cdn.media.ccc.de/.../wikidatacon2019-1009-eng-Experimenting_with_Wikidata_in_the_Saami_languages_hd.mp4

(Susanna) ...Wikimedia Finland,
and we have during this year

started working
with the Saami communities,

the culture and language,
starting experimenting

doing the groundwork for future projects.

(Kimberli) Well, actually she started
working this year.

I've been working since 2006 so...
(laughter)

(Susanna) Well, it's at
the end of chapter...

Yep here we go. Let's see what we have.

I don't know which one it is.

[inaudible]

So usually when we give presentations,
we realize nobody knows

what we're talking about,
the Saami languages.

So this is Norway, Sweden,
Finland and Russia.

And the yellow part--
and it starts quite far down here--

is the Saami dialect continuum
or language continuum.

And the languages
that have Wikipedias are five--

or there's actually only one,
Northern Saami Wikipedia.

And then the other languages
that we work with are six and seven,

and Jon Harald is from Wikipedia Norway,

and they work with the other ones
in Norway and Sweden

and the Northern Saami one.

Sää'mjânnam is the name
for this area in Skolt Saami.

This is somehow...

Yeah, so.

(Susanna) Oh yes, while thinking

about how to serve
these language communities,

as Kimberli was showing there--
maybe we'll go back to the map,

the biggest language community
in Saami area is the Northern Saami.

And when we think of Saami,
we think of Northern Saami,

but there are at least
eight other Saami communities

and language groups.

So we are working with two,

which is here--it's Inari Saami

as well Skolt Saami,
they both have around 300 speakers.

So we cannot expect--

now going to the next slide--

there are two different types
of language communities,

those that have Wikipedias
and therefore are served

within the Wikimedia ecosystem

and those that don't have a Wikipedia,

and therefore it's
much more difficult for them.

And we find that working
with structured data,

we can serve
these language communities as well.

So Kimberli may tell you
about this sticker that you have got.

So the sticker says--

in Skolt Saami
which is spoken by about 300 people--

it says Wikimedia Finland wishes
everyone a happy United Nations

International Year
of Indigenous Languages 2019.

And the sticker was created
for an event that we went to

at the end of August in Northern Finland.

(Susanna) So, it wasn't that easy.

So we started setting up language code

for Skolt Saami and Inari Saami
and found out that it's not

a straightforward process.

It's not really documented.

It was really, really hard
to find out how to do it.

So we made this elephant metaphor
here as a reindeer.

So there are different parts
of this Wikimedia environment

that look at some specific area
of this language,

definitions and there doesn't seem
to be an overall way

and process of how to deal
with adding your languages.

So what we did was we made
a lot of noise

and tried to ask everyone
to help us, and in the end,

we managed to first have
Skolt Saami and Inari Saami

for monolingual properties;

then to labels in Wikidata;

and then only to find out
that they wouldn't work

in structured data on Commons.

Then again after another process
for that, maybe six months after,

we find out that they wouldn't work
in Wikipedias

so I think that's still unsolved.

(Kimberli) When we first started,
you could only use Northern Saami

and Southern Saami
in Wikimedia projects.

And as a bonus part of this,
we have now the ability to use

the Finnish Romani language also

within the Wikimedia projects.

This trying to get your language--
the ability to be able to use

your language in a Wikimedia project
is not straightforward.

It's really difficult,
and when you talk to people,

they're like, "Oh yeah, I'll fix it.
It'll take me five minutes."

And then, yeah, it takes them
five minutes to fix one thing.

but then the next thing is not working,

the next thing, something else breaks,
things like that.

And if we, people who have been
in the Wikimedia projects forever,

can't figure out how this thing works

and how to get things
straightforwardly working,

then we can't expect communities--

language communities that aren't
familiar with the Wikimedia projects

to be able to figure out where to start

and how to navigate this process.

It's not possible.

And there are actual pages

that people are like, "Oh yeah,
there's a page for this."

And you're going, "But it doesn't come up
in Google Search for instance,

so it's not findable."

- Do you want to say something about that?
- (Susanna) No, that's fine.

So well we tried to come up
with some things

that should be looked into.

This is not an exhaustive list,

but well, obviously, the process
needs to be streamlined.

(Kimberli) The one that I really hate
are the language codes.

Because for instance I did research
with [inaudible]

which is a specific language of its own.

And there is no ISO code for it.

There is an ISO code for [inaudible].

And they've lumped together
two different languages

that are completely
unintelligible to each other.

And so Wikimedia projects use ISO codes
for these type of things.

And we really think
that there should be

a more fine-grained level to this.

For Skolt Saami, even though
there's only 300 people that speak it,

we have a lot of data for it.

And there's four main dialects,

and the words aren't the same
in the four dialects.

So I would really like to be able to put
this is from the Paaččjokk dialect,

this is from the Suõ´nn’jel dialect,
and that type of stuff.

But we can't do that.

We can't do that for Spanish.

We can't do it for English even.

And so something has to be done
about the language codes

in the Wikimedia projects.

Yeah, and something that started to happen

I think is to engage maybe
the broader language,

linguist language communities
into the decision-making process,

and maybe they're like the decisions
that need to be made.

The bureaucracy maybe has
to be somehow assessed.

What are the decisions that are needed
in this sphere?

Like what are the application processes?

What are the... yeah, so.

Thanks to Benjamin's presentation today,

I think PanLex needs
to be added to this too.

(laughing)

(man) We have individual ISO codes

for all the languages you mentioned.

Are you using IETF or... ?

(man) We start with [inaudible] codes
and [inaudible] codes

and then they can just get
a variety ID [inaudible].

[inaudible]

(Kimberli) Good. We'll talk
about it more in the Q&A then.

(moderator) If we can repeat
that for the stream

because it was...

(Susanna) Okay, I can't. (chuckles)

- (moderator) We can do it after.
- (Susanna) Right.

(Kimberli) So some of the ways
that we work together...

We work with the communities themselves,

and we were invited
to this 70-year anniversary

of the Skolts living in Finland.

They were relocated to Finland

from when the border was closed off.

And so they've been living in this area
for seven years,

and there was a big party going on,

and we were there.

She was working with little kids
putting in Moomin characters

in the different Saami languages
and different words like that.

Do you want to say
something else about that?

(Susanna) Yeah, just
to also pinpoint that.

We can find new ways of working
with data or language

so we can go to this--

We can go together with the communities.

We want to create participatory methods

in which we can add more information.

I think we have come up with this idea
of the term of "depictathons"

now that we can work with images
or translateathons which have been

done earlier as well,
but these are the kinds of events

together with the communities
that we can work with the language.

(Kimberli) So some
of the solutions that we have.

(Susanna) Here are two ideas
for next year that we have.

We are developing and seeing
what can be done with them.

One of them comes
as a collaborative project

together with the Saami archives

and the Saami museum in Inari
in the North of Finland,

and we could collect
cultural heritage concepts

across these Nordic countries
in different Saami languages,

but not only Saami languages
but also in the Nordic languages

because we share
a similar cultural heritage/history

that we have similar monuments.

This, of course, came up
with a Wiki Loves Monuments competition

and archeological finds
across the area are similar.

And the other one is place names,

that is a fortunate new project
starting at Wikimedia.

Norway, that we could expand
to be Pan Nordic,

to include place names in all these.

- Pan Saami.
- Pan Saami, ooh.

(Kimberli) So these are depictathons.

The Skolt Saami--
there are thousands of pictures

of the Skolt Saami in Commons.

They come from different archives,
and they have data,

the structured data on them
is basically from 100 years ago

so it's describing things
in the way that they would have been

described 100 years ago.

We don't want those,
those ways of description there anymore

because a lot of them are racist,
quite racist.

We don't want them.

The community doesn't want them.

The community wants to be able
to write what they want to say

about the pictures in their own language,

or in Finnish or Norwegian or Swedish.

And so we've been having depictathons
as an idea that--

well, we've done it.

So people can change the captions,
change the descriptions

of these pictures in Commons,

and you work with structured data
so I'll let you talk about that.

(Susanna) Yeah, and well,
let's see our next slide

because this is just as--

you all know structured data on Commons
so for you this is no news.

And I think, well from these,
we also enter delicate questions

of what are the descriptions,

but we'll come back to that.

(Kimberli) In the Northern Saami,
we've been creating

autogenerated Wikidata info boxes.

They've been pulling in data
from Wikidata

because I'm the one person
that's correcting everything

in the Northern Saami Wikipedia,

and I don't have time
to change every mayor,

the population of every country,
things like that.

So I've been really blessed
with the people

that have come up and started helping
create these info boxes.

And it's expanded the amount of knowledge

we have in the Northern Saami
Wikipedia greatly.

So this is Nils-Aslak Valkeapää,

who is one of the most famous Saami
multi-talent--he's a polymath.

I mean, he was a singer, a writer,

artist, and we now have
this info box there for him,

all of the data which is pulled
from Wikidata.

Before we had maybe three lines
and no picture.

(Susanna) And this applies specifically

of course to the languages
that have a Wikipedia.

(Kimberli) Yeah, but doesn't work
in an incubator.

(Susanna) Yep.

This is quite exciting now.

Once we have the--

well, we are not working
with lexicographical data,

like specifically.

We will extend to it,

but we are concerned mainly
about labels and items so far.

So what this makes possible
is tagging content,

museums, libraries
as well as broadcasters.

Yle, the Finnish Broadcasting Company

as they are already using
the Wikidata for tagging,

this might be an opportunity
for the small Saami languages

in the Nordic area.

And this is my opportunity to show

my project Wikidocumentaries as well

because it is a project that reads--

well, it's difficult to make the change...

Let me have [inaudible] help.

Yeah, there.

So here we have a page
in Wikidocumentaries,

which is now in English.

This is a project that consumes
information from the Wikimedia sphere.

Every item in Wikidata has a page,

or can be made into a page

or is automatically created into a page.

Then it gathers all this information
across Wikimedia projects,

and the interface exists already
in 40 plus languages,

and I would be able
to change the interface

and then see all the same data
in another language.

I could also, as you can see,
or you were able to see

in the English one,
that there is no article on this

in the English Wikipedia.

Therefore you could go to see
which languages it exists,

and this one is in Northern Saami.

So you would be able to switch
only the article language.

But also then it can also display
any language

that is encoded in Wikidata.

So we also get it
in the same page in Skolt Saami.

Although, there is no Wikipedia,

you get all the same content

in these languages.

(Kimberli) There is actually
an article about her

in Skolt Saami on the incubator,

but it doesn't work with Wikidocumentaries

because of the way
the incubator is encoded.

(Susanna) Oh yeah.

And just briefly, I'm very excited
in thinking about an app

that will gamify this
or like collecting these terms

into Wikidata.

But I haven't landed on one,
and I'm sure there are experiences

of that across this community,

and it would be interesting
to put together our thoughts on that.

(Kimberli) So there's
quite a few challenges

that we have in these projects.

This picture, if you come across it

on any Wikipedia please delete it.

It's two Finns dressed as Saami people.

It's labeled fake Saami clothing,

and people still use it
on Wikipedia projects.

I don't know why.

So we have false data.

We have racist--and with the Saami,

we have a lot of eugenics-based data.

So when they were trying to prove
that the Saami were a lower race

so they could sterilize them
and things like that,

we have a lot of that data

because that's the stuff
that comes out of archives.

Data usage--data has been used
without the consent

of the communities,

and for instance, the Skolt community
was kind of shocked to see

that their relatives are in Commons,

and they weren't very appreciative of it.

Sensitive data,
which Stacy can talk more about.

Yeah, this is used
on the Hungarian Wikipedia.

Here's that lovely picture

describing that these people
are Saami people.

Please delete it.

Yeah, this is more
what Stacy will talk about.

(Susanna) Leave it to you?

(Kimberli) Sensitive data.

TK labels--you want to talk about before.

(Susanna) You're not addressing them.

I think we could also look
into identifying content

already on Commons
or just about to enter Commons,

how to tag and identify, tag
and perhaps delete

or then find out restricting
the usage of this media.

Well, it's very short,

but let's see if we have
more opportunities to discuss that.

(Kimberli) We can skip this part.

Sorry.

I want to say that this is the week

of the Saami Language Week this week

so please feel free to use hashtags
for Saami languages.

Gæjhtoe!

(Susanna) Spä'sseb!

(Kimberli) Spä'sseb!

Takkâ.

(applause)