User talk:Oravrattas

Previous discussion was archived at User talk:Oravrattas/Archive 2 on 2016-04-02.

Hello,

I'm pretty new to Wikidata, but thought I'd try to update Wikidata:WikiProject every politician/United States of America to have entries for every congress (i.e. term) of the U.S. Congress. Separately I've collected data on all historical U.S. Senators and plan on uploading that data shortly.

I'm a little confused as to why my recent change to pass the terms parameter to the Template:Every Politician/Legislature table row template isn't working. Being new, I'm not sure how to debug the issue. I only added Wikidata:WikiProject every politician/United States of America/Congresses moments ago. Do I need to wait for some bot to run for these terms to show up on Wikidata:WikiProject every politician/United States of America the way they do on Wikidata:WikiProject every politician/Scotland?

Any tips you could offer would be appreciated.

Thanks

It looks like it's worked to me. Perhaps you had a cached version?

Possibly, but I'm still not seeing anything more than the Current Members link under Member lists, where I was expecting to see a bullet item for each Congress. I tried refreshing the browser cache. Is there something else I should try?

Do I need to manually create a Listeria list for each Congress like this one? Wikidata:WikiProject every politician/Scotland/data/Parliament/1st Parliament

Ah, adding a 'terms' parameter doesn't change anything there: it changes the 'all terms' link in the first column.

Once relevant pages are created under United States of America/data/Senate/ they'll appear in the second column.

FWIW, It's pretty clearly harassment to me. I have a pretty thick skin, so I thought I'd give him every opportunity to drop the bullying. It also helps to lay bare what he's doing. It's also clearly a conduct issue because he avoids all discussion of the content.

Oddly, he seems to concede that Teester's model is acceptable here. [https://www.wikidata.org/w/index.php?title=Wikidata:Requests_for_deletions&diff=1257610452&oldid=1257609418]

Also, I explained my plan here [https://www.wikidata.org/w/index.php?title=Wikidata:Project_chat&diff=1258054898&oldid=1258050860], but he keeps asking me to link old chats. It's clearly meant as busy work to scare me off wanting to contribute.

Moreover, as I mentioned here [https://www.wikidata.org/w/index.php?title=User_talk:Gettinwikiwidit&diff=1257471105&oldid=1257466606] this obviates his complaints about Teester's model, namely that the addition disrupted bots. Separating the property used for the two types of position entries means both can co-exist. I'm not sure he really cared that much the first time around though since he has done nothing to resolve it in over a year (two?)

Thanks for your help.

FWIW, on MisterSynergy's talk page Jura1 tried to have me completely blocked and reiterated his complaints about your funding. I for one don't care too terribly much about what people's motivations to contribute are. I would just like the data to be useful.

Hello Vunj9wqvucy74tfx,

Really sorry for the inconvenience. This is a gentle note to request that you check your email. We sent you a message titled "The Community Insights survey is coming!". If you have questions, email surveys@wikimedia.org.

You can see my explanation here.

Hi, many articles you edit with QuickStatements (Lluís Puig i Gordi (Q31837305), Raquel Casviner Cañellas (Q63351546), Meritxell Lluís i Vall (Q59343390), Raül Romeva (Q2065232), ...), were already correct

I can't quite understand the modelling here. What is the connection between Q61744680 and Q63825349?

Q63825349 is the specific data from Q61744680 in the electoral district of Barcelona

In which of these did a given candidate stand?

Thanks for the hint! I've been looking for such a page since I began working with Wikidata over two months ago. It confirms my suspicion that there is still much work to be done on the meta-technical side; I'm not at all satisfied with how novice editors are left to decide for themselves whether an item should be an instance of (P31) or a subclass of (P279) another without the tools necessary to visualize the ontological structure of the domain they are working in. I think I understand the concepts well enough to decide how to do my own edits, but I often find myself at odds with Wikipedia editorial decisions possibly made years ago, perhaps with little concern for Wikidata requirements.

As an example, please have a look at office (Q12823105) and explain to me what that particular word "or" is doing in a Wikidata item description? With your experience, you should be able to tell what this looks like from a non-English-language perspective. Like, what if attic (Q244490) were described as "space found directly below the pitched roof of a house or other building or flow of gases or air on a large scale"; would that make sense? Shouldn't this sort of thing have been resolved already at the disambiguation page level? It has even crept into the lexeme entry. And a curriculum (Q207137) is not a Latin phrase, but an educational plan. And...

Sorry for ranting on your talk page; my bad. It's no fault of yours. And I do appreciate your work on political data, whether you were involved with getting all the members of the Swedish parliament since 1970 into Wikidata or not!

I've removed the second half of that 'or', as it's clearly a mistake. position (Q4164871) is the item for that other concept.

I'll leave the Lexeme for someone else to resolve, as I'm not so familiar with the nuances around those.

Ok, thank you! A word of caution though: My reason for not fixing all those errors as soon as I encounter them is that I want to understand how they came to be in the first place, and what other errors they may have resulted in, in order to fix those as well. I guess the Office case isn't much to worry about; the description is apparently taken from the Office disambiguation page, which itemizes other interpretations of the word, but doesn't suggest that "position" might be yet another item (and there is no English Wikipedia article linked to the position (Q4164871) item anyway). I was looking for some kind of "public office" though, but I threw the whole issue aside when I found that many of the national public offices have their own unique items, sometimes subclasses of position, and I don't recommend merging them without a very thorough understanding of all the aspects described in the different linked articles.

Case in point: I'm now considering undoing an earlier merge of the Swedish Resecentrum (public transit centre) article item with transport hub (Q2298537) (and probably also of the Finnish equivalent) as I don't think the merging editor realized that a transport hub may include also cargo handling. There are oil and container shipping terminals among its subclasses! But the ancestral class tree doesn't seem to be very carefully laid out either (I have spent almost a full week investigating the 238 subclasses of transportation stop (Q548662) since I learned about the new public transit stop property and wanted to know what it referred to before suggesting some appropriate qualifiers to it)...

But I very much enjoy this. Wikidata is quite close to what I imagined an Internet web of information might look like, before the World-Wide Web appeared.

Hi. I don't understand your recent edits at Q63928321 and Q63890553 where you changed the value of P31 from Q12311825 to Q19571328. The two items are linked to Wikipedia articles at dawiki which is about the Folketing elections 2019 in Faroese Islands respective Greenland. The articles aren't only about the results but also describes election rules, names the candidates etc. So why did you make the edits https://www.wikidata.org/w/index.php?title=Q63928321&curid=63633752&diff=1225897894&oldid=1225897511 and https://www.wikidata.org/w/index.php?title=Q63890553&curid=63600763&diff=1225898010&oldid=1225897401?

PS. I don't understand the statements with P1269 either. I think that the statements with P361 is more accurate. The elections on Faroese Islands and Greenland was a part of the Folketing elections. Nothing more, nothing less.

My starting point here is that these shouldn't show up in an election calendar: they're not the actual elections themselves, but represent pages that talk about an aspect of the election within a particular area. I've no particular attachment to using electoral result (Q19571328) if they're wider than that, but I don't think they should be an instance of election (Q40231) (or any subclass of that)

I don't agree that they aren't instances of elections. The Folketing is elected in three separate elections in Denmark proper (175 seats), Faroese Islands (2 seats) and Greenland (2 seats). Each of these elections are organized after different rules according to their own election laws. They don't even necessarily happen on the same date. Why shouldn't they show up in an election calendar?

BTW. The situation is exactly like it is for 2019 European Parliament election (Q16999180) which has parts that happens in different countries like e.g. 2019 European Parliament election in Denmark (Q55436512). Both of these items (Q16999180 and Q55436512) have instance of (P31) election to the European Parliament (Q1128324). I see no difference between this and the Danish Folketing election having three parts in different countries/autonomous territories.

How would you suggest generating a list like that at https://en.wikipedia.org/wiki/2019_national_electoral_calendar (not including referendums) from Wikidata?

I will look into it and reply with a query later.

There doesn't seem to be any meaning in the statement you added at . Given that this may be a problem that occured before, please propose any additions to items with Swiss parliament ID (P1307) at Property_talk:P1307 before doing them. This ensures that Wikidata quality remains consistent.

This was an upload from https://fr.wikipedia.org/wiki/Liste_des_membres_du_Conseil_national_suisse_(2019-2023), where some of the entries in the 'group' column go to that item, rather than one for the group itself. Those would have been corrected in a follow-on set of updates, but given your resistance, I'll simply back those all out and leave this data alone.

Thanks, I will clean up once more.

Hi! Thanks for the help on migrating qualifiers. Would you mind sharing the code used for this batch? https://editgroups.toolforge.org/b/wikibase-cli/9194d5a15ae2b/ I am trying to learn the WD tooling and examples are very helpful. ~~~~

Sure. I have a P1268-to-P4100.rq file containing the query:

SELECT ?st WHERE { ?st ps:P39 wd:Q10655178 ; pq:P1268 [] }

Try it!

This gets the statement IDs that we need to migrate. Then those can be switched using wikibase-cli:

wd sparql P1268-to-P4100.rq | xargs -n 1 | awk '{ print $1, "P1268 P4100" }' | wd mq --batch --summary "Migrate P1268 to P4100 for members of the Riksdag"

That's neat! Thanks!

Hi! Why you removed independent (Q2415493) from MP items? I know that in English independent (Q2415493) and independent politician (Q327591) have the same label, but independent (Q2415493) is specific for Parliaments (they do not belong to any of the parliamentary groups, different from not being affiliated with any political party). It is called in that way by Parliament of Catalonia and the Europea Parliament, for instance.

Do you have some examples of where I did this? Mostly I was doing this the other way around: migrating P39 records from the general "not part of a political party" to the specific "not part of a parliamentary group", so I suspect any the other way around may have been a mistake, but examples would help me look more closely. Thanks!

Q11924122, Q20006587, Q4889041, ...

The definition of those have changed quite a bit even in the time since those changes. I do agreed that we'd benefit from being precise in the difference between someone standing for election as an independent candidate, and someone sitting in a legislature as part of one. In some countries there is very little difference between those, but in many the difference is significant.

BTW, with respect to Members of the European Parliament, I suspect we should be using Non-Inscrits (Q1440626) rather than either of the ones mentioned.

It is possible to revert editions that changed non-attached (Q2415493) for independent politician (Q327591) in Member of the Parliament of Catalonia (Q18714088)?

Instead of removing instance of (P31): election (Q40231), consider add subclass of (P279) <type of election, like gubernatorial, presidential etc>. Thank you

That's exactly what I have been doing… Do you have an example of one where I've removed a plain election (Q40231) and not added something more suitable? That would imply something is going wrong somewhere, so it would be good to track it down. Thanks.

Acre gubernatorial election, 2014 (Q17813784) and Acre gubernatorial election for Senate, 2014 (Q51330919) are the ones that caught my eyes (I'm modeling the Brazilian elections). Both of them you removed "election" and put "gubernatorial election" as instance, but the first is about the election event itself, which is divided into 2 rounds, which in turn are umbrelas for the other elections (each position is an election itself). I wouldn't remove instance of election, since this is what they are. In fact, office contested (P541) and applies to jurisdiction (P1001) fullfills perfectly the distinction you are trying to declare. Please take into consideration that not an electoral process is equal to another (that's why I limited myself to the Brazilian one) and let me know if you feel the need of some insight, I'd be glad to help. Good contributions

The problem here seems to be that these items are about quite different things depending on what language you're working in! In English Acre gubernatorial election, 2014 (Q17813784) is only about the election for Governor, but in Portuguese it's about all the elections at that time. This seems to have led to some slightly odd descriptions then, where Acre gubernatorial election for Senate, 2014 (Q51330919) in English is the "Gubernatorial election for the Senate", which doesn't make any sense. So based on the English description and link, the correct P31 for Q17813784 would indeed be gubernatorial election (Q15261477), but then that's incorrect for the Portuguese version. We're going to need to split that item, and move either the English or Portuguese Wikipedia link to a different one (I'm not sure whether something suitable already exists, or would need to be created).

I read the article in enwiki and it is about the election of the governor only. I moved it to State elections in Acre in 2014 for Governor (Q51336711), although I'm reluctant on the articles being orphan. But, besides the Wikipedia interlink, I would suggest to you base these choices of instances etc in the items statements themselves, not the label and/or descriptions, once those can be subjective and misled by wrong assumptions or translations. I didn't know "gubernatorial election" would only apply to the governor, I thought it was the "english" way to represent "state elections". Thank you for your insights!

Do you have a source for this - Special:Diff/793566387? And this - Special:Diff/793566289.

This was a bulk import of people in https://ro.wikipedia.org/wiki/Categorie:Deputați_români

It looks like some of the paths there aren't to actual members — Ecaterina Iorga (Q53998086) has come in via Nicholas lorga having his own sub-category, which is easily fixed, but the Members of the Transylvanian Diet, Hungarian Diet, and Grand National Assembly are a bit more problematic. I'll fix those up now.

Thank you. Looks better now.