As violence continues in Jerusalem, it is worth remembering what people on both sides are fighting about is not just land or access or rights, but also symbolism.

This ancient city has many contested meanings attached to it. Israelis call it Yerushalayim in Hebrew, while Palestinians refer to it as Al Quds. Both sides also see it as the capital of their respective states.

Pretty much nobody else sees it as the capital of any state. The city does not host one single foreign embassy. But if you type “Jerusalem” into Google, you’re likely to see an infobox that declares the city as the “capital of Israel.”

The Wikimedia Foundation, for instance, has built a project called Wikidata that aims to turn a lot of the information from Wikipedia into linked, structured, data. When new statistics for the population of Nairobi, or any other city, are released, these can be edited into Wikidata and then propagated to many of the hundreds of versions of Wikipedia, instead of humans slowly, manually, editing each of those versions.

Google is trying to go a step further by building what they call a Knowledge Graph: a knowledge base that can gather information from Wikidata, Wikipedia, Freebase (another user-generated knowledge base), and a range of other sources. On releasing it in 2012, Google described the system as “a critical first step towards building the next generation of search, which taps into the collective intelligence of the web and understands the world a bit more like people do.”

This is the relevant bit:

First, because of the ease of separating content from containers, the provenance of data is often obscured. Contexts are stripped away, and sources vanish into Google’s black box. For instance, most of the information in Google’s infoboxes on cities doesn’t tell us where the data is sourced from.

Second, because of the stripping away of context, it can be challenging to represent important nuance. In the case of Jerusalem, the issue is less that particular viewpoints about the city’s status as a capital are true or false, but rather that there can be multiple truths, all of which are hard to fold into a single database entry.

Finally, it’s difficult for users to challenge or contest representations that they deem to be unfair. Wikidata is, and Freebase used to be, built on user-generated content, but those users tend to be a highly specialized group—it’s not easy for lay users to participate in those platforms. And those platforms often aren’t the place in which their data is ultimately displayed, making it hard for some users to find them. Furthermore, because Google’s Knowledge Base is so opaque about where it pulls its information from, it is often unclear if those sites are even the origins of data in the first place.

Jerusalem is just one example among many in which knowledge bases are increasingly distancing (and in some case cutting off) debate about contested knowledges of places. Google searches conducted in London show a range of places in which Google’s databases pick sides in contested political situations. A search for “Londonderry”(the name used by unionists) in Northern Ireland is corrected to “Derry” (the name used by Irish nationalists). A search for Abu Musa lists it as an Iranian island in the Persian Gulf. This stands in stark contrast to an Arab view that the island belongs to the United Arab Emirates and that it is instead in the Arabian Gulf. In response to a search for Taipei, Google claims that the city is the capital of Taiwan (a country only officially recognized by 21 U.N. member states). Similarly, the search engine lists Northern Cyprus as a state, despite only one other country recognizing it as such. But it lists Kosovo as a territory, even though it’s formally recognized by 112 other countries.

My point is not that any of these positions are right or wrong. It is instead that the move to linked data and the semantic Web means that many decisions about how places are represented are increasingly being made by people and processes far from, and invisible to, people living under the digital shadows of those very representations. Contestations are centralized and turned into single data points that make it difficult for local citizens to have a significant voice in the co-construction of their own cities.

The potential problem is, if search engines and other platforms come to rely on Wikidata to the same degree people have become happy to plagiarise Wikipedia, then an edit-war won by anonymous accounts in an obscure corner of the internet (i.e. Wikidata) could, with one fell swoop, redefine truth for the entire internet.

It seems a remarkably vulnerable system, especially when you bear in mind that at present 80% of Wikidata is either unreferenced, or only referenced to some language version of Wikipedia.

Coincidentally, I just made a similar point here: viewtopic.php?f=8&t=7113&p=164924#p164912

I'm sure they must be aware of this potential weakness. Do you know of any discussions about addressing it, or any steps taken?

It's been on the Wikimedia mailing list for the past few days.

It's been on the Wikimedia mailing list for the past few days.

That Gerard Meijssen guy is especially guilty of this, and he seems to be a really uppity tool, doesn't he?

That's "eventualism". "Quality is terrible, but eventually it will be great, because ... we're all trying, and it's a wiki!"
To me that sounds more like religious faith or magical thinking than empirical science.

Such controls would appear absent at Wikidata today, given that the site managed to tell the world, for five months in 2014, that Franklin D. Roosevelt was also known as "Adolf Hitler".

Another monumental exposé by Andreas in the most recent signpost op-ed.

Markus,

On Wed, Dec 9, 2015 at 10:32 AM, Markus Krötzsch <
markus at semantic-mediawiki.org> wrote:

What this page suggested was that that Freebase being shutdown means that
> Google will use Wikidata as a source. Note that the short intro text on the
> page did not say anything else about the subject, so I am surprised that
> this sufficed to convince you about the truth of that claim (it seems that
> other things I write with more support don't have this effect). Anyway, I
> am really sorry to hear that this quickly-written intro on the web has
> misled you. When I wrote this after Google had made their Freebase
> announcement last year, I really believed that this was the obvious
> implication. However, I was jumping to conclusions there without having
> first-hand evidence. I guess many people did the same. I fixed the
> statement now.
>
> To be clear: I am not saying that Google is not using Wikidata. I just
> don't know. However, if you make a little effort, there is a lot of
> evidence that Google is not using Wikidata as a source, even when it could.
> For example, population numbers are off, even in cases where they refer to
> the same source and time, and Google also shows many statements and sources
> that are not in Wikidata at all (and not even in Primary Sources).
>
> I still don't see any problem if Google would be using Wikidata, but
> that's another discussion.
>
> You mention "multiple sources".
> {{Which}}?
>
> Markus
>


For the record, here is what your university webpage used to say.[1]

---o0o---

Wikidata is the free, collaborative knowledge base behind Wikipedia and
many other Wikimedia projects. The Web site has been online since late 2012
and has since become an important data provider for Wikipedias in all
languages. Ten thousands of users have contributed statements about
millions of entities. In December 2013, Google announced that their own
collaboratively edited knowledge base, Freebase, is to be discontinued in
favour of Wikidata, which gives Wikidata a prominent role as an inut for
Google Knowledge Graph. The research group Knowledge Systems is working in
close cooperation with the development team behind Wikidata, and provides,
e.g., the regular Wikidata RDF-Exports.

Development of Wikidata started in April 2012 with a team of developers
based on the Berlin offices of Wikimedia Germany. The project was heavily
inspired by Semantic MediaWiki and Markus Krötzsch has been acting as an
architectural advisor to the project since its inception.

---o0o---

You were well placed to know. The source I quoted in the op-ed was a
different one though, a snippet from an IRC chat[2].

---o0o---

16:33:55 <dennyvrandecic> also, Wikidata is not a free ticket into the
Knowledge Graph as Freebase was
16:34:07 <dennyvrandecic> it is just one source among many
16:34:27 <Lydia_WMDE> i think we really need to highlight this
16:34:30 <dennyvrandecic> benestar: actually I think that companies editing
Wikidata might be very beneficial
...

---o0o---

As a Google employee working on Wikidata, Denny can be presumed to know
what is and isn't a source for the Knowledge Graph.

Noam Shapiro in SEJ commented on the above IRC chat, saying:[3]

---o0o---

As one of the insiders notes above, “Wikidata is not a free ticket into the
Knowledge Graph as Freebase was.” It may very well be that the direct
relationship observed between Freebase and the Knowledge Graph will not be
replicated in Wikidata’s relationship with the Knowledge Graph. That being
said, it is still “one source among many,” and likely an important one.
After all, the Knowledge Graph thrives on the existence of structured data,
and - especially in the absence of Freebase - that is exactly what Wikidata
provides.

---o0o---

In May of this year, Tony Edward published an article in Search Engine Land
titled "Leveraging Wikidata to gain a Google Knowledge Graph result".[4]

---o0o---

Back in December 2014, Google anounced that it would be shutting down
Freebase <http://wiki.freebase.com/wiki/Main_Page>, a repository of
structured data that helps power Google’s Knowledge Graph, and working to
migrate all its data to Wikidata.

But how does Wikidata measure up? How can marketers leverage Wikidata to
help a business become an entity and gain a Knowledge Graph result? I have
personally had success with gaining Knowledge Graph entries for my clients
and myself. Below, I have outlined the steps you can take to both gain and
enhance a Knowledge Graph result. [...]

---o0o---

Another article in Search Engine Land, by Barry Schwartz, reporting on the
closure of Freebase:[5]

---o0o---

This means that the data won’t be lost but instead will be transferred to
Wikimedia Foundation’s project Wikidata, which will have their own API to
so that developers who want to retrieve facts automatically, as they did
with Freebase, can still do so. This would include Google also pulling
data from Wikidata, to help power its Knowledge Graph.

---o0o---

There are more articles like that ... I actually only came across your
university web page after I'd written the op-ed.

One other point. Denny said today on the Kurier talk page in the German
Wikipedia that he stands by his opinion, quoted earlier in this thread,
that Wikidata, being under the CC0 licence, must not import data from
Share-Alike sources. It would be irresponsible to do so, he said.[6]

If Wikidata with its CC0 licence must not import data from Share-Alike
sources, then I don't understand why there are mass imports from Wikipedia,
which is a Share-Alike source.

[1] https://archive.is/O8h8K
[2] https://archive.is/LoQXX#selection-2479.0-2519.74
[3] http://www.searchenginejournal.com/wiki ... ph/130459/
[4] http://searchengineland.com/leveraging- ... lt-219706/
[5] http://searchengineland.com/google-clos ... aph-211103
[6] https://archive.is/bu9Io#selection-12005.450-12005.662

According to Denny, Wikidata, under its CC0 licence, must not import data
from Share-Alike sources. He reconfirmed this yesterday when I asked him
whether he still stood by that.

In practice though we have Wikidata importing massive amounts of data from
Wikipedia, which was a Share-Alike source last time I looked. Isn't
Wikidata then infringing Wikipedia contributors' rights?

Why is it okay to import data from the CC BY-SA Wikipedia, but not from
European CC BY-SA population statistics?

There are inchoate and uncomfortable parallels to licence laundering here,
which I would hope is not something the WMF stands for. Could someone
please explain?

I'm unimpressed by the patronising, insulting tone adopted by Denny, Gerard and Markus toward Andreas. Very similar to the tone WMF Technical used to adopt when dealing with criticism of their project. I hope Lydia at least reads what Andreas and Mark Graham said and addresses their main questions and criticisms.

I'm unimpressed by the patronising, insulting tone adopted by Denny, Gerard and Markus toward Andreas. Very similar to the tone WMF Technical used to adopt when dealing with criticism of their project. I hope Lydia at least reads what Andreas and Mark Graham said and addresses their main questions and criticisms.

Btw, if you google for "Edward Said", google will tell you with their little info-box that he was born in 1935, in "Jerusalem, Israel".

Google have had a zillion complaints about it, but has never changed it.

Outside Wikipedia, facts can't be determined by campaigns.

I might be misunderstanding Poetlister, but I think they were implying the flip side of the logical statement. Inside Wikipedia, "facts" can be determined by campaigns. Which presumably someone did to manipulate the Wikidata (or whatever) to show the nonsensical statement.

I might be misunderstanding Poetlister, but I think they were implying the flip side of the logical statement. Inside Wikipedia, "facts" can be determined by campaigns. Which presumably someone did to manipulate the Wikidata (or whatever) to show the nonsensical statement.

The story with Jerusalem is very simple. I created the Wikidata item. The English description was "city in Israel". Then POV pushers came. Some of them wanted "city in Palestine", and others wanted "capital of Israel". Then one user, who later was elected to the board of Wikimedia Israel, canvassed a number of users in Hebrew Wikipedia. When there were too many POV pushers, I just unwatched the page, and it became "capital of Israel". Later on, someone managed to change it to smth neutral. That's it. There is nothing automatic here.

Regarding campaigns and Wikidata, in the wikimedia-l thread (Andreas then points out that the Wikidata Jerusalem page still says it's the capital of Israel.)

Lydia is working on a response on-wiki.

It's been deleted or moved. (link)

...the world is complicated and there is no single truth...

On Thu, Dec 10, 2015 at 4:18 AM Andreas Kolbe <jayen466 at gmail.com> wrote:

> According to Denny, Wikidata, under its CC0 licence, must not import data
> from Share-Alike sources. He reconfirmed this yesterday when I asked him
> whether he still stood by that.
>
> In practice though we have Wikidata importing massive amounts of data from
> Wikipedia, which was a Share-Alike source last time I looked. Isn't
> Wikidata then infringing Wikipedia contributors' rights?
>
> Why is it okay to import data from the CC BY-SA Wikipedia, but not from
> European CC BY-SA population statistics?
>
>
Andreas, what I said was that Wikidata must not import data from a data
source licensed under Share-Alike date source.

The important thing that differentiates what I said from what you think I
said is "import data from a data source". Wikipedia is not a data source,
but text. Extracting facts or data from a text is a very different thing
than taking data from one place and put it in another place. There was no
database that contains the content of Wikipedia and that can be queried.
Indeed, that is the whole reason why Wikidata has been started in the first
place.

In fact, extracting facts or data from one text and then writing a
Wikipedia article is what Wikipedians do all the time, and the license of
the original text we read has no effect on the license of the output text.

So, there is no such thing as an import of data from Wikipedia, because
Wikipedia is not a database.

I have repeatedly pointed you to
http://simia.net/wiki/Free_data
and you yourself have repeatedly pointed to
https://meta.wikimedia.org/wiki/Wikileg ... ase_Rights
so I would assume that you would have by now read these and developed an
understanding of these issues. I am not a lawyer, and my understanding of
these issues is also lacking, but I wanted at least to point out that you
are misquoting me.

Please, would you mind to correct your misquoting of me in the places where
you did so, or at least point to this email for further context?

Denny,


I quoted your statement verbatim and in full in the op-ed. Moreover, your
statement had a context. Alexrk2 had said,[1]



---o0o---

Read the above.. at least under European Union law databases are protected
by copyright. CC0 won't be compatible with other projects like OpenStreetMap
or Wikipedia. This means a CC0-WikiData won't be allowed to import
content from Wikipedia, OpenStreetMap or any other share-alike data
source. The worst case IMO would be if WikiData extracts content out of
Wikipedia and release it as CC0. Under EU law this would be illegal. As a
contributor in DE Wikipedia I would feel like being expropriated somehow.
This is not acceptable! --Alexrk2 (talk) 15:32, 16 June 2012 (UTC)

---o0o---



Note Alexrk2's three (3) specific references to Wikipedia.

Alexrk2 referred to imports of content from Wikipedia, and how it would
make her or him feel expropriated if WikiData extracted content out of
Wikipedia and released it under CC0.

You replied,



---o0o---

Alexrk2, it is true that Wikidata under CC0 would not be allowed to import
content from a Share-Alike data source. Wikidata does not plan to extract
content out of Wikipedia at all. Wikidata will provide data that can be
reused in the Wikipedias. And a CC0 source can be used by a Share-Alike
project, be it either Wikipedia or OSM. But not the other way around. Do we
agree on this understanding? --Denny Vrandečić (WMDE) (talk) 12:39, 4 July
2012 (UTC)

---o0o---



Alexrk2 specifically mentioned Wikipedia. So did you in your reply,
assuring Alexrk2 that Wikidata did not in fact plan to extract content out
of Wikipedia at all. Does this lend itself to the interpretation that you
were talking only about databases, and not about Wikipedia?

Alexrk2 then replied to you,



---o0o---

@Denny Vrandečić: I agree. But I thought, the aim (or one aim) of
WikiData would be to draw all the data out of Wikipedia (infoboxes and
such things).

---o0o---



You did not respond to that post, or participate further in that section.
And these bot imports of Wikipedia infobox contents etc. have happened and
are ongoing. They have been mentioned in many discussions. There are
millions of statements in Wikidata that are cited to Wikipedia.

Just a few days ago, Jheald said on Project Chat,[2]



---o0o---

But my own view is that we should very definitely be trying, as urgently as
possible, to capture as much as possible of the huge amount of data in
infoboxes, templates, categorisations, etc on Wikipedia that is not yet in
Wikidata -- and that (at least in most subject areas) calls to restrict to
only data from independent external sources are utterly utterly misguided,
and typically bear no relation to either what is desirable, what is
available, or what is still needed in order to utilise such sources
effectively. Jheald (talk) 23:49, 8 December 2015 (UTC)

---o0o---



It's not plausible to my understanding to argue that Wikipedia's templates,
infoboxes etc. are not "data sources" when contributors speak of capturing
"the huge amount of data" contained in them. Much of the existing content
of Wikidata consists of data extracted from Wikipedias.

If you feel I have misquoted you anywhere on-wiki, please point me to the
corresponding place (here or via my talk page in that project), and I will
do whatever is necessary.



[1]
https://meta.wikimedia.org/wiki/Talk:Wi ... or_data.3F
[2]
https://www.wikidata.org/w/index.php?ti ... =281906226

Lydia sez...



Sounds like they're setting us up for "it's okay if there are a ton of errors in Wikidata".

Oh yes, I see her personal motto now... "I help people make awesome happen."

This is doomed.

In fact, extracting facts or data from one text and then writing a
Wikipedia article is what Wikipedians do all the time, and the license of
the original text we read has no effect on the license of the output text.

The plot thickens (as it always does, when Wikimedians are presented with thoughtful criticism of their projects).

"East Jerusalem is in State of Palestine"

Mark Graham, Slate, 30 Nov 2015: Why Does Google Say Jerusalem Is the Capital of Israel?

Mark Graham, Slate, 30 Nov 2015: Why Does Google Say Jerusalem Is the Capital of Israel?

I'm pretty sure Tehran and Jerusalem aren't sister cities, or are any of the others added to Wikidata by Yamaha5.