WeRelate start looking into connecting to WikiData

Question

See

• Wikidata (genealogy) <-> WeRelate

==> we get another player in the genealogy Echosystem 

Comments

Echosystem = muckspreading system.

The internet can now turn any old junk into official truth in no time flat.

Wikipedia has this rule of "Verifiability not Truth", ie you can post any old rubbish so long as you can cite a "source" for it, even if the "source" is total tosh.

And we all correct profiles to match WikiData.  Source?  WikiData is the source!

---

I hadn´t realised WeRelate was still alive and kicking. Thought it had fizzled out some time ago.

---

I have checked it a little bit and its like WIkiTree a Wiki based tool using an older version

• Werelate Special:Version 1.7.1
• WikiTree Special:Version 1.11.1
• Wikipedia Special:Version 1.30

---

>> And we all correct profiles to match WikiData.  Source?  WikiData is the source!

Not if you are doing genealogy research and have a quality process.... 

Internet and the semantic web is just speeding up the process of finding related information you still need to do the genealogy research and hopefully you have and following something like GPS.

Lesson learned in WikiTree with all unsourced profile is that the greatest "benefit" today with internet is that we get more unsourced badly researched profiles faster....or? 


Linked Open Data what is it and why is it good
 

---

Glad that I never came across it but Wiktree instead!

---

But what about Linked Open Junk?

WikiData already needs editing.  When it's too big to be edited, what use will it be?

---

What about web junk? Isnt internet too big ;-) What use is WikiTree? Do we trust WikiTree? Just because every child has been corrected to be born after the mother can we trust it ;-)

Tim Berners Lee on the subject open Linked data

The future will tell?!?!? But its like the web you don't read and trust everything that is said on twitter and web pages... you need to select. I guess you dont trust everything inside Wikitree and not everything inside Wikitree is good genealogy...

The quality process in WIkidata is based on the following theory link...  

I have a feeling what Aleš done with FindAGrave hopefully also will get FindAgrave people interested in WikiTree and collaborate...

I would like to see that we measure quality inside WikiTree to understand

 

 

---

I thought Findagrave was now owned by Ancestry in which case I wonder how much they would be motivated to work with Wikitree.

---

I guess not Ancestry I speak about users...

Also the new FindAGrave plattform seems non visionary....

---

Now I'm lost.  If I want to select, I can just Google.  If I'm consulting a work of reference, I'm kind of hoping that somebody else has already done the selecting so I don't have to.  If Wikipedia says "let the reader decide", what's it for?

---

But Google also starts using the semantic WEB because text retrieval is not good enough see Linked Open Data What is it for me 

I'm kind of hoping that somebody else has already done the selecting so I don't have to.

Then the person selecting and you need to have the same interests.... if you are interested in genealogy I guess Von_Sydow-8 is always better than Wikipedia Oscar_von_Sydow or than Wikidata Q53628

What are we discussing?!?!

1. But what about Linked Open Junk?
   1. I guess if people can create it will be junk. I trust less and less Wikipedia, I never trust WikiTree and always check sources when reading a Swedish profile.... etc.... 
       
2. I would like to have a quality approved list of the best references
   1. Maybe Wikidata can help you problem is when sites like Wikidata starts linking to FindAgrave and WikiTree they will link to sites with no quality process.... ==> you more often get junk compared to quality approved sources   
      
      Solution is what Tim Bernes Lee saýs in the video select the sources you trust and skip the rest... its a loosely coupled system... you dont have to agree with everyone else because that will never work....
      
      Do you think FindAgrave is garbage skip check with them. If WikiTree is always right then just read WikiTree....
       
3. How to find everything related to a person?
   1. Then the semantic web makes it easier
4. Wikipedia vs. Wikidata
   1. In Wikipedia articles you will never have a link to WikiTree they have somekind of quality check/guidelines
      
      but in WikiData you have this person in Wikipedia/Wikidata is the same as WikiTree profile xxx and FindAgrave profile yyy
      1. Example WikiTree Bethell-207 is same as Wikidata Q5111947
         1. Links
             

            Oxford Biography Index Number	101002304
            ISNI	0000 0...9 0914
            Open Library ID	OL6977235A
            VIAF ID	63813847
            GND ID	1053075928
            National Portrait Gallery (London) person ID	mp70771
            WikiTree ID	Bethell-207

 

---

What are we discussing?!?!

I think it's whether, if you want data quality, you should address it at the input stage, or jusr assume it will sort itself out later.

WP, WD and WT now agree about Bethell.  But who knows whether any of it's correct?  I don't, and I created the WT profile.  And for all anybody knows. I might have created the WP article and the WD record as well.

 

---

Data quality is a problem.... and I feel the US "fake news" debate tell us that its much bigger than just genealogy.... In genealogy 2017 with exporting importing unsourced profiles referencing each other its crazy...

I listen a lot to Dario Taraborelli head of the research department at Wikimedia foundation and my understanding is that they focus on sourcing/adding references to all facts. 

Wikidata is an important piece together with an upcoming platform WikiCite

• WikiCite introduction 10 minutes 
  • possibility to reuse citations
  • Wikidata today 77% of all facts are without good sources
• His twitter account