Profile

Cover photo
Antoine Isaac
137 followers|1,463 views
AboutPostsPhotosVideos

Stream

Antoine Isaac

commented on a post on Blogger.
Shared publicly  - 
 

We can create events or statuses that "cause the claim to be true", and create predecessor/successor between these events. I guess that this could be done using an interface for capturing data, which would be as simple as the current one for qualifier, but which would handle the more complex structure behind the scene.

Trying to be clearer on the position I expressed before: I have no doubt that any solution to this problem will look like qualifiers, from a pattern perspective. What I'm really puzzled is the use of the exact same technical solution for presenting (1) qualifiers that describe the claim itself and (2) relations that pertain to what is described in the claim.
The French wikipedia page says that Abdallah II was preceded by Ahmad, in his role of sultan. The wikipedia page does not say that the claim "Abdallah was sultan" was preceded by Ahmad.

But actually looking again at the wikipedia page another solution for answering the challenge comes to mind. It becomes more simpler if one considers the dynasty that is indicated in the table:
https://fr.wikipedia.org/wiki/Dynastie_Alaouite
Just consider that this dynasty is an ordered list, in which Abdallah II appears 6 times. This is what the data should be. you may still indeed have something that look like qualifiers on the Wikidata page, but it would be just a matter of apparences, as it would just be interface sugar (either to enter the data or display it) on top of that ordered dynasty list.
1
Dan Matei's profile photoGerard Meijssen's profile photoAntoine Isaac's profile photo
5 comments
 
Ok, then the Office may be the level that gives the ordered list. By the way I pretty much agree with your "keeping it with the person". It's really at the level of things described, not the claims.

Also, a warning: these challenges are interesting, but one has to keep in mind that the data has to remain usable. And that it may well imply sacrificing some of the richness of the original information.
Keeping everything at the risk of blowing up your knowledge representation and exchange framework does not sound like a good idea. Don't forget that RDF was not created ex-nihilo, even if it may look like it. There are really decades of knowledge representation work before, and the idea of RDF was to avoid the high complexity that plagued many of these efforts, rendering them not usable. And even then many people blame RDF for not being practical enough!
Add a comment...
Have him in circles
137 people
Christophe Guéret's profile photo
Joris Pekel's profile photo
Andrew MacLean's profile photo
benoit rasset's profile photo
Yann Sénécal's profile photo
People
Have him in circles
137 people
Christophe Guéret's profile photo
Joris Pekel's profile photo
Andrew MacLean's profile photo
benoit rasset's profile photo
Yann Sénécal's profile photo
Links
Basic Information
Gender
Male