Hello! Unfortunately, you have not made the required 5 admin actions in the last six months and your admin permissions were removed per WD:A. Thanks for your service to the project!
User talk:ValterVB
About this board
 |
On this page, old discussions are archived. An overview of all archives can be found at this page's archive index. The current archive is located at 2015. |
Previous discussion was archived at User talk:ValterVB/Archive 1 on 2015-08-10.
Dear colleague,
Some time ago your bot added the "diameter" property from Gazetteer of Planetary Nomenclature to many features on planetary bodies. But some of these features are not round (canyons, cracks, faults), and thus don't have diameter in the sense of Property:P2386. The values in Gazetteer are actually size, not diameter. Moreover, these numbers usually have false precision (for example, "3761.28 km" for Valles Marineris). Couldn't you, please, correct this: mark these numbers as "size" or "length" instead of "diameter" for non-round features (i.e., all planetary features except craters) and round them appropriately?
With best regards, ~~~~
Technically, any form has its diameter. By definition, diameter of a set is a maximum of distances between two points of it.
In discussed property:P2386, in geology, in planetology and in the ordinary human language non-round objects do not have diameter. And it is only part of the discussed issue.
I have used the terminology used in the site. Example: for Rupes Recta tha page is this. You can see that the word used is "Diameter" and the value is "115.95 KM". In each item that I have edited, you can see the link to the page used for retrive the values in property Gazetteer of Planetary Nomenclature ID (P2824)
I understand that you copied data from the site, but this data is obviously incorrect (it is a site about names of the objects, not about the objects). When filling Property:P2386 we must use single terminology, according to the definition of this property: "the diameter of a circular or spherical object". Cracks and ridges can't have diameter. Moreover, size of such objects can't have 5 or 6 significant digits, it is obvious false precision. Moreover, their values can have nothing in common with reality at all (e.g., size of Caloris Planitia was twice undervalued before I asked Gazetteer staff to correct it). Moreover, size of many large features (hundreds and thousands km) is stated as 0 km in Gazetteer (1 , 2), and you copied even this: 1, 2.
I don't think that the data is incorrect, probably is the diameter of the circle that include the object.
Sorry, you are not even sure what did you copy, but you "don't think" that it is incorrect? Yes, it is diameter of the circle that include the object, it is not diameter of the object itself, as you presented it. Do you think that size of Shangri-La "0 km" is also correct? And 10-meter "precision" in the "diameter" of Valles Marineris "3761.28 km" is also correct? And there is meaningful difference between this "diameter" and length of 3770 km?
I know what I have imported but it was 2/3 years ago. I have more of 20 million of edit with my user+my bot, so I can't remember all I have done. In that period I have downloaded all the db of "Gazetteer of Planetary Nomenclature" and I have reported all the data here. I don't see error, 0 km is the value reported by "Gazetteer", so it is correct respect to the source. We only report the data as indicated in the databases.
Shangri-La stretches from 150 to 200° W and from 20°S to 20°N. It is more than 2000 km long. You can see it on any map. Gazetteer of Planetary Nomenclature is a source about names of the features, not features itself. Its goal is to provide possibility to match the name and the feature, nothing more. It contains lots of incorrect and totally wrong sizes. They just did not measure albedo features on Mercury, Mars and Titan because these features are often diffuse, hence "0". For other features, they just give results of semi-automatic measurement, hence the false precision. Imagine measurement of Sahara with 10-meter precision - it is an equivalent absurd. I rewrite illiterate planetological articles in Wikipedia for 7 years, and now we have an additional flow of illiterate information from Wikidata. Data from the Gazetteer can be made useful: all zeros must be removed, all "diameters" must be rounded to 2-3 significant digits and "diameters" of non-crater features must be renamed into length (and ignored when there are length from other sources).
ciao per caso ho trovato due identiche elementi: Q64902630 e Q38641643 si tratta della stessa persona ovvero un ricercatore.. uno dei due va cancellata o spostare in redierct :)
Hanno ORCID iD (P496) diversi, fino a quando Orcid non li unisce noi non possiamo unirli. Ho provato a scrivere al sito vediamo se li uniscono.
ok aspetteremo.. :)
Mi hanno risposto, il senso è più o meno questo "Abbiamo esaminato gli account e trovato informazioni sufficienti che indicano che appartengono a due diversi ricercatori, sebbene il loro campo di ricerca possa essere simile." Quindi direi che è corretto che siano item differenti.
bene quindi sono due voci differenti.. grazie ValterVB :)
@GerardM: I have deleted Q30253119 2 years ago because was only name and surname in label english and dutch, P31=Q5 and nothing else. No indication in edit summary and no what's link. --~~~~
It is linked in a Scholia with several others. That is only possible when these links existed at the time. ~~~~
This is the Scholia with several missing items.. https://tools.wmflabs.org/scholia/author/Q57863411 My problem is not so much in them having been deleted. I do not know how to correct things.
@GerardM: I restore them: Q30253119 Q30253121 and Q30253127. But the strange thing is that "What links here" did not reported the link. Now I have forced the update of Q28984688 with an edit and now Q30253119 appear. Probalby when the items was created there was a cache issue.
Thank you :)
the way SourceMD works is that new items for authors or papers are created in the first run. A second run will bring authors and papers together.. As a consequence, it makes sense to run a SourceMD job for all authors that are new and do not have linked papers.
Hi,
I see that you added a lot of "melldroader en Italia" as Breton description for italian footballer. It's not really incorrect but literraly it means "footballer in Italy". Is it on purpose? Am I missing something?
If not, could I replace these descriptions with "melldroader Italia" (footballer of Italy) which is the more usual way to say it in Breton.
I have used translation in MediaWiki:Gadget-autoEdit.js. If is more correct "melldroader Italia" is better ask in discussion page of the gadget
Ciao,
non capisco perché il nome "Teymouraz" non può esser considerato come semplice variante locale di Teimuraz (Q7695055)... Difatti, sono due traslitterazioni dello stesso pronome persiano, თეიმურაზ.
Buona serata,
Perché anche in francese si usa "Teimuraz" e quindi in quella voce va indicato quella versione, se è una variante di quel nome in georgiano no persiano al limite lo puoi mettere negli alias. La gestione dei pronomi è un po' particolare ma ha una sua logica. In Wikidata:WikiProject_Names trovi maggiori informazioni. Intanto ripristino e aggiungo l'alias
Grazie mille per il ripristino e le spiegazioni! Ma in quali fonti francesi si usa anche "Teimuraz"... ?
Ho controllato con il "Puntano qui" e ho visto che c'erano item che usano "Teimuraz" come label in lingua francese. Ora però mi è venuto un dubbio, provo a chiedere al progetto.
Vedi Wikidata_talk:WikiProject_Names#Doubt_about_not_latin_name
OK, proverò a seguire la discussione
Hi Valter, thanks for your note on the portal. Can you please point me to a simple example to fetch an entity using page name and finding its QID? thanks, ~~~~
Premise: I have created this framework to elaborate many items.
Below you can see 3 examples:
- Find single item using only Wikidata API
- Find multiple item using only Wikidata API
- Find multiple item using only Wikipedia API
If something isn't clear ask me.
/// <summary> Find single item from wiki and page using only Wikidata </summary>
private void button39_Click(object sender, EventArgs e)
{
Site WD = new Site("https://www.wikidata.org", user, password); //Login
string strJson=WD.LoadWD("enwiki", "Douglas Adams"); //Search page "Douglas Adams" in english wikipedia
Entities EntityList = new Entities();
EntityList = JsonConvert.DeserializeObject<Entities>(strJson, new DatavalueConverter()); //create a collection of entity: only one in this case
Entity entity = EntityList.entities.Values.First(); //extract the first and only item
string id = entity.id; //extract Q number
string label=entity.labels["en"].value; //extract english label
string description = entity.descriptions["en"].value; //extract english description
txtMessage.Text =
"Qnumber: " + id + Environment.NewLine +
"Label: " + label + Environment.NewLine +
"Description: " + description;
}
/// <summary> Find a list of item from wiki and page using only Wikidata </summary>
private void button39_Click(object sender, EventArgs e)
{
Site WD = new Site("https://www.wikidata.org", user, password); //Login
string strJson=WD.LoadWD("enwiki", "Douglas Adams|Italy|Jimmy Wales"); //Search page "Douglas Adams" in english wikipedia
Entities EntityList = new Entities();
EntityList = JsonConvert.DeserializeObject<Entities>(strJson, new DatavalueConverter()); //create a collection of entity: three in this case
if (EntityList.entities != null)
{
foreach (Entity entity in EntityList.entities.Values) //loop on each entity
{
string id = entity.id; //extract Q number
string label = entity.labels["en"].value; //extract english label
string description = entity.descriptions["en"].value; //extract english description
txtMessage.AppendText(
"Qnumber: " + id + " - " +
"Label: " + label + " - " +
"Description: " + description + Environment.NewLine);
}
}
}
/// <summary> Find a list of item from wiki and page using only Wikipedia </summary>
private void button39_Click(object sender, EventArgs e)
{
Site WP = new Site("https://en.wikipedia.org", user, password); //Login
string strJson = WP.LoadWP("Douglas Adams|Italy|Jimmy Wales");
Pages pages = new Pages();
pages = JsonConvert.DeserializeObject<Pages>(strJson);
foreach (Page page in pages.query.pages.Values) //loop on each page
{
txtMessage.AppendText(page.title + " = " + page.item + Environment.NewLine);
}
}
That worked! Thank you very much for the example. Is it possible to package the entire library into a single DLL? Right now I am having to include all the classes (site, pages, utilities etc). I am also having to include the change namespace on the classes.
I havent time to do it now, maybe in future...
Understood, no problem.
One more request, can you provide me with sample code on how to create a new item like Q60204555 with all its statements and interwiki links?
You can skip this one. I have decided to batch add items using Quickstatements.
OK, but if you want, you can see this page The example is for edit one item, if you wan create a new item you must copy only from // Edit entity to the end and change last row in WD.EditEntity("", Sitelinks, Aliases, Descriptions, Aliases, Claims, "BOT:Test"); BOT:Test is the object of the edit.
Hi Valter, the getentities requires that a enwiki link is present. How do I retrieve wikidata items that do not have an interwiki link for enwiki using just the text and language?
I solved the above by creating a search class that can work with wbsearchentities.
public class Searchinfo
{
public string search { get; set; }
}
public class Matches
{
public string type { get; set; }
public string language { get; set; }
public string text { get; set; }
}
public class Search
{
public string repository { get; set; }
public string id { get; set; }
public string concepturi { get; set; }
public string title { get; set; }
public int pageid { get; set; }
public string url { get; set; }
public string label { get; set; }
public string description { get; set; }
public Matches match { get; set; }
}
public class Searches
{
public Searchinfo searchinfo { get; set; }
public List<Search> search { get; set; }
public int success { get; set; }
}
public string SearchWD(string lang, string pages)
{
string post = "";
post = string.Format("action=wbsearchentities&format=json&language={0}&search={1}", WebUtility.UrlEncode(lang), WebUtility.UrlEncode(pages));
return PostRequest(_url + _api, post);
}
Yes, this is a possible solution but wasn't implemented in my BOT because I've never needed it. I had implemented the simple search:
private void button39_Click(object sender, EventArgs e)
{
Site WD = new Site("https://www.wikidata.org", user, password); //Login
var res = ListGenerator.ListFromSearch(WD, "Torino", "0"); //Return a tuple res.count=n° of items res.list=item separated by pipe
if (res.count != 0)
{
txtMessage.Text = "Numebr of items: " + res.count.ToString() + Environment.NewLine;
txtMessage.AppendText(res.list); //list of items to be passed to LoadWD
}
}
Ciao, rieccomi per il 2019!
Dunque, oggi vorrei sottoporti i seguenti elementi: Cy Young (Q362264), Race Imboden (Q389550), Lyle Bettger (Q2008186), Billy Hatcher (Q2903651), Lyle Talbot (Q3039327), Taylor Mays (Q7690055). A seguito di unioni con elementi contenenti unicamente la corrispettiva categoria di Commons, questi elementi presentano un altissimo numero di descrizioni "categoria di un progetto Wikimedia" evidentemente sbagliate. Ora, in casi del genere tu mi consigli:
- di annullare l'unione, ripristinando i due elementi separati
- nel qual caso c'è un modo per caso di salvare le modifiche successive all'unione o bisogna procedere manualmente?
oppure
- di mantenere l'unione, rimuovendo le descrizioni sbagliate
- nel qual caso c'è un modo per rimuovere solo le descrizioni "categoria di un progetto Wikimedia" o bisogna procedere manualmente?
Grazie mille!
Io in questi casi normalmente cerco il merge sbagliato e annullo quello, poi ripristino la voce che è diventato un redirect, sposto il sitelink senza usare il merge e cancello la voce che essendo stata creata come redirect non ha molto senso unire. Forse è meglio se mi spiego con un esempio:
- Q1 è un item normale
- Q2 è un item di categoria
- Q2 è stato unito in Q1 generando il casino delle descrizioni.
- Io normalmente annullo il merge, e ripristino Q2, poi sposto il link da Q2 a Q1 e poi cancello Q2.
Perfetto ... e, per concludere, secondo la tua opinione tali unioni sono da considerarsi sbagliate? Perché io sono sinceramente in dubbio sul da farsi.
Personalmente unire categorie o disambigue o liste o template in altri tipi di item, io lo considero sempre sbagliato e cerco di ripristinare, poi verifico se doveva essere un semplice spostamento di link o un errore.
Ah, aggiungo una curiosità: più in generale ti risulta che esista un qualche tool col quale si possano individuare e rimuovere le descrizioni "categoria di un progetto Wikimedia" in elementi privi di instance of (P31) Wikimedia category (Q4167836)?
Partiamo da questa: visto che sei amministratore puoi incollare mw.loader.load( '//www.wikidata.org/w/index.php?title=MediaWiki:Gadget-dataDrainer.js&ctype=javascript&action=raw' ); nel tuo Special:MyPage/common.js a questo puno nel menù dove trovi anche il merge trovi una nuova voce "Svuota" selezionadola di darà la possibilità di cancellare: etichetta e/o descrizione e/o alias e/o sitelink. Quindi fa parzialmente quello che chiedi:rimozione descrizione.
DataDrainer l'ho già attivato e l'ho usato molto spesso in questi anni (non ricordo, forse tu stesso me l'avevi consigliato) ... il problema è che cancella indiscriminatamente tutte le etichette/descrizioni/alias/sitelink, quindi anche le (poche) descrizioni buone. Non esisterebbe anche uno strumento più duttile?
Che io sappia no.
Ciao! Ho appena trovato circa 600 categorie di es.wiki, queste, prive di P301 e della reciproca P910: basta che ti basi sul fatto che la voce principale della categoria in es.wiki dovrebbe avere esattamente lo stesso titolo della categoria. Ovviamente ci saranno dei casi in cui non troverai una voce in es.wiki, quei casi poi li sistemerò io a mano (es. ti anticipo già che per la Germania tra 1968 e 1988 non capisco il senso di quelle categorie, perché esistevano Germania Ovest e Germania Est, non la Germania unita). Grazie mille!
Queste sono le categorie senza voce corrispondente:
Per le altre ho il file pronto ma QuickStatements non funziona (non mi fa fare il login). Se non riparte tra poco le incollo in una pagina così si può usare quando si riprende.
Quickstatements sembra non funzionare, nessuna delle 2 versioni, per ora ho parcheggiato i dati in User:ValterVB/sandbox