Hoi, Again, your English Wikipedia background shows. Wikipedia exists in 200+ languages, knowledge panels in English are of interest to some 50% of our public. The aggregated data at Wikidata is able to service all the languages we support. For the nds.wikipedia.org just recently, it was announced that they now support Listeria lists. Given a mission of "sharing the sum of all knowledge", there is more data to share in Wikidata than there is in all the Wikipedias combined.
When you consider the data stream provided to those who want to pay for it; this data stream makes much of the bot traffic we used to have redundant. So it is not as if any party could not get at the data, this way is more convenient to us and to them. Thanks, GerardM
On Mon, 27 Sept 2021 at 15:03, Andreas Kolbe jayen466@gmail.com wrote:
The Cunctator's point about Wikidata's copyright-free CC0 licence is actually one issue that I had meant to include in the list of WMF ethical lapses in my other post .
Wikidata has imported very large amounts of content from Wikipedia, which, as The Cunctator points out, has a Creative Commons Attribution-ShareAlike License. This was not the original plan, as it was thought doing so would infringe the licence under which Wikipedia contributors had released their contributions.
In 2012, for example, while he was still a Wikimedia Deutschland employee, Denny wrote on Meta,[1]
Alexrk2, it is absolutely 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? --[[User:Denny Vrandečić (WMDE)|Denny Vrandečić (WMDE)]] ([[User talk:Denny Vrandečić (WMDE)|talk]]) 12:39, 4 July 2012 (UTC)
Denny then moved to Google in October 2013, and subsequently argued strongly in favour of making Wikidata CC0, which is the viewpoint that prevailed and led to large-scale importation of Wikipedia content in Wikidata.
The legal situation is admittedly fairly complex[2] but it stands to reason that when a person moves from Wikimedia to Google, loyalties and priorities will change along with such a move. That is only natural. Nobody would or should make such a move if they weren't prepared to be loyal to their new employer. (I've cc'ed Denny as a courtesy.)
What is equally certain is that the CC0 licence served the interests of Google and other Big Tech companies. All of this of course happened at a time when Google and Silicon Valley were particularly strongly represented on the WMF board.[3]
As far as the Wikimedia projects are concerned, Wikidata's shift to CC0 substantially increased the risk of disintermediation that Guillaume mentioned in his post. If content is CC0, there is no need for attribution, so unlike the present Knowledge Graph panels, which at least have a link to Wikipedia, there is no need for any attribution to a Wikimedia site at all when others use Wikidata content.
Content is then widely disseminated and presented as truth without any indication that it comes from a Wikimedia volunteer project. As Heather Ford has pointed out in her chapter of the Wikipedia @ 20 book, "Rise of the Underdog"[4], this obscuring of provenance is undesirable for other reasons as well – it becomes harder to contest information. Users lose agency.
Now, in the context of the grand aim of Knowledge Equity, I believe it is absolutely the wrong thing for the WMF to enter into any association with Big Tech companies that results in any preferential treatment being extended to them.
Companies like Google, Amazon, Apple and Facebook are surveillance capitalists. Their entire business model is based on tracking user behaviour. It is diametrically opposed to professed WMF core values concerning privacy and data protection.
Moreover, these companies have become trillion-dollar companies – really the 21st-century equivalent in many ways of what oil companies were in the last century, and wielding the same kind of covert influence – in part because of their diligent effort to avoid paying taxes in the countries they operate in.
The way these companies are set up, this will never change: shareholders will always demand maximum return on their investments, which necessitates minimising tax. I believe anyone who would try to change these companies' tax-avoidance behaviour, volunteering to pay the billions of dollars of tax these companies morally owe the global south and other jurisdictions, would simply be axed.
What this means, given that all these companies are based in the US, is that as their already overwhelming market share grows globally, the economic imbalance disadvantaging the global south – which is the root cause of unequal access to knowledge – will only grow. It's a bit like 21st-century colonialism: wealth streaming out of poor countries into a rich one.
For both of these reasons, privacy and tax avoidance, I believe the WMF has absolutely no business aiding these companies to any extent where it would give them any additional advantage over regional or global competitors. Of course I acknowledge that it is impossible to avoid interacting with Big Tech, but where competitors such as DuckDuckGo are available whose values are at least partially more aligned with the WMF's own, they should be clearly preferred as WMF partners.
For that reason I was really glad to read about a joint WMF/DuckDuckGo study the other day that shed some interesting light on another aspect of disintermediation. This study found that knowledge panels increase rather than diminish click-throughs to Wikipedia[5], much the opposite of what I and others thought a few years ago. While this is a single study whose conclusions may not necessarily hold true in all contexts, it is an interesting and encouraging result.
Andreas
[1] https://meta.wikimedia.org/w/index.php?title=Talk:Wikidata&diff=3876137&... [2] https://meta.wikimedia.org/wiki/Wikilegal/Database_Rights#Conclusion [3] https://en.wikipedia.org/wiki/Wikipedia:Wikipedia_Signpost/2016-01-13/News_a... [4] https://wikipedia20.pubpub.org/pub/fcgjp9ul/release/2 [5] https://diff.wikimedia.org/2021/09/23/searching-for-wikipedia-duckduckgo-and...
On Sun, Sep 26, 2021 at 7:35 AM Gerard Meijssen gerard.meijssen@gmail.com wrote:
Hoi, Facts are in and of themselves not copyrightable. Collections of data may be copyrighted.We choose not to and as a result Wikidata is the powerhouse that it has become.The CC-by-sa license is our license of choice for Wikipedia however, the way it has been enforced so far has been defensive, we are quite happy when our material is used.
At Wikidata we are long past the point where the majority of the data is from a Wikipedia. From day one Wikidata has provided essential services to every Wikipedia, Wikidata can provide superior services to Wikipedia. Because like Commons, we have to maintain the data only once and have it available everywhere. Wikidata is instrumental in sychronising death information among our projects. It has been shown over and over again to have more complete information as can be found in Wikipedia lists and categories. Wikipedians choose to stick with their arguably substandard practices.
The notion that a Google or an Amazon are not capable of extracting facts from a Wikipedia is silly. They have the capacity and the skills and the software to do just that. Wikidata provides them additional information making their information more complete. They have their reasons to be model citizens and contribute to the Wikimedia Foundation. We now provide paid for services to them making their bot activity less of a strain to our services and provide them a (paid for) service. Thanks, GerardM
On Sun, 26 Sept 2021 at 04:32, The Cunctator cunctator@gmail.com wrote:
Sorry for not being explicit; the connection is that protection of copyleft would be inconvenient to major endowment donors such as Google and Amazon. WikiData is a Wikimedia project that converts copylefted content into (what Wikimedia asserts to be) copyright-free content.
On Thu, Sep 23, 2021 at 5:40 PM Andy Mabbett andy@pigsonthewing.org.uk wrote:
On Thu, 23 Sept 2021 at 19:27, The Cunctator cunctator@gmail.com wrote:
It's really disappointing to me that the Structured Data work has
been used to blow up Wikipedia's copyleft.
Your message has nothing to do with the endowment
You offer no evidence that "the Structured Data work has been used
to blow up Wikipedia's copyleft."
- You do not explain what you mean by "blow up Wikipedia's copyleft."
If you wish to discuss copyright and/or structured data, please start a new thread; and be clear there about the point you wish to make.
-- Andy Mabbett @pigsonthewing https://pigsonthewing.org.uk _______________________________________________ Wikimedia-l mailing list -- wikimedia-l@lists.wikimedia.org, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l Public archives at https://lists.wikimedia.org/hyperkitty/list/wikimedia-l@lists.wikimedia.org/... To unsubscribe send an email to wikimedia-l-leave@lists.wikimedia.org
Wikimedia-l mailing list -- wikimedia-l@lists.wikimedia.org, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l Public archives at https://lists.wikimedia.org/hyperkitty/list/wikimedia-l@lists.wikimedia.org/... To unsubscribe send an email to wikimedia-l-leave@lists.wikimedia.org
Wikimedia-l mailing list -- wikimedia-l@lists.wikimedia.org, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l Public archives at https://lists.wikimedia.org/hyperkitty/list/wikimedia-l@lists.wikimedia.org/... To unsubscribe send an email to wikimedia-l-leave@lists.wikimedia.org
Wikimedia-l mailing list -- wikimedia-l@lists.wikimedia.org, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l Public archives at https://lists.wikimedia.org/hyperkitty/list/wikimedia-l@lists.wikimedia.org/... To unsubscribe send an email to wikimedia-l-leave@lists.wikimedia.org