[Wikidata-bugs] [Maniphest] [Commented On] T95441: Fix representation of somevalue and novalue
gerritbot added a comment. Change 206955 merged by jenkins-bot: https://phabricator.wikimedia.org/T95441: represent novalue in the same way as WDTK https://gerrit.wikimedia.org/r/206955 TASK DETAIL https://phabricator.wikimedia.org/T95441 REPLY HANDLER ACTIONS Reply to comment or attach files, or !close, !claim, !unsubscribe or !assign username. EMAIL PREFERENCES https://phabricator.wikimedia.org/settings/panel/emailpreferences/ To: Smalyshev, gerritbot Cc: Bene, JanZerebecki, gerritbot, Aklapper, Manybubbles, mkroetzsch, daniel, Smalyshev, jkroll, Wikidata-bugs, Jdouglas, aude ___ Wikidata-bugs mailing list Wikidata-bugs@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-bugs
[Wikidata-bugs] [Maniphest] [Commented On] T95441: Fix representation of somevalue and novalue
gerritbot added a comment. Change 206955 had a related patch set uploaded (by Smalyshev): https://phabricator.wikimedia.org/T95441: represent novalue in the same way as WDTK https://gerrit.wikimedia.org/r/206955 TASK DETAIL https://phabricator.wikimedia.org/T95441 REPLY HANDLER ACTIONS Reply to comment or attach files, or !close, !claim, !unsubscribe or !assign username. EMAIL PREFERENCES https://phabricator.wikimedia.org/settings/panel/emailpreferences/ To: Smalyshev, gerritbot Cc: Bene, JanZerebecki, gerritbot, Aklapper, Manybubbles, mkroetzsch, daniel, Smalyshev, jkroll, Wikidata-bugs, Jdouglas, aude ___ Wikidata-bugs mailing list Wikidata-bugs@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-bugs
[Wikidata-bugs] [Maniphest] [Commented On] T95441: Fix representation of somevalue and novalue
Smalyshev added a comment. Another option for novalue - doing what WDTK does: subject [ a http://www.w3.org/2002/07/owl#Class http://www.w3.org/2002/07/owl#complementOf [ a http://www.w3.org/2002/07/owl#Restriction; http://www.w3.org/2002/07/owl#onProperty property; http://www.w3.org/2002/07/owl#someValuesFrom http://www.w3.org/2002/07/owl#Thing ] ] Of course, we can extract the big scary class out and do something like this: wd:Q23 a wdno:P23. wdno:P23 [ a owl:Class owl: complementOf [ a owl: Restriction; owl:onProperty p:P23; owl:someValuesFrom owl:Thing ] ] and reuse `wdno:P23` everywhere for the same property. TASK DETAIL https://phabricator.wikimedia.org/T95441 REPLY HANDLER ACTIONS Reply to comment or attach files, or !close, !claim, !unsubscribe or !assign username. EMAIL PREFERENCES https://phabricator.wikimedia.org/settings/panel/emailpreferences/ To: Smalyshev Cc: Bene, JanZerebecki, gerritbot, Aklapper, Manybubbles, mkroetzsch, daniel, Smalyshev, jkroll, Wikidata-bugs, Jdouglas, aude ___ Wikidata-bugs mailing list Wikidata-bugs@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-bugs
[Wikidata-bugs] [Maniphest] [Commented On] T95441: Fix representation of somevalue and novalue
daniel added a comment. Well, there's rdf:nil, but i'm not sure it's correct here. I thin the choice is: either use the same predicates with a special value, but then you'd no longer find no value entries when looking for entities with no statements about a given property. Or you use special properties, which explodes the number of properties. It seems to be one or the other... can you think of a third alternative? TASK DETAIL https://phabricator.wikimedia.org/T95441 REPLY HANDLER ACTIONS Reply to comment or attach files, or !close, !claim, !unsubscribe or !assign username. EMAIL PREFERENCES https://phabricator.wikimedia.org/settings/panel/emailpreferences/ To: Smalyshev, daniel Cc: Bene, JanZerebecki, gerritbot, Aklapper, Manybubbles, mkroetzsch, daniel, Smalyshev, jkroll, Wikidata-bugs, Jdouglas, aude, GWicke ___ Wikidata-bugs mailing list Wikidata-bugs@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-bugs
[Wikidata-bugs] [Maniphest] [Commented On] T95441: Fix representation of somevalue and novalue
Smalyshev added a comment. Thanks for mentioning rdf:nil, however I think it's not a good match, since it's used for representing lists, and reusing it here may imply semantics that we do not want to bring in. can you think of a third alternative? Well, there's an obvious alternative of just not producing any statements when there's novalue. I'm not saying it's the best one, but it's there. TASK DETAIL https://phabricator.wikimedia.org/T95441 REPLY HANDLER ACTIONS Reply to comment or attach files, or !close, !claim, !unsubscribe or !assign username. EMAIL PREFERENCES https://phabricator.wikimedia.org/settings/panel/emailpreferences/ To: Smalyshev Cc: Bene, JanZerebecki, gerritbot, Aklapper, Manybubbles, mkroetzsch, daniel, Smalyshev, jkroll, Wikidata-bugs, Jdouglas, aude, GWicke ___ Wikidata-bugs mailing list Wikidata-bugs@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-bugs
[Wikidata-bugs] [Maniphest] [Commented On] T95441: Fix representation of somevalue and novalue
daniel added a comment. @Smalyshev perhaps instead of a special value, use a special predicate for novalue/somevalue. Then novalue would behave just like there being no values, if you are looking for values. But if you are interested in novalue specifically, you can still check for it directly. TASK DETAIL https://phabricator.wikimedia.org/T95441 REPLY HANDLER ACTIONS Reply to comment or attach files, or !close, !claim, !unsubscribe or !assign username. EMAIL PREFERENCES https://phabricator.wikimedia.org/settings/panel/emailpreferences/ To: Smalyshev, daniel Cc: Bene, JanZerebecki, gerritbot, Aklapper, Manybubbles, mkroetzsch, daniel, Smalyshev, jkroll, Wikidata-bugs, Jdouglas, aude, GWicke ___ Wikidata-bugs mailing list Wikidata-bugs@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-bugs
[Wikidata-bugs] [Maniphest] [Commented On] T95441: Fix representation of somevalue and novalue
Smalyshev added a comment. The problem with special predicates is that right now we already have 8 types of predicates. Adding novalue ones would mean taking it to 16, with each having its own prefix and URL. For somevalue, another 8. I'm not sure it is a good scheme to have so many different prefixes and URLs, and also may be hard for tools to handle. We could use suffixes in this case, but we concluded suffixes are generally worse idea... I think for somevalue bnode is not a bad idea, but for novalue currently don't have better ideas. TASK DETAIL https://phabricator.wikimedia.org/T95441 REPLY HANDLER ACTIONS Reply to comment or attach files, or !close, !claim, !unsubscribe or !assign username. EMAIL PREFERENCES https://phabricator.wikimedia.org/settings/panel/emailpreferences/ To: Smalyshev Cc: Bene, JanZerebecki, gerritbot, Aklapper, Manybubbles, mkroetzsch, daniel, Smalyshev, jkroll, Wikidata-bugs, Jdouglas, aude, GWicke ___ Wikidata-bugs mailing list Wikidata-bugs@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-bugs
[Wikidata-bugs] [Maniphest] [Commented On] T95441: Fix representation of somevalue and novalue
Bene added a subscriber: Bene. Bene added a comment. Btw, what is the difference between setting no value for property children versus just having no statement for property children? I don't think it's a good idea to add children = no value to all persons which do not have children. The same applies to all other properties which a subject possibly could have but actually hasn't. TASK DETAIL https://phabricator.wikimedia.org/T95441 REPLY HANDLER ACTIONS Reply to comment or attach files, or !close, !claim, !unsubscribe or !assign username. EMAIL PREFERENCES https://phabricator.wikimedia.org/settings/panel/emailpreferences/ To: Smalyshev, Bene Cc: Bene, JanZerebecki, gerritbot, Aklapper, Manybubbles, mkroetzsch, daniel, Smalyshev, jkroll, Wikidata-bugs, Jdouglas, aude, GWicke ___ Wikidata-bugs mailing list Wikidata-bugs@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-bugs
[Wikidata-bugs] [Maniphest] [Commented On] T95441: Fix representation of somevalue and novalue
JanZerebecki added a comment. No statement does not say anything, just that nobody added any information about this yet or that no information can be found. Only with no value does it say that the person does not have children. For many people there are no sources that say whether the person had children, which means no statement would be the most we can do. (All this is offtopic for this ticket, so if you want to further discuss it lets move it to the mailing list, IRC or wikidata.org.) TASK DETAIL https://phabricator.wikimedia.org/T95441 REPLY HANDLER ACTIONS Reply to comment or attach files, or !close, !claim, !unsubscribe or !assign username. EMAIL PREFERENCES https://phabricator.wikimedia.org/settings/panel/emailpreferences/ To: Smalyshev, JanZerebecki Cc: Bene, JanZerebecki, gerritbot, Aklapper, Manybubbles, mkroetzsch, daniel, Smalyshev, jkroll, Wikidata-bugs, Jdouglas, aude, GWicke ___ Wikidata-bugs mailing list Wikidata-bugs@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-bugs
[Wikidata-bugs] [Maniphest] [Commented On] T95441: Fix representation of somevalue and novalue
Smalyshev added a comment. @Bene, we do not add anything, we just represent the data as they are in the wikidata DB. So this is a topic for broader discussion in the community about data representation. Here, the question is when the value in the DB is novalue, how we represent it in RDF. TASK DETAIL https://phabricator.wikimedia.org/T95441 REPLY HANDLER ACTIONS Reply to comment or attach files, or !close, !claim, !unsubscribe or !assign username. EMAIL PREFERENCES https://phabricator.wikimedia.org/settings/panel/emailpreferences/ To: Smalyshev Cc: Bene, JanZerebecki, gerritbot, Aklapper, Manybubbles, mkroetzsch, daniel, Smalyshev, jkroll, Wikidata-bugs, Jdouglas, aude, GWicke ___ Wikidata-bugs mailing list Wikidata-bugs@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-bugs
[Wikidata-bugs] [Maniphest] [Commented On] T95441: Fix representation of somevalue and novalue
daniel added a comment. NOVALUE needs to be represented explicitly. People should be able to query for kinds with no children or states with no capital. In particular, they should also be able to query for states with no statements about the capital property, and *not* get states that are explicitly recorded to have no capital. TASK DETAIL https://phabricator.wikimedia.org/T95441 REPLY HANDLER ACTIONS Reply to comment or attach files, or !close, !claim, !unsubscribe or !assign username. EMAIL PREFERENCES https://phabricator.wikimedia.org/settings/panel/emailpreferences/ To: Smalyshev, daniel Cc: Aklapper, Manybubbles, mkroetzsch, daniel, Smalyshev, jkroll, Wikidata-bugs, Jdouglas, aude, GWicke, JanZerebecki ___ Wikidata-bugs mailing list Wikidata-bugs@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-bugs
[Wikidata-bugs] [Maniphest] [Commented On] T95441: Fix representation of somevalue and novalue
Smalyshev added a comment. Also, I'm not sure what that would mean if we have both novalue and other value on the same property. TASK DETAIL https://phabricator.wikimedia.org/T95441 REPLY HANDLER ACTIONS Reply to comment or attach files, or !close, !claim, !unsubscribe or !assign username. EMAIL PREFERENCES https://phabricator.wikimedia.org/settings/panel/emailpreferences/ To: Smalyshev Cc: Aklapper, Manybubbles, mkroetzsch, daniel, Smalyshev, jkroll, Wikidata-bugs, Jdouglas, aude, GWicke, JanZerebecki ___ Wikidata-bugs mailing list Wikidata-bugs@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-bugs
[Wikidata-bugs] [Maniphest] [Commented On] T95441: Fix representation of somevalue and novalue
Smalyshev added a comment. I agree but I'm not sure how to do it in a way so that the query all people who have children would not return people with no value in children field. Making people use and it's not Novalue each time doesn't sound good as people would have no idea they have to do this. TASK DETAIL https://phabricator.wikimedia.org/T95441 REPLY HANDLER ACTIONS Reply to comment or attach files, or !close, !claim, !unsubscribe or !assign username. EMAIL PREFERENCES https://phabricator.wikimedia.org/settings/panel/emailpreferences/ To: Smalyshev Cc: Aklapper, Manybubbles, mkroetzsch, daniel, Smalyshev, jkroll, Wikidata-bugs, Jdouglas, aude, GWicke, JanZerebecki ___ Wikidata-bugs mailing list Wikidata-bugs@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-bugs
[Wikidata-bugs] [Maniphest] [Commented On] T95441: Fix representation of somevalue and novalue
daniel added a comment. In https://phabricator.wikimedia.org/T95441#1191057, @Smalyshev wrote: Also, I'm not sure what that would mean if we have both novalue and other value on the same property. It would mean that there is a disagreement, e.g. some historians claim that king Foobar had no children, others claim that Xyzzy was Foobar's child. Wikibase is explicitly designed to allow this kind of disagreement. This would be expressed by having two statements about the property, citing different sources. TASK DETAIL https://phabricator.wikimedia.org/T95441 REPLY HANDLER ACTIONS Reply to comment or attach files, or !close, !claim, !unsubscribe or !assign username. EMAIL PREFERENCES https://phabricator.wikimedia.org/settings/panel/emailpreferences/ To: Smalyshev, daniel Cc: Aklapper, Manybubbles, mkroetzsch, daniel, Smalyshev, jkroll, Wikidata-bugs, Jdouglas, aude, GWicke, JanZerebecki ___ Wikidata-bugs mailing list Wikidata-bugs@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-bugs
[Wikidata-bugs] [Maniphest] [Commented On] T95441: Fix representation of somevalue and novalue
daniel added a comment. Btw: Such disagreements should not be a problem in truthy output either, because we don't define any semantics for the properties used as predicates, on the level of RDFS/OWL. So even if a human sees a contradiction there, a triple store shouldn't. TASK DETAIL https://phabricator.wikimedia.org/T95441 REPLY HANDLER ACTIONS Reply to comment or attach files, or !close, !claim, !unsubscribe or !assign username. EMAIL PREFERENCES https://phabricator.wikimedia.org/settings/panel/emailpreferences/ To: Smalyshev, daniel Cc: Aklapper, Manybubbles, mkroetzsch, daniel, Smalyshev, jkroll, Wikidata-bugs, Jdouglas, aude, GWicke, JanZerebecki ___ Wikidata-bugs mailing list Wikidata-bugs@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-bugs
[Wikidata-bugs] [Maniphest] [Commented On] T95441: Fix representation of somevalue and novalue
gerritbot added a subscriber: gerritbot. gerritbot added a comment. Change 202963 had a related patch set uploaded (by Smalyshev): https://phabricator.wikimedia.org/T95441: Make somevalue use bnode https://gerrit.wikimedia.org/r/202963 TASK DETAIL https://phabricator.wikimedia.org/T95441 REPLY HANDLER ACTIONS Reply to comment or attach files, or !close, !claim, !unsubscribe or !assign username. EMAIL PREFERENCES https://phabricator.wikimedia.org/settings/panel/emailpreferences/ To: Smalyshev, gerritbot Cc: gerritbot, Aklapper, Manybubbles, mkroetzsch, daniel, Smalyshev, jkroll, Wikidata-bugs, Jdouglas, aude, GWicke, JanZerebecki ___ Wikidata-bugs mailing list Wikidata-bugs@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/wikidata-bugs
