NO TITLE edit

Tjena!

Vet inte om du sett Wikidata:Cultural heritage task force. -- Lavallentalk(block) 15:30, 7 August 2013 (UTC)Reply

Jodå. Har bara inte hunnit koma igång med den =). Behövs nått mer med svenska geografiska properties?/André Costa (WMSE) (talk) 17:18, 7 August 2013 (UTC)Reply
Har inte studerat I detalj, men identifierare för kulturarvsdata hade ju varit trevligt, men är kanske fel att introducera just nu när det är så mycke prat om att sammanföra flera identifierar-properties till ett enda eller några få, så man kanske ska ligga lite lågt med det. RAÄ-nummer kanske går att använda någon befintlig eller redan föreslagen property för, annars kanske den är lämplig att föreslå. Ngn property som borde kunna användas för återgivning (citat) av runstenar tror jag det redan borde finnas. sockenkod finns redan om du tittar på min disk. -- Lavallentalk(block) 19:29, 7 August 2013 (UTC)Reply
Jag har lagt till constraints på vissa av dessa samt putsade på Pastoratkodsbeskrivningen. Ta gärna en kik och se att det ser rätt ut. Vet inte var/om källorna ska in på diskussionssidorna.
För kulturarvsdata finns det på rak hand följande authorities som används på sv.wiki:
  • BBRa: Bebyggelseregistrets anläggningsid
  • BBRb: Bebyggelseregistrets byggnadsid
  • FMIS-id: Fornminnesregistrets objekt-id
  • RAÄ-nr: FMIS traditionella id (kan egentligen fås fram automatiskt från FMIS-id
  • Signum: Samnordisk runtextdatabas identifierare för runristningar. Detta är även den vanligaste källan för runtexttolkningar.
/André Costa (WMSE) (talk) 03:29, 8 August 2013 (UTC)Reply
Samnordisk databas, det kommer de gilla! Är det Uppland/Östergötland 1,2,3 etc? -- Lavallentalk(block) 06:25, 8 August 2013 (UTC)Reply
Precis. Den är främst Sverigefokuserad men inkluderar även några andra länder. Jag skulle dock rekommendera att enbart ta med de två första delarna av signumet (dvs. den delen som man vanligen använder typ "U 293"). Jag länkade ovan till artikeln. /Lokal Profil (talk) 09:37, 8 August 2013 (UTC)Reply

Welcome edit

 

Welcome to Wikidata, André Costa (WMSE)!

Wikidata is a free knowledge base that you can edit! It can be read and edited by humans and machines alike and you can go to any item page now and add to this ever-growing database!

Need some help getting started? Here are some pages you can familiarize yourself with:

  • Introduction – An introduction to the project.
  • Wikidata tours – Interactive tutorials to show you how Wikidata works.
  • Community portal – The portal for community members.
  • User options – including the 'Babel' extension, to set your language preferences.
  • Contents – The main help page for editing and using the site.
  • Project chat – Discussions about the project.
  • Tools – A collection of user-developed tools to allow for easier completion of some tasks.

Please remember to sign your messages on talk pages by typing four tildes (~~~~); this will automatically insert your username and the date.

If you have any questions, don't hesitate to ask on Project chat. If you want to try out editing, you can use the sandbox to try. Once again, welcome, and I hope you quickly feel comfortable here, and become an active editor for Wikidata.

Best regards! --Tobias1984 (talk) 15:42, 7 August 2013 (UTC)Reply

Babel edit

Use Babel on Wikidata to view and edit more than one language e.g. {{#babel:de-N|en-3}} --Tobias1984 (talk) 15:42, 7 August 2013 (UTC)Reply

Many thanks. This made my life way easier =) /André Costa (WMSE) (talk) 03:29, 8 August 2013 (UTC)Reply

P107 edit

P107 är på G att bli raderad. Det är instance of (P31) som gäller. Exakt hur den ska användas för biografier vet jag inte. Just "Person" är nog inte första valet, då det tydligen kan åsyfta flera saker. -- Lavallen (talk) 08:11, 31 December 2013 (UTC)Reply

Tack för tipset. Verkar vara instance of (P31) -> human (Q5) som gäller nu. Jag ville mest städa upp efter ÖDOKBot som skapar wikidata entries för svenska artiklar utan sådana när den stöter på dem i sitt vanliga arbete. /André Costa (WMSE) (talk) 11:27, 15 January 2014 (UTC)Reply

WD:PC edit

Moved to User talk:Lokal_Profil /André Costa (WMSE) (talk) 14:08, 3 March 2014 (UTC)Reply

Your edit to Hals portrait edit

 
In Nationalmuseum or not?
 
the correct Nationalmuseum NM 6421

Hi I noticed this edit and was wondering why. Is this painting not in the collection because it is on loan, or has it been sold? According to the RKD documentation it is in the museum catalog with code NM6421. Thanks, Jane023 (talk) 17:23, 20 March 2015 (UTC)Reply

I went to the rkd entry linked to from the wikidata entry (not the one you linked to) where it just says private collection. NM6421 is this one (which also doesn't fit the description at rkd). /Lokal Profil (talk) 18:27, 20 March 2015 (UTC) (on my private account)Reply
Sorry for the delay - In Slive she is listed as being in the collection of the King of Sweden, is that collection online? Here is the other painting. Jane023 (talk) 17:50, 17 August 2015 (UTC)Reply
I haven't heard of that collection before so I'm not sure whether it is online, seems unlikely though. /André Costa (WMSE) (talk) 08:27, 25 August 2015 (UTC)Reply

ÖdokBot edit

To run a bot here, please file a request at Wikidata:Requests for permissions/Bot.--GZWDer (talk) 17:35, 20 July 2015 (UTC)Reply

Added. /André Costa (WMSE) (talk) 13:50, 23 July 2015 (UTC)Reply

Alternate accounts edit

See User talk:Lokal Profil#Alternate accounts.--GZWDer (talk) 02:32, 28 July 2015 (UTC)Reply

Data upload for the National Museum, Sweden edit

Hi André,

A small issue with some of the data you uploaded last year for collections and subcollections of the Nationalmuseum.

It looks like you've been using the applies to part (P518) qualifier to indicate the subcollection, whereas normally this qualifier would indicate a part of the overall work.

A couple of standard approaches would be either to have the collection (P195) property point to the sub-collection, and then on the sub-collection item note that it was part of (P361) the overall collection; or, alternatively, to have the work have collection (P195) = the overall collection, and then use location (P276) to indicate the museum the work is actually assigned to.

Any chance you could take a look at these, and consider what the best way forward would be ?

Also raised at Wikidata_talk:WikiProject_sum_of_all_paintings#Collection_.2F_applies_to_part.

Thanks! Jheald (talk) 16:18, 28 October 2015 (UTC)Reply

References edit

Hi André, I noticed your edits on Q20857660. Can you please include an url in each reference and the date retrieved? Multichill (talk) 13:06, 24 February 2016 (UTC)Reply

They should all be referenced. The only exception is if the bot crashes between adding a statement and sourcing it... in which case it gets handled on the next pass through.
@Multichill: Or do you mean a reference URL (P854) should be added in addition to the existing reference statements? (and publication date (P577) changed to retrieved (P813)) André Costa (WMSE) (talk) 13:27, 24 February 2016 (UTC)Reply
Actually changing and publication date (P577) changed to retrieved (P813) wouldn't make sense since I rely on the timestamp in the document rather than the current timestamp (since that says more about how old the info is). /André Costa (WMSE) (talk) 13:29, 24 February 2016 (UTC)Reply
Sorry, I wasn't clear. You're currently adding stated in (P248) and publication date (P577). Could you also add (not replace) reference URL (P854) and retrieved (P813) so that the references have 4 parts? Multichill (talk) 13:35, 24 February 2016 (UTC)Reply
Ah. Now I get it. Sure I'll add those parts. I can't easily do it retroactively though, but I'll look into it... once wdq is back up. /André Costa (WMSE) (talk) 15:11, 24 February 2016 (UTC)Reply
@Multichill:: First one with all four now up. /André Costa (WMSE) (talk) 15:54, 24 February 2016 (UTC)Reply
Excellent! That way it's easy to follow where the data came from without having to puzzle. Multichill (talk) 16:23, 24 February 2016 (UTC)Reply
Things seem to be moving quite fast now! I noticed that User:Multichill/Nationalmuseum painters missing id is much smaller now after your fix. Do you plan to create the missing painters? Maybe with this list or a bot? Multichill (talk) 16:57, 24 February 2016 (UTC)Reply
There are still quite a lot of more connections which can be done through the mix'n'match dataset (even just verifying the auto matched statements should give quite a lot more connections). But once that has slowed down then your tool would be an excellent start for identifying which artists to create. I don't think the whole of the mix'n'match dataset is suitable for creation though since it also includes publishers and a few other categories which are less likely to be relevant. /André Costa (WMSE) (talk) 10:11, 26 February 2016 (UTC)Reply
In case you didn't realize yet: My tool uses the mix'n'match database. Not sure if I mentioned that before. I agree that it probably doesn't make sense to create everyone in the set. Multichill (talk) 11:32, 26 February 2016 (UTC)Reply
I created a bunch of the missing artists. Only a few left to complete it. I might have created some duplicates, but once your bot adds more information, these should surface for merger. Multichill (talk) 15:33, 28 February 2016 (UTC)Reply
Created the missing ones and connected all remaining painters. Can you run your bot to add more information to these artists and get it sourced? Multichill (talk) 12:34, 5 March 2016 (UTC)Reply
Thanks. I've imported them into mix'n'match so the bot should do a pass during the day. /André Costa (WMSE) (talk) 12:47, 5 March 2016 (UTC)Reply

National Museum in Mix'n'match edit

Hi, I see that you marked entries like this one and this one as "not relevant for Wikidata". I imagine that is because they are duplicates of [1] and [2] ? If they are unintentional duplicates in the museum's database, I think both ids should be added to Wikidata, so as to get the most comprehensive mapping possible, and also, if they happen to be watching us, to help them improve fix their database :). - Zolo (talk) 21:33, 24 February 2016 (UTC)Reply

@Zolo: Hi. Yes it's correct that I marked them due to the duplicates. I also added a suggested edit in their exposed dataset (over at here). I'm in contact with the person responsible for digital coordinator to see how they will act upon such discoveries. If they ignore the suggestion or keep the entries anyway then I'd agree that both should be added to Wikidata. If they merge their results (and one of the public link goes dead) then I'm less sure.
If we go for double id's I would have to patch the code for User:AndreCostaWMSE-bot, which is enriching the items based on those connections, since it relies on WDQ and WDQ doesn't pass on information about the rank of the statement.
I'll ping in an update here once I hear back from the museum. /André Costa (WMSE) (talk) 10:04, 26 February 2016 (UTC)Reply
I agree with Zolo here. If they have accidental duplicates, we should have them both here. If they solve it, we'll just remove the identifier that's no longer in use. Or you can go even one step further and do it the same as the Netherlands Institute for Art History (Q758610). They keep an eye on the constraint report and when duplicates pop up, they update their database and also Wikidata. I'm quite happy with how that is going!
Something else while talking about data quality. The date of death for Hendrik van Steenwijk I (Q709038) on 7209 is mixed up with Hendrik van Steenwijk II (Q709393). It should be 1603, not 1649 (ulan & RKD). Multichill (talk) 12:16, 26 February 2016 (UTC)Reply

AndreCostaWMSE-bot adding nationality edit

This edit is problematic, because the person in question was never a citizen of Italy - a country which didn't exist at that time. It's better to avoid automatization in this case; references (and Wikipedia categorization) usually refers to ethnicity when they say he or she is an Italian, while the property country of citizenship (P27) refers to a different meaning of the word "nationality". — Yerpo Eh? 08:53, 26 February 2016 (UTC)Reply

@Yerpo: Hi. The problem here is that an authoritative source (the database of an art museum), incorrectly, states that the nationality is Italian. I believe it would be better to keep the statement but set it to deprecated (like this). That way it is recognized that some (credible) sources claim this but that it has been identified to be incorrect. /André Costa (WMSE) (talk) 09:52, 26 February 2016 (UTC)Reply
Actually we have a massive unsolved issue with those "nationalities" in Wikidata. It is quite common for museums to refer to 16th centuy people as Italians, and it was common at the time as well, and even I think, of the "Italian nation". Obviously, that is no nationality in the current technical sense, but things become messy pretty fast when moving away from the contemporary context of nation-states with a clear citizen/non-citizen distinction. There were discussion about it in Wikidata, but afaik no consensus decision on how to handle it. --Zolo (talk) 08:07, 27 February 2016 (UTC)Reply
@André Costa (WMSE): thanks, I agree that makes sense, but there's a lot of wrong info around like this and with infoboxes auto-transcluding data from Wikidata, it can be frustrating to users that like things to be precise (that kind of users we need) but are less computer-savvy. Even I didn't know about the "deprecated" setting. Maybe a bot could help - setting as deprecated every nationality where the country didn't exist at the time when the subject was alive.
@@Zolo:, I believe we do refer to historic inhabitants of the Apennine peninsula as Italians, they just lived in different states. Wikidata treats "nationality" as equivalent of "citizenship", but there is another meaning: "ethnicity". — Yerpo Eh? 15:22, 27 February 2016 (UTC)Reply
Please refrain from doing any (mass) changes until you probably figured out how to handle this and have community consensus. Multichill (talk) 15:50, 27 February 2016 (UTC)Reply
@Zolo, Yerpo, Multichill: I've removed nationality from the changes done by the bot. If anything this shows that an RFC is needed to figure out what we actually intend country of citizenship (P27) to signify. Should we use it at all for anyone born before say 1900? The current (English) description "country of citizenship" essentially means the state which issued your passport. What most people actually interpret it as is e.g. "this person is French". Whether France here means the the First French Republic, the Kingdom of France or modern day France is probably not something they think about. It also raises all kinds of issues for some people with inexact birth dates where their actual citizenship can never be known. I believe there are a lot of similar issues with how country (P17) is being used so that should probably be part of the same RFC. André Costa (WMSE) (talk) 16:29, 28 February 2016 (UTC)Reply

EU Surface Water Body Code (P2856) edit

Hi, I tried to convert some of the Lake ID (Sweden) (P761) values to EU Surface Water Body Code (P2856). I noticed then that many of the links provided by Lake ID (Sweden) (P761) are not valid, for instance try the links at Q18335126, Q18332676, Q18335193, Q18335192, Q18335194, Q18177679, Bjäresjö (Q16502448), Q18177719 and Q18177675 (9 out of 10 items of my trial...). Can you look into that? Are the svwiki values reliable? Lymantria (talk) 14:44, 2 June 2016 (UTC)Reply

That looks bad, and both the P761 and P2856 are broken so it's likely an issue at the source. Maybe @Nasko: can shed some light on it?/André Costa (WMSE) (talk) 15:22, 2 June 2016 (UTC)Reply
Indeed, that's what I was afraid of - perhaps codes are withdrawn?? Lymantria (talk) 17:16, 2 June 2016 (UTC)Reply
EU Surface Water Body Code (P2856) is regulated by the EU in Water Framework Directive (Q1508115). In short: lakes (or rivers and groundwaters reservoir for that matter) that are so large that they aren't affected by the surrounding basin and smaller streams are regarded as "sv:vattenförekomst" and should have EU Surface Water Body Code (P2856). As a rule of thumb: lakes >=1 km2 are regarded as EU Surface Water Body Code (P2856) which is around 7500 lakes in Sweden. Lakes <1 km2 are bundled together with other lakes and streams in the subbasin (sv:delavrinningsområde) of about 5-20 km2. It is incorrect to give all lakes a EU Surface Water Body Code (P2856)
Some facts:
  1. All lakes in Sweden have an id numer that identifies it. It's called sjöid (lake id) The id is the same as the lakes coordinates (the out flow) in the now obsolete coordinate system RT90 with 10 m precision. The id for Börringesjön is 615464-134175. The lake is situated 6154.640 km north of the equator...
  2. All lakes with descent map data (shows up on a map with the scale 1:250000) has the id "Vattenytans ID (VYID)" (water surface id). VYID is the same as the lakes coordinates (the approximate center point) in the now obsolete coordinate system RT90 with 10 m precision. Börringesjön [3] has the VYID 615343-134249. It is located 1210 m south of the out flow.
  3. Lakes >= 1 km2 have the EU Surface Water Body Code (P2856). It is identical to sjöid (lake id) with the prefix SE. Börringesjön has sjöid 615464-134175 and EU Surface Water Body Code (P2856) SE615464-134175.
  4. Lakes < 1 km2 use "Vattenytans ID (VYID)" (water surface id). It is identical to "Vattenytans ID (VYID)" (water surface id) with the prefix NW. Åmossarna has sjöid 614773-133228 and "Vattenytans ID (VYID)" (water surface id) 614785-133228 and with the added prefix it becomes NW614785-133228. Åmossarna [4] dosn't have any EU Surface Water Body Code (P2856). The lake is not large enough.
  5. I have never heard of VISS ID. I think it has been mixed up with Sjöid. Sjöid lake id is on the form ######-######. EU Surface Water Body Code (P2856) is on the form SE######-######. P761 should be called sjöid. VISS is only a database with external data, just like Wikipedia and Wikidata.
It is not surprising that only 1 out of 10 worked. Only 7500 (13%) of the 57000 Swedish lakes on svwp are 1 km2 or larger. 30000 or so (all lakes larger than about 0.04 km2) are in modern databases in the EU (=VISS). All lakes >= 0.01 km2 (100000 or so) exist in older databases maintained by SMHI.
Info in swedish can be found here and here.
I hope I could shed some light. Nasko () 19:10, 2 June 2016 (UTC)Reply
Thank you very much, Nasko. I think this should mean that we keep Lake ID (Sweden) (P761) and rename into sjöid, while removing its formatter URL, while we have to do additional investigations on which of the codes can be added to EU Surface Water Body Code (P2856). Lymantria (talk) 05:37, 3 June 2016 (UTC)Reply
Thanks Nasko and Lymantria. I talked with my contact at VISS and as he explained it Lake ID (Sweden) (P761) are the sjöid as mentioned. A subset of these were then reused in VISS for their surface water bodies. A subset of that subset was exported to WISE and is hence reachable through EU Surface Water Body Code (P2856) (with SE prefix). The renaming of Lake ID (Sweden) (P761) is indeed correct /André Costa (WMSE) (talk) 14:44, 14 June 2016 (UTC)Reply

WLM DB edit

Hi, so about WLM and Wiki Loves Monuments ID (P2186):

  • Are these identifiers stable?
  • For which countries are they available?
  • Is there a database on Labs I can access read-only?

Having done WLM-UK two (?) years ago, which included the creation of respective Wikidata items, I might have a go at matching them. --Magnus Manske (talk) 16:14, 13 June 2016 (UTC)Reply

@Magnus Manske: this is a made up id. I would touch it with a ten foot pole. Take a look at Wikidata:WikiProject Cultural heritage for some of the correct properties to use. On Commons:Commons:Monuments database you can find all the info about the database. It's best to use the api, but you can access the SQL database directly too. Multichill (talk) 21:03, 13 June 2016 (UTC)Reply
Thanks! --Magnus Manske (talk) 08:14, 14 June 2016 (UTC)Reply
@Magnus Manske, Multichill: It's an id that, AFAIK, should only ever be used for items that don't have a proper id. I.e. where artificial ids were created for the sake of WLM. Italy had to do this and Sweden had to do it for a handful of objects. Don't know how widespread it is. /André Costa (WMSE) (talk) 14:39, 14 June 2016 (UTC)Reply

WLM lists edit

Hi! I'd like to discuss some topics about such lists and help with it. I am working on Latvian WLM lists (IMHO it is the only lists not migrated to WLM database :-). I wrote pywikibot code which load Latvian WLM lists to WD and created example list for an region (by Listeriabot with @Jura1: and @Magnus Manske: help). Surely I see global problems - it is too hard to edit the items and create new items in these lists now, so WLM community will prefer to work with original lists, as well as local issues - for example I can't group monuments in the same way as it was done in original list. Yet another global issue - we can't migrate local (non Commons) image to Wikidata. So where is the best place for discussions about it? --Voll (talk) 20:03, 28 June 2016 (UTC)Reply

Hi Voll. Thanks for reaching out and appologies for the delayed reply. Does all of the monuments exist both on Wikidata and in Wikipedia lists (such as w:lv:Valsts aizsargājamie kultūras pieminekļi Alojas novadā)? If so that is great! If it is not yet in the Monuments database I can fix that for you.
As for the Listeriabot lists yes I agree that they are not great for the WLM community (we wrote about that here). The non-Commons images are an issue of course. If htey can be migrated to Commons they definitely should. If they are used as fair-use (or not suitable for Commons for another reason). Then there is nothing we can do.
As for the best place to discuss these things. Probably Wikidata_talk:WikiProject_WLM? /André Costa (WMSE) (talk) 12:11, 3 July 2016 (UTC)Reply

Nationalmuseum English labels edit

Hi André, it looks like for quite a few items the English label is missing. Take for example Q18602071, I just added it. Maybe you can have your bot add the remaining labels? Multichill (talk) 07:23, 21 August 2016 (UTC)Reply

Hi Multichill. I'm currently doing a batch upload of the image associated with the Nationalmuseum Wikidata import. As a part of the preparation process I also got my hands on a better data source than I had for the original import. Once the image are up I'll add the new images, any available English titles and data about who the paintings depict to Wikidata. /André Costa (WMSE) (talk) 11:53, 7 October 2016 (UTC)Reply
Ok. Good to see that you're working on this again. Are you planning to do any other museums too? Multichill (talk) 15:33, 7 October 2016 (UTC)Reply
Boom. Lots of (English) labels, image (P18) and depicts (P180) claims =)
Only Nationalmuseum for the time being, although I might import more data from them now that I have some that has not gone via Europeana. I'll also try to remember to raise the issue with LSH the next time I get images from them to see if they can prep a dataset from their collections (at least one has a bunch of paintings). /André Costa (WMSE) (talk) 20:06, 16 October 2016 (UTC)Reply
nice jump. I hope other collections will follow. Multichill (talk) 17:00, 17 October 2016 (UTC)Reply

Re-run sourcing job edit

Hi André, can you re-run the job that did this edit? Does that also add missing claims? Someone deleted a lot of claims seemingly after a lot of junk was imported from the Italian Wikipedia. :-( Multichill (talk) 15:04, 30 November 2016 (UTC)Reply

I can re-run any of the jobs and they should add both missing claims and sources. Currently I'm patching a bug in the library though so until that is resolved I'm not comfortable with a new run.
That said I no longer add nationality due to this comment so a re-run wouldn't re-add that. /André Costa (WMSE) (talk) 08:01, 1 December 2016 (UTC)Reply
You have 4 accounts and you still manage to edit as an ip? ;-)
Right, the whole country thing sucks. That should be resolved. Somehow we need to model a way to get from Rembrandt (Q5598) to Dutch/Netherlands and Raphael (Q5597) to Italian/Italy. Preferably in a way that scales, is understandable and doesn't have two competing systems. Multichill (talk) 10:56, 1 December 2016 (UTC)Reply
Way more than four ;) but non logged in on my phone obviously.
Any solution would be nice. A "Modern day equivalent nationality" property/qualifier may be a way to get around any political issues. The problem is that the "true" nationality might actually contain less information than the modern day equivalent. E.g. Soviet Union or Roman empire. On the other hand nationality is such a fuzzy concept anyway. Are you Bavarian, German or an EU-citizen? /André Costa (WMSE) (talk) 11:19, 1 December 2016 (UTC)Reply

Missing collection edit

Hi André, why did you remove the collection from these paintings? Just clicking some random paintings shows these are all in the National Museum. The collection should be the same as collection (P195) statement and as qualifier of inventory number (P217) and should be the collection of which it's part, not some subcollection. In this case that's clearly Nationalmuseum (Q842858). Multichill (talk) 10:05, 27 December 2016 (UTC)Reply

That was due to #Data_upload_for_the_National_Museum,_Sweden above. And the related discussion on SOAP. /Lokal Profil (talk) 10:29, 27 December 2016 (UTC)Reply
That's unfortunate. What was your motivation for the creation of these subcollections? Are these historic different collections that got merged or just different departments? All the inventory numbers are in the same namespace so makes it inconsistent to put them in different collections. Multichill (talk) 11:05, 27 December 2016 (UTC)Reply
(Most) are historically different collection which were all brought under the same umbrella. Many of them are for that reason also exhibited primarily in different venues (think historical castles mostly). I don't particularly mind if they are under the main collection, but what I don't want is for the subcollection information to be lost (and of course I would need to find out so that I can update my scripts). /Lokal Profil (talk) 11:28, 27 December 2016 (UTC)Reply
Would be interesting to have full provenance information like with Q27024071. Do you have that kind of data? Multichill (talk) 12:17, 27 December 2016 (UTC)Reply
Don't think I have that detailed data anywhere =( /André Costa (WMSE) (talk) 13:23, 16 January 2017 (UTC)Reply
Sure they do. Probably hiding somewhere deep in the collection management system or on paper. Multichill (talk) 15:55, 16 January 2017 (UTC)Reply

Missing Wikidata items? edit

Hi André, I came across Commons:Category:Media contributed by Nationalmuseum Stockholm: connect to wikidata item and looks like we don't have any items for these works here on Wikidata. Do you know what happened here? Can you create the missing works? Multichill (talk) 08:30, 27 April 2017 (UTC)Reply

Hi @Multichill:. Remembering correctly these were images for which I hadn't already created the wikidata entries. Meaning they were not available via Europeana back in the day. It looks like (at least some) are available today e.g. this entry which goes with this image. I'm not sure whether the old code is still functional though and whether the later Europeana export differed in it's structure (which is likely since the first had several issues). It built something (at the same time as I did the image uploads) but I likely set it up to not create any new items. (Or rather it works of existing items). It should be possible to modify it to run of the full list of files(or at least the ones which were uploaded) and create entries where missing... No promises of when I'll get to it though :/ /André Costa (WMSE) (talk)
@Multichill: I took a closer look at the code I had to see if there was a quick patch to also create new items for the paintings (with images) which didn't have them. While it would be easy to add that to the wikidata component of the image upload the main issue is that it currently has no components for the claims that were originally included in the Europeana import e.g. P31, collection, inventory number and most importantly creator (which is a rich field containing multiple qualifiers). While this is definitely doable (I did it once from the same data for Commons) it makes it more than just a quick hack and as such it will have to wait for some holiday time. /André Costa (WMSE) (talk) 12:17, 26 May 2017 (UTC)Reply
I'll tell John you're really in need of vacation. Multichill (talk) 20:06, 26 May 2017 (UTC)Reply

I just ran into two of these paintings. Hoping you can get them on to Wikidata shortly! Jane023 (talk) 14:22, 28 November 2017 (UTC)Reply

@Jane023: Alicia Fagerving just did some more imports but I'm not sure whether the ones corresponding to these images were included in the data we got access to. /André Costa (WMSE) (talk) 15:58, 1 December 2017 (UTC)Reply
Hm nice to have an update, but no - the two files I was looking for were NM 548 & NM549. Strange that only half this category got done: c:Category:Paintings by Adriaen van Ostade in the Nationalmuseum Stockholm. Jane023 (talk) 17:40, 1 December 2017 (UTC)Reply

Svar på mejl edit

Svarar här istället för på wmse: då jag fastnade i Flow där!

Jag avstår. Frugan är fortfarande sjukskriven efter sviterna av en cancer-behandling och vi är på väg att flytta, så det är fullt upp. Jag gick själv in I väggen i våras, och det i sig är nog tillräckliga skäl att ligga lågt ett tag till. Det jag kan komma på är att vi skulle behöva utveckla bättre WD-integrerade mallar. Mall:Ortsfakta Sverige WD och liknande lösningar är inte särskilt kostnadseffektiva mätt i använda serverkrafter. -- Innocent bystander (talk) 07:45, 19 June 2017 (UTC)Reply

Hej. Jag har full förståelse för att du väljer att avstå under de omständigheterna. Tack för förslaget kring mallarna, dessa fanns inte på m radar sedan tidigare. Är det då främst Lua-modulerna som skulle behöva effektiviseras eller tror du att det mer rör sig om utveckling i integrationslagret för Wikibase/MediaWiki? Hälsningar André Costa (WMSE) (talk) 14:37, 20 June 2017 (UTC)Reply
Det är Lua-kod som behöver skrivas som kan stödja en hel mall. Med nuvarande lösning I "Ortsfakta Sverige WD" och andra liknande mallar så måste hela objektet laddas ner minst en gång för varje parameter I mallen. En lösning som tillåter att WD-objektet bara behöver laddas ner en enda gång för hela mallen vore betydligt bättre. Sedan tillkommer att data även måste hämtas från andra objekt, kanske även det flera gånger från samma ställe. Det kanske går att skapa billiga lösningar där också, men löser vi så bara det förstnämnda så ligger vi betydligt bättre till prestandamässigt. Idag klarar "sv:Stockholm (tätort)" inte att använda WD hela vägen just för det här problemet, och det är många fler funktioner som det vore önskvärt om vi kunde få in i en sådan mall.
Jag jobbar med geografi, men detta skulle kunna vara till stöd även för exempelvis biografier. -- Innocent bystander (talk) 15:29, 20 June 2017 (UTC)Reply
Tack för förslaget och underlaget. Vi tog fram en önskelistan-post baserad på den som jag skickade i ett wiki-mail till dig. /André Costa (WMSE) (talk) 11:08, 12 July 2017 (UTC)Reply

Alicia Fagerving edit

Since you're also part of WMSE, could you find User:Alicia Fagerving (WMSE) and ask them to respond to messages people are leaving? I don't see any responses from them on their user talk page or their bot's talk page since May even though both accounts are still actively editing. I assume they're not getting notifications, but whatever the problem is, they should find a solution, especially if they're going to continue running a bot. - Nikki (talk) 10:25, 5 September 2017 (UTC)Reply

Thanks for letting me know. I believe the bot talk page had fallen of both of our watchlists (I had assumed that it was redirected). /André Costa (WMSE) (talk) 13:59, 8 September 2017 (UTC)Reply

AndreCostaWMSE-bot edit

Hi, thanks for your edits. Wrt to these edits, please be careful. I don't know how that info is stored in Wiki Loves Monuments monuments database (Q28563569), but Iglesia de San Pedro Apóstol (Q22695464) it's not a Bien de Interés Cultural (Q23712) (BIC) and its P808's value is not 01110060007 (in fact, it does not have a BIC code). 01110060007 is a code granted by Andalusian Institute of Historical Heritage (Q5917182) (IAPH) and we haven't a property in Wikidata for that one yet (we have Guía Digital del Patrimonio Cultural de Andalucía ID (P3318), but it's not exactly the same). Having this "IAPH code" does not mean necessarily an item being considered BIC. strakhov (talk) 15:06, 6 November 2017 (UTC)Reply

@Strakhov: Answering you at Wikidata talk:WikiProject WLM/Mapping tables/es (es) to keep the discussion in one place. /André Costa (WMSE) (talk) 15:11, 6 November 2017 (UTC)Reply

drainage basin (P4614) edit

Hello. Would you mind moving the watersheds and water districts of Swedish bodies of water from part of (P361) to the new drainage basin (P4614)? I was about to move everything just like I've just moved this one but I then realized that you probably had put some serious efforts in adding those references. Can you have a look? Thierry Caro (talk) 00:57, 6 December 2017 (UTC)Reply

Hi @Thierry Caro. I've updated the code used by the bot and the instructions to make use of drainage basin (P4614) instead. Sadly there doesn't seem t obe an easy way of moving the statement from one property to another (with the reference). I might try doing a one-off bot run for that but right now I don't have much spare time so I don't know ehen I'll get to it. /André Costa (WMSE) (talk) 13:19, 29 January 2018 (UTC)Reply

Vandalism on soup kitchen? edit

Can you explain this edit to soup kitchen (Q2142654)? Daask (talk) 10:32, 15 June 2018 (UTC)Reply

Upon further review, I'm sure this was an innocent mistake rather than vandalism, but I'm still not sure whether the Peruvian bit should be split into a new item. See Wikidata:Project chat#Soup Kitchen as Peruvian cultural monument? Daask (talk) 14:42, 18 June 2018 (UTC)Reply
Answered in Project chat. Thanks for also pinging here. /André Costa (WMSE) (talk) 14:45, 18 June 2018 (UTC)Reply
Snap =) Yes, all of the Peruvian data should ideally be moved out into its own item. Would have been nice with a "split" functionality on items (or even just "copy to new item"). /André Costa (WMSE) (talk) 14:52, 18 June 2018 (UTC)Reply

Wrong ATE for cultural monuments edit

I've found a bulk of wrong values added by you, look e.g. https://tools.wmflabs.org/reasonator/?q=Q26233065. Can you please correct them? --Infovarius (talk) 09:43, 10 August 2018 (UTC)Reply

@Infovarius: I need a bit more informaton about what you think is wrong with Q26233065 and how you think I'm involved? /André Costa (WMSE) (talk) 08:21, 27 August 2018 (UTC)Reply
Because it is human and you've added him as a region for some monuments: like this. --Infovarius (talk) 10:43, 13 September 2018 (UTC)Reply
@Infovarius:Thanks for the clarification (the ATE abbreviation was unknown to me). The source for the error was an incorrect link on es.wikipedia which I've now fixed. I've also fixed the errors here. /André Costa (WMSE) (talk) 08:12, 14 September 2018 (UTC)Reply
Oh, sorry, I've forgotten to translate АТЕ from Russian (this means "Administrative division") :) And thanks for all the fixing! --Infovarius (talk) 14:33, 17 September 2018 (UTC)Reply

individual listed building complex and enskilt byggnadsminne edit

Good morning! I have been working with Cultural monuments lately and can find some editions made by members of WMSE like enskilt byggnadsminne (Q20871913) created by User:Väsk and individual listed building complex (Q24284072) created by User:André Costa (WMSE). Are there any specific difference between these two items. As for Agusastugan (Q20871954) and Agusastugan (Q30316427) i can not dsee that there should be any differece as the building is described as Agusastugan är en vinkelbyggd korsvirkesgård från de första decennierna av 1800-talet med möbler, husgeråd, kläder, och verktyg från slutet 1800-talet Can you help me to Discover the difference between individual listed building complex and enskilt byggnadsminne?
For the sake of good order I am also pinging @Alicia Fagerving (WMSE): Breg Pmt (talk) 10:01, 6 October 2018 (UTC)Reply

@Pmt: hi and sorry for the delayed reply. enskilt byggnadsminne (Q20871913) is the classification/legal motivation for the protection status (there are three classifications in Sweden). Enskilt referee to it not being a church or old governmental building but rather an individual picked building/group of buildings. individual listed building complex (Q24284072) simply refers to the grouping level of the item. Listed buildings are grouped into complexes which in turn get grouped on environmental level. A complex may consist of one or more buildings and protection status may have been determined on either building level or complex level. The complexes are what we (Wikimedia Sverige) imported to Wikidata.
The two items you linked to were examples of a complex with only one building. I've edited them a bit to clarify their relationship. Of course for a one building complex the distinction is just one of semantics. /André Costa (WMSE) (talk) 07:19, 15 October 2018 (UTC)Reply
Thank you very much for Your answer. The reason why I am asking is that I am trying to compare and find a relation between Norwegian Directorate for Cultural Heritage (Q1468738) and the Q17749638. As I understand now there is a concordance between Q17749638 and enskilt byggnadsminne (Q20871913)?. I think I also can find a relation to Rijksmonument (Q916333) and the Dutch Law Monumentenwet van 1988. Can you also provide the two other classifications together with enskilt byggnadsminne (Q20871913). I have made some editing on Q17749638 (With constraints) Breg Pmt (talk) 08:05, 15 October 2018 (UTC) .(With constraint)Reply
The other two are governmental listed building (Q19845339) and ecclesiastical listed building in Sweden (Q10551478). Note that technically the latter is not a subclass of listed building in Sweden (Q328070) even though it is frequently referred to as such. I don't know much about how the Norwegian system works though. /André Costa (WMSE) (talk) 15:59, 15 October 2018 (UTC)Reply

Torre de Recena and Castillo de Recena edit

Dear André Costa,

could you please merge the next entries in Wikidata: "Torre de Recena" and "Castillo de Recena"? I did not realise, when I created the one for Castillo de Recena in 2018, that there were already an entry for Torre de Recena created by you with the bot. I would suggest you to keep my entry, since the name "Castillo de Recena" is more realistic (it is not an only tower (Torre), but a castle, with many towers, walls, etc.), and it is more completed than the one you created in 2017.

Thank you very much in advance. Regards, --Veinticuatro de Jahén (talk) 10:35, 17 February 2019 (UTC)Reply

@Veinticuatro de Jahén: Hi. Sorry for the delayed reply, I've been away from work for an extended time. Now merged. /André Costa (WMSE) (talk) 08:31, 3 June 2019 (UTC)Reply

Kulturnav edit

Change [5] issued by (P2378) we are looking into how displaying Wikidata Properties of type External in Wikipedia would work see WikidatalightSalgo60 (talk) 07:39, 1 July 2019 (UTC)Reply

I like it. I think a combination of 3 and 3B is probably the most natural. Some identifiers might be famous enough to have their own articles whereas the "issuer" does not (or is less relevant). One would also have to be careful about the situation where there are multiple issuers for an identifier (e.g. ISBN). /André Costa (WMSE) (talk) 12:33, 29 August 2019 (UTC)Reply

Pages Q87721075 and Q43135303 to be merged edit

Hello André,

could you please merge both pages into one? I created https://www.wikidata.org/wiki/Q87721075 last week without knowing that it was already created. It is more completed than the other one, I would suggest to delete the page https://www.wikidata.org/wiki/Q43135303.

Thank you in advance.

Kind regards, --Veinticuatro de Jahén (talk) 11:23, 22 March 2020 (UTC)24deJahénReply

@Veinticuatro de Jahén: Merged. But there is a discrepancy with heritage designation (P1435) which I'd be happy if you could resolve. /André Costa (WMSE) (talk) 21:39, 23 March 2020 (UTC)Reply
@André Costa (WMSE): It is already solved. Thank you very much. Regards, --Veinticuatro de Jahén (talk) 22:25, 24 March 2020 (UTC)Reply

Page to be deleted edit

Hello André,

could you please delete this page https://www.wikidata.org/wiki/Q43137429? I already created https://www.wikidata.org/wiki/Q95628674, which is now much more detailed and complete. I didn't know that there was already one item in wikidata for it. Thanks in advance. Kind regards, --Veinticuatro de Jahén (talk) 18:00, 27 May 2020 (UTC)Reply

@Veinticuatro de Jahén: merged. /André Costa (WMSE) (talk) 12:10, 28 May 2020 (UTC)Reply

Page to be merged edit

Hello André,

sorry, the same occurred to me again. Could you please merged this page https://www.wikidata.org/wiki/Q105080216 with https://www.wikidata.org/wiki/Q43137133? I noticed too late that there was already one item in wikidata for it. Thanks in advance. Kind regards,--Veinticuatro de Jahén (talk) 17:24, 26 January 2021 (UTC)Reply

Looks like you, or someone else, already got to it =) /André Costa (WMSE) (talk) 22:55, 29 January 2021 (UTC)Reply

ÖdokBot edit

Hi, André Costa (WMSE), The last edit by your bot ÖdokBot dates from April 16th, 2017 (total number of edits only 64). I would propose to remove its bot flag due to inactivity. Do you oppose to that? --Lymantria (talk) 14:40, 6 December 2023 (UTC)Reply

@Lymantria Hi. Removing the bot flag is fine. The project that the bot was linked to was paused due to legal reasons and should it start up again it will likely work in a very different manner.
Cheers, André Costa (WMSE) (talk) 08:10, 7 December 2023 (UTC)Reply
Thank you. --Lymantria (talk) 10:49, 7 December 2023 (UTC)Reply
How about AndreCostaWMSE-bot? It has not been used since November 15th, 2017. --Lymantria (talk) 07:44, 31 January 2024 (UTC)Reply
@Lymantria It should be ok to remove that flag. @Alicia Fagerving (WMSE) has largely taken over the bot uploads at WMSE and should I need it I can a apply for a flag for AndreCostaWMSE-bot again. /André Costa (WMSE) (talk) André Costa (WMSE) (talk) 08:09, 2 February 2024 (UTC)Reply
Thanks again for your reply. Indeed, when needed you can apply again for a bot flag. --Lymantria (talk) 09:39, 2 February 2024 (UTC)Reply