 language groups, the "older" minority languages, indigenous to Finland for a long time. We also looked at the "newer" minority languages, used by recent newcomers to Finland, such as Arabic, Somali, Vietnamese and so on
- 1 weekend, 2 people, 22 potential languages. It quickly became apparent that of the top 8 languages we could only manage the 3 Saami languages since we couldn't get a hold of people because of the weekend. We managed 11 languages from the list of 22 plus an extra 4 bonus languages.
- Thanks to the information about how to translate these infographics being propagated through social media, we found that other languages were also being translated to.
- Northern Sámi was important because the first case in Finland was a tourist in the north. In Norway, the first case was also in the Sámi-speaking areas. The north of Sweden is particularly hard hit right now.
- Localised: We changed the text from talking about not going to barbecues to talk about not going to ski centres, because that's what people started doing when the pandemic arrived in FInland
- 9 spin off animations, 30 languages
- In the Covid-19 guidelines by language, we have 94 categories of guidelines in 94 languages (plus 2 more languages not yet categorised). This is thanks to initial infographics and guidelines being released under licences we could use on Commons.
- [Licensing issues]
- We also had the challenge that we needed to create new terms for things that didnt exist in the language before, such as "hand sanitiser"
- Professional translators don't get paid if we're doing crowdsourcing, which is obviously an issue too
- Before the next epidemic, we really need to have response team with multiple skillsets — language skills (translation, proofreading, terminologists), graphic design, medical expertise, contacts with national health agencies, contacts with WHO
- font issues mentioned; Google Noto was suggested as a solution, but even it doesn't work with Skolt:
- ʹʼâčʒǯđǧǥǩŋõžšåäö (https://fonts.google.com/specimen/Noto+Sans?preview.text=%CA%B9%CA%BC%C3%A2%C4%8D%CA%92%C7%AF%C4%91%C7%A7%C7%A5%C7%A9%C5%8B%C3%B5%C5%BE%C5%A1%C3%A5%C3%A4%C3%B6&preview.text_type=custom).
- The first character is one of the most commonly used chars and it comes up as a box. In regular and bold italics, the first two chars are missing.
- ʹʼâčʒǯđǧǥǩŋõžšåäö (https://fonts.google.com/specimen/Noto+Sans?preview.text=%CA%B9%CA%BC%C3%A2%C4%8D%CA%92%C7%AF%C4%91%C7%A7%C7%A5%C7%A9%C5%8B%C3%B5%C5%BE%C5%A1%C3%A5%C3%A4%C3%B6&preview.text_type=custom).
- font issues mentioned; Google Noto was suggested as a solution, but even it doesn't work with Skolt:
Subhashish
- Odia Wikimedians user group, from the eastern part of India.
- The state of Odisha has a long coastline and there are a lot of migrations have happened in the recent past, particularly to neighbouring states, labelled on the map. As a result, lots of people living outside the province headed back to Odisha when the pandemic started. Obviously, with migration means that Covid-19 cases came with them.
- Odisha is a tiny state (in the context of India), with 45m people has many COvid-19 cases. 12k cases, ~7,500 recovery, 56+ deaths. The Odia community on Wikipedia is relatively small
- [something about high-qualiuty articles from a Wikipedian who is a retired physician]
- We also had a collaboration with the local government, who agreed to licence their Covid dashboard under CC BY 4.0 licence.
- The majority of contributions in the Commons category are originally from the state government.
- Last year the govt released a bunch of dictionaries under CC BY 4.0, this means we can extract a lot of tet, image and video content from their site.
- We saw this opportunity last year, with a tool allowing people to upload content from speocifically0-licenced Twitter accounts into Commons. This meant we could import from the govt's and the health department's Twitter accounts.
- A doctor in the community u[name?] has been active in the community for many years and has contributed a lot of content about Covid-19. He has also made several videos, the slide here shows a snapshot from a video where he explains about the pandemic — very useful for language communities with poor literacy rates, for example.
- Issues we faced are shared as with the other communities — small numbers of editors means that a lot of work needs to be done, so new work is more likely to contain spelling mistakes for example. When there is only 1 or 2 subject-matter experts there is a lot of them to do.
- Languages lack terms for the coronavirus pandemic, so we needed to decide if we would use new terms or directly import terms from other languages
Netta Hussain
- It's interesting to see how you created partnerships with the govt even before the pandemic, meaning that you could import information quickly once the pandemic arrived.
- Netta is involved in translation initatives and with Medicine WikiProjects.
- [3 links on slide]
- Misinformation is a major problem — it has become such an issue with the Covid-19 pandemic that the WHO described is as an infodemic.
- Information about the pandemic is available in 130 [check num?] languages, which is a great achievement. But that is less than half the languages of Wikimedia and it is disproportionately likely to be a problem with minority languages — language communities that are smaller often lack the resources to create these articles. Also, as most scientific communication happens in large languages, a lack of references is very common.
- We started a translation taskforce as a part of WikiProject Covid-19 from the English Wikipedia WP:COVID-19TTF
- The first thing to translate was a summary version of the enwiki article, then proofread. These need to be translated in a culturally appropriate manner — the English WIkipedia is very Anglocentric, of course.
- It is almost important to make articles time-proof, so they do not need frequent updates that put a strain on small language communities. Being updated only once a month or so is best.
- WIkipedia language editions can use COntent Translator to help with this.
- The TTF has a short and long list of articles.
- The TTF desperately needs experts to help create the short version of English-language information to be translated
- SWASTHA (the Sanskrit word for health, an acronym for Special Wikipedia Awareness Scheme for The Healthcare Affiliates). This taskforce watches Covid-19 related articles to ensure that experts who do not know WIkipedia editing can help fight misinformation in articles.
- meta:WP:COVID-19INDIC localised images, graphics, labels and charts into Indian languages. They also assess the quality of all Covid-related articles.
- While we are mostly aware of the gap between English and non-English WIkipedias, that gap in images is not so widely known and is particularly relevant for pandemic information, where topics like social distancing
Panel discussion
- [Missed question about notifications to users to return to their engagement with these projects]
- Netha: I only know about TTF projects
- Subhashish: Kim, you work with a lot of languages that have a small internet presence, so they might not have an article about "virus", for example?
- Kim: We had to create some of those articles, definitely — virus, epidemic, pandemic and so on. It's a huge rabbit-hole you can disappear into. But you have to remember that getting articles translated is urgent, so we started with Covid-19 in Northern Sámi. In the Inari Saami community, they started with the Spanish flu (https://incubator.wikimedia.org/wiki/Wp/smn/Espanjatavdâ) and then "pandemic" (https://incubator.wikimedia.org/wiki/Wp/smn/Pandemia), as they had been heavily affected by the "Spanish flu" pandemic from 1918
- Subhashish: [question missed]
- Kim: We had the fortunate situation where the elders in the Sámi community are very knowledgeable about viruses, for example. Other communities have had problems with elders spreading misinformation about what can be done in these kinds of cases; I don't know how those have been solved, I'm afraid. Social distancing messages of "don't go visit your elders to protect your elders" was a useful way of phrasing things
- Subhashish: Question to anyone. Should there be some retrospective charges to national authorities for all this work, even if just to get funding for local Wikimedia groups?
- Iñaki: For Basque WMUG is very associated with the education programme, which receives some funcding from the autonomous region's govt. This work was not done within that framework, it was entirely volunteer effort. The information being put out byt the Basque Aut. Govt. was not always relevant to other Basque-speaking communities. [something about media] also working on graphics was on a voluntary basis. I;m not sure we could go back and request money for that. I would have to give that more thought; perhaps other panellists might have thoughts
- Kim: It would be useful to get something in place ahead of the next epidemic
- Netha: WM Sweden had success in bringing posters in 24 [num? languages into Commons (https://commons.wikimedia.org/wiki/Category:COVID-19_instruction_posters_from_the_Swedish_Public_Health_Agency).
- Subhashish: Standards and licensing need to be agreed beforehand — everyone is too busy during a pandemic
- Subhashish: During emergencies, communities often lack the time to translate things manually. Do people use machine translation to speed up editors' work?
- Iñaki: We have a machine translation engine integrated with the Content Translation extension, which has been tested 1–2 years ago, called "Matxin", which was not so good. More recently, applications are being developed more rapidly, using "neural translation". Since Basque is a language isolate, it is not like translating between closely-related languages like es ↔ ca.
✨✨✨✨✨
More information about the Celtic Knot Conference 2020: https://meta.wikimedia.org/wiki/Celtic_Knot_Conference_2020
The Friendly Space Policy also applies on this space: https://meta.wikimedia.org/wiki/Celtic_Knot_Conference_2020/Friendly_Space_Policy