Thanks for the stats, Gerard. Two thoughts:
- With so many items without description I wonder why we don't have the automatic descriptions gadget enabled by default.
- There are many items without statements, but not that many articles without a category --> would it be possible to have a game that suggests instance of/subclass of based on the statements of the items in the same or in the upper category?


On Tue, Aug 19, 2014 at 10:39 AM, Gerard Meijssen <> wrote:
I am not in research. I am into making Wikidata into a reasonable resource. To achieve this I add gazillions of statements and I am happy when people focus on templates. However without the data, templates that make use of Wikidata are niche applications. Without some mature understanding of what Wikidata is at this time concentrating on templates is an exercise in building enabling technology.

My point is for the "community" to have reasonable expectations. So many people consider Wikidata to be useless. That is fine but imho it helps when the baseline of where Wikidata is at this time is understood. 

The statistics that explain this best can be found here ..

On 19 August 2014 10:27, Luca Martinelli <> wrote:

Ok, I got the point. What you probably  need to consider is that focusing on one goal does not mean at all that we have to dismiss all the others. At least, *I* do not think so.

You want to focus on research? Fine, do it. I'd like to focus on templates. That's fine too, I guess. We're both working to let Wikidata be appreciated - by separate audiences.


Il 19/ago/2014 07:57 "Gerard Meijssen" <> ha scritto:

What is the point of Wiktionary, WIkipedia, Wikispecies et al as a WMF project? Like Wikidata they all help us share in the sum of all knowledge. Wikidata already provides an application in being the vehicle for interlanguage links. 

The low hanging fruit of Wikidata is not sharing info in templates, it is in providing search results where a Wikipedia does NOT have an article. It is used for this and it does have a measurable impact.  It is "nice" to have the ambition to share data in templates but be realistic. The quality of the data in Wikidata does not merit this at this time. The "community" insists on sources and frankly it is assassine to expect that in the first few years it will be available near the level that some "demand". This is only based on the data that is there. That is the next problem we do not have enough data. We are still at the stage where we are harvesting data for the first time. Harvesting big amounts, not one item at a time.

It is important to have goals, and it is nice that at the start providing data to templates was seen as an initial goal. However it will not be like with Pallas Athena when she came from the head of Zeus in full armour. This goal is achievable and we are making big strides in that direction BUT we need smaller goals, small applications that grow our content in both quality and quantity. As I wrote on my blog, we need to think in terms of confidence in our data and not so much in sources. Amir is finishing a tool that will allow us to compare data for "humans" in the English, German and Italian Wikipedia. That will be a massive step in the right direction.

I care about Wikidata and I know that at this time those freakingly hard templates are the least of our worries. More problematic is that people think of Wikidata as a service product for Wikipedia and limit their thinking to templates. The existing search extension with WDQ is there. It works really well. It is dismissed probably because it demonstrates that ALL Wikipedias cover less than 50% of the subjects known to us. We know all of them because of Wikidata.

So yeah by all means blow the horn about our aspiration of servicing templates in those projects that can handle this. It is fine. It is not realistic and even counter productive as an aspiration when we do not appreciate the reality as we have it at this time.

On 18 August 2014 14:41, Luca Martinelli <> wrote:
2014-08-17 17:00 GMT+02:00 Gerard Meijssen <>:
> Hoi,
> Importing data from Wikidata (where do you want it??) is just one
> application. There are so many potential applications for structured data
> and Wikidata implicitly covers the sum of all knowledge as we know it (in
> the Wikimedia projects) so there are opportunities galore.
> For people "not to know how to" is a given. I do not care to know about
> Wikipedia templates because they are freaking impossibly hard.

Yet, if we don't use the Wikidata data in the "freaking impossibly
hard" Wikipedia templates, what is the point of Wikidata as a
Wikimedia project?

I remember that this project had among its first goals to help
disseminate structured data on all Wikimedia projects, in order to
relieve the less-crowded WMF projects of their burden in managing such
data and to let their few users focus on writing/translating/expanding
their articles. Now, if we don't show to people on the WMF project -
even the bigger ones - that Wikidata IS useful by helping them in
retrieving these data, what is the point of this project?

"There are so many potential application", I know, yet THIS IS ONE OF
THEM -- and in my personal and humble opinion a damn important one.

Luca "Sannita" Martinelli

Wikidata-l mailing list

Wikidata-l mailing list

Wikidata-l mailing list

Wikidata-l mailing list

Etiamsi omnes, ego non