Wikidata talk:WikiProject Cycling

@ TomT0m : j'ai avancé une première version. Est-ce que tu pourrais y jeter un coup d'œil pour voir quelles problématiques se dégagent ? J'ai pas mal de course en ce moment, mais j'essaye de passer chaque jour sur Wikidata. Cordialement, Jérémy-Günther-Heinz Jähnick (talk) 14:55, 21 August 2015 (UTC)

French

Could anyone finish translating this page into English? Matěj Suchánek (talk) 11:16, 18 November 2016 (UTC)

An user start this but doesn't finish, I avoid, because my english is horrible (and it is worst when I speak...). I will contact you on another topic to translate lines to have new functions for CS Wiki. Jérémy-Günther-Heinz Jähnick (talk) 12:26, 18 November 2016 (UTC)

FYI GZWDer changed the source language to French (thanks!), so the semantics is no longer a problem. But as there are less people speaking French, an English translation should still be priority. Matěj Suchánek (talk) 19:38, 15 January 2017 (UTC)

New template

Template:Cycling properties replaces Template:Sports properties where needed. Thierry Caro (talk) 16:41, 13 February 2017 (UTC)

step races (or not)

Hi, added a statement in cycling race (Q15091377) saying that races are eiher one day race or have several steps. Together with the item documentation template, in the talk page, the query

select ?item where {
    ?item wdt:P31/wdt:P279* wd:Q15091377 .
    {
        ?item wdt:P31/wdt:P279* wd:Q2912397 .
        ?item wdt:P31/wdt:P279* wd:Q1318941 .
    } .
}

Try it!

was automatically further and gives a list of races that are both one step and several steps at the same time interestingly it finds races who changed their configuration over time. I promoted the most recent one as preferred rank. More importantly that second query is also generated :

select ?item ?itemLabel where {
    ?item wdt:P31/wdt:P279* wd:Q15091377 .
    {
        ?item wdt:P31/wdt:P279* wd:Q2912397 .
        ?item wdt:P31/wdt:P279* wd:Q1318941 .
    } .
  SERVICE wikibase:label { bd:serviceParam wikibase:language "[AUTO_LANGUAGE],en". }
}

Try it!

who give a list of race instance we don’t know if they are one day race of several day from their type. This resulted in this edit : https://www.wikidata.org/w/index.php?title=Q506424&diff=625777674&oldid=622540094 to classify the world championship as a one day race :) @Jérémy-Günther-Heinz Jähnick, Thierry Caro: author  TomT0m / talk page 17:24, 2 February 2018 (UTC)

Merci @TomT0m:. Je ne connaissais pas cette propriété disjoint union of (P2738). Jérémy-Günther-Heinz Jähnick (talk) 08:26, 3 February 2018 (UTC)

Sous titre jaune « Chronologie »

@Psemdel: Bonjour, je propose de rajouter un sous titre optionnel « Chronologie» pour l'Wikidata:WikiProject Cycling/Documentation/infobox au dessus de précédent et suivant. En effet, cela rendrait plus lisible ces deux items (souvent masqué par la carte). Voici un exemple d'infobox qui utilise un sous titre comme ça sur Wikipédia fr : [1]. Cordialement, Nemo Le Poisson (talk) 08:49, 25 August 2019 (UTC)

Pourquoi pas.

The 'instance of' property is being overused

I've been trying to do some general queries of sports data, and I've noticed that cycling data makes heavy use of the instance of (P31) property in problematic ways. P31 is the simplest property there is on Wikidata, and should be used to classify entities in the broadest possible way. In cycling data, however, it appears to be used for much more minuscle classifications. There are other properties such as sports season of league or competition (P3450) and competition class (P2094) which should be used instead of instance of (P31). By way of example, let's look at 2018 Amstel Gold Race (Q51678681):

Current	Proposed
instance of (P31)Amstel Gold Race (Q478174) instance of (P31)1.UWT (Q22231106)	instance of (P31)cycling race (Q15091377) sports season of league or competition (P3450)Amstel Gold Race (Q478174) competition class (P2094)1.UWT (Q22231106)

You might ask, why does this matter? I'm sure Module:Cycling race has no trouble dealing with this schema; however, tools and queries which aren't designed specifically for cycling have more problems. The impetus for this post came earlier tonight, when I was trying to write a query to figure out what types of items use participating team (P1923). When I queried against instance of (P31), I got this:

If you click through to the original SVG file, you'll be able to hover your mouse over any bubble to see the full name. Or here's my original SPARQL query.

Not all of those tiny little bubbles are cycling-related, but it seems that a majority (or at least a significant plurality) of them are. And that's the thing – it's so hard to tell what all of those items are when there's so many of them. It's hard to get a high-level perspective on the data when the most general properties are bogged down with overly specific values.

So could this project please move towards changing the schema to fit better within the larger Wikidata ecosystem? Thanks, IagoQnsi (talk) 01:25, 13 August 2020 (UTC)

It is of course doable. To go back a bit, around 2015 I had asked for the creation of properties that had been refused to me at the time, and which subsequently appeared. If it is possible to adapt the algorithm, it is also necessary that all users are informed in their language of the property changes to be used. For my part, I can take care of updating the documentation and detailing it with specific examples, it will just be necessary that the Wikimedia foundation provides a small budget because it takes time. Also, don't forget that some versions of Wikipedia still use old algorithms. Jérémy-Günther-Heinz Jähnick (talk) 09:07, 16 August 2020 (UTC)

I would say that it should be done and the code will be adapted. This should not even a breaking change if modules are not updated in all languages — I guess that only some data will disappear from the infobox. @IagoQnsi: would you be able to move the value to a different propery using a bot? --Papuass (talk) 09:36, 31 August 2020 (UTC)

I read this now. The proposition above seems reasonable to me. I will have a look at the code, to make it compatible. Psemdel (talk) 15:45, 22 October 2020 (UTC)

@Papuass:, it took really a while, mainly because we had to agree on the properties to use, because it needed a big rework of lot of code and also to be sure that all WP don't have any issue. But now, the "1.1" and so on are gone from the P31! Psemdel (talk) 05:44, 8 June 2023 (UTC)

National federation data

I just finished a small project to provide accurate data about the national cycling federations and the continental confederations of the UCI, the results are described here: User:Rz98/Cycling/UCI Member Federations. Hopefully this can be of some use to this WikiProject. -- Rz98 (talk) 22:12, 30 June 2021 (UTC)

Just to say that the above-mentioned page was updated to reflect some recently admitted federations. -- Rz98 (talk) 21:41, 27 January 2022 (UTC)

Notebook to explore Tour de France's winners

I've developed a notebook which explores all Tour de France's stage winners : https://observablehq.com/@pac02/tour-de-frances-stage-winners. It could easily be done for other competitions such as Vuelta, Giro or any other competitions with multiple stages. PAC2 (talk) 06:38, 18 January 2022 (UTC)

Very interesting! To respond to the consistency issue that you raise on your page: all Tour stages from 1903 onwards have their own Wikidata item, date, winner, start and finish location, countries traversed etc. As for days with multiple stages, which were a regular occurrence from the 1930s to the 1980s, you should be able to rely on the follows/followed-by relations to get them in order; or just sort them alphanumerically. I did spend a couple of days last summer to process the vast amount of data that was already there and complete the missing bits, check consistency, and cross-check between multiple sources. So the TdF stages ought to be in a fairly usable state for database queries. I meant to write up some documentation about this but got distracted, your work may just give me the motivation to finish this part of the job. I'll leave a message here when I've written it up. -- Rz98 (talk) 19:29, 27 January 2022 (UTC)

Thanks for your feedback and for your work on the data. PAC2 (talk) 19:37, 27 January 2022 (UTC)

I've been documenting the query here : Wikidata:WikiProject Cycling/SPARQL/List of all tour de France's stage winners. PAC2 (talk) 06:12, 29 July 2022 (UTC)

National championships

Dear all,

I wanted to share with you some thoughts about the way the data about national championships is organized, and propose some improvements. Indeed I encountered frustrations trying to figure out how exactly the data is organized and how to use it for statistical evaluations, and other new users may have felt the same. I am ready to do the necessary work to implement the modifications that I'm proposing, but I would like to hear from the community first. My apologies in advance for the length of this post.

Last summer, I tried to get my teeth into national championships, e.g. I wanted to write some SPARQL queries to figure out who was national champion of a given country at some point, or compile somebody's palmarès. But I did not manage to penetrate the jungle of all the relevant items and gave up. Three nights ago I tried again, with more success, and these are the results.

The cycling module is doing an excellent job at automating the updating of many Wikipedia articles. I believe that it is equally important to help Wikidata users with understanding the data set, extending it, and evaluating it for their own interests, see e.g. the recently developed notebook of Tour de France winners. So my post is about these aspects. Currently, there seems to be no document describing the data model that new contributors are expected to work to. The documentation of the Cycling module provides partial answers, but it primarily addresses how the module uses data, but not the more general question of how the cycling-related Wikidata items are formatted. Users mostly need to navigate the Wikidata items in their browsers to understand how data is modelled, and that is not an easy task as I experienced.

Example

Wikidata items surrounding the 2021 Australian national road cycling championships

As an example, here is a graph depicting the data items surrounding the 2021 Australian Road Cycling Championships, with the relations instance of (P31), subclass of (P279), part of (P361), and has part(s) (P527) connecting them. If you open this version, then you can click on any node to jump to the Wikidata item it represents. I picked Australia simply because they were high up in the alphabet, with good data, but the picture for other countries would be similar. A championship race has three attributes:

• the nation (always Australia in this example)
• the year (2021 in this example)
• the competition, such as men's elite time trial; I am using M for men, W for women, R for road race, and T for time trial.

As an example, the node 2021 AU WR stands for 2021 Australia National Road Cycling Championships Women RR (Q104306250), and I call it a node of type NYC, because it stands for the national championship race in a certain nation, in a certain year, in a given competition. A YC node would be for a given year and competition, such as 2021 national road cycling championships - women ITT (Q104305086), and list the respective races in each country. Nodes of the same type have the same colour. The NYC nodes at the bottom contain the actual results, the others contain lists that allow, e.g., the nationalchampionships function to do its job.

These three attributes naturally define a partial order, where national road cycling championship (Q2306612) is the highest element (NRCC in the graph), and e.g. a Y-node like 2021 (Q104303043) is above an NY-node but incomparable to an N-node. There are complications such as AU T, representing time trial at Australian championships (Q3497134). These exist because people wrote Wikipedia articles about road race and time trial disciplines separately; let's call them ND-type nodes, and they can go between types N and NC, but these do not exist for all countries. (And indeed the interpretation of AU R is unsure, as we shall see.)

The arrows indicate the Wikidata relations that exist within this sample:

• Thick black arrows indicate a has part(s) (P527) relation from the upper to the lower element, and instance of (P31) in the other sense.
• Thick red arrows indicate a has part(s) (P527) relation from the upper to the lower element, and part of (P361) in the other sense.
• Thin arrows indicate one-sided relations, with solid red being has part(s) (P527), dashed red being part of (P361), dashed black instance of (P31), and dashed green subclass of (P279).

Finally, a red border indicates items linked to Wikipedia articles, a green border those referenced by the nationalchampionships function, and a dashed border items created by CyclingInitBot. All relations touching the latter items were generated by the bot. Within the bot, items with red names are fixed constants from a table, and items with green names are looked up by their French Wikidata label. The bot and the module use has part(s) (P527) of some of these items. The other items and relations seem to have been created by hand, and do not seem to be exploited by any module, as far as I can tell.

Problems

In the example, the relations broadly bear out the aforementioned partial order, but not uniformly:

• Some relations are both ways, some are one-way top-down, and others are one-way bottom-up. One-way relations make it particularly difficult for new users to find their way through the data when browsing.
• Some nodes are completely unreachable, notably 2021 MR and 2021 MT. I only found them because I suspected their existence, using the What links here function. This way I found not one of each but two, serving two distinct sets of countries.
• Top-down relations are uniformly has part(s) (P527), but bottom-up relations are either instance of (P31) or part of (P361), seemingly chosen at random. This too makes it more difficult to navigate the data. The occurrence of subclass of (P279) is a one-off.
• Women and men nodes have different links, see e.g. the uplinks from 2021 MR and 2021 WR. AU MT links to AU R instead of AU T, and the downlinks from AU are quite erratic.
• Also, remember that nobody tells people that this hierarchy exists, or what nodes they can expect to be there at all. The graph may make things look reasonably straightforward, but I had to invest some hours to infer it. The interpretation of what an item represents is just mine since nodes do not "speak" for themselves; e.g. their "colour" and attributes are only inferrable from labels, which are sometimes contradictory (see below). E.g., a SPARQL query that would find, say the 2018 women's ITT from Norway, would have to search for the correctly constructed label in French, instead of properties attached to the item.
• One simple thing in particular confused me enormously: the labels for NY-type nodes (e.g. 2021 AU), generated by CyclingInitBot, are not equivalent in English and French! See Q104306235:
  • Its French label correctly describes its function: Championnats d'Australie de cyclisme sur route 2021.
  • Its English label says 2021 Australian National Road Race Championships – note road race (equivalent to course en ligne) and not road cycling (equivalent to cyclisme sur route). For anyone using English labels, this suggests that the type is NYC and not NY, and one would not expect to find time trials under this label.
• The resulting confusion is magnified because the uplink from 2021 AU is AU R (Q2564187) and not AU (Q105397574), as one would expect. The English and French labels have the same difference of race vs championships, and that reflects different approaches in their wikipedias: the English wikipedia has created separate articles for road races and time trials in most nations' championships, whereas the French wikipedia tends to have a single article comprising both disciplines. And somehow, in Wikidata, both the English road race and the French road championships articles got mixed up. So maybe my interpretation of AU R for Q2564187 is wrong and this is actually AU (of which there then would be two). But anyhow, this shows that interpreting items based on just their labels is prone to errors.

Objectives

The overall aim of the proposal (below) is to make the data model more robust and user-friendly:

• Simplify the structure and adapt the existing items to the new structure.
• Document the structure within the WikiProject Cycling, so that other people can use it.
• Write a verifier that checks if the structure is intact, in case of inadvertant edits that could corrupt it.

I am willing to put in most of this work, but I'd like to hear other views first. Some modifications would be necessary in CyclingInitBot, if Psemdel (its creator) agrees to them. No modifications would be required for the Cycling Module, at least as far as I can see.

Proposals

Proposal for restructuring the national cycling championships data

Proposals for the structure:

• All relations in the above diagram should go both ways. This helps new users to navigate and understand them intuitively, and to find all the places they need. This should also contribute to fewer item duplications (like 2021 MR resp. 2021 MT).
• Uniformly use has part(s) (P527) and part of (P361) for these bidirectional relations. In cases where one item would have links to two or more different types of items, qualifiers can be used to say which type of item one links to. This would not concern the nationalchampionships function, whose link targets would not change. (Even better would be to have three different pairs of relations, but which ones?)
• The instance of (P31) relation can now be used to indicate the type of node (N, Y, C, NC etc), e.g. cycling championship by country and category. This helps create a verifier and unambiguously states the function of an item, independently of the labels, and independently of Wikipedia articles attached to it, which may differ or lack in various languages.
• For nodes pertaining to a competition, give them tags identifying discipline and rider category. For the former, individual road race (Q51541990) and individual time trial (Q2266066) exist. For the latter, one would create six competition classes (ME, WE, MU23, WU23, MJ, WJ), in analogy with the approach followed by other sports, e.g. women's U22 association football (Q106584244) or men's U20 basketball (Q80841061). These items only need to be translated once into every language, but they would give meaning to every competition item they are used in. Also, users could write queries that would directly identify, say, the above-mentioned 2018 Norwegian women's ITT, by specifying these attributes in a query. The relations that could be used are sport (P641) and competition class (P2094)
• French and English labels of NY-items should be made equivalent.

The resulting structure is depicted on the right. Only one competition type is shown because the others would be analogous. The ND-type nodes are optional and would only exist where people have written such Wikipedia articles, since they are entry points from Wikipedia. This scheme also allows to deal with cases where we have existing Wikidata items representing separate junior events, or separate men/women events.

Finally about the verifier I mentioned. With the new structure, one could unambiguously determine the type of an item, and with the properties for country/year/discipline/rider category, determine what it stands for. It is then possible to check that the item links to all the items that it needs to link to, and is linked by them in return, and to check their types and attributes. In the example, this would allow to detect, e.g., that there are two different 2021 MT items, or that AU WT and AU MT items have different uplinks. I could write a tool that periodically automates these checks.

That's all. Sorry again for being so lengthy, and thanks in advance for your comments. -- Best regards, Rz98 (talk) 22:44, 26 February 2022 (UTC)

Discussion

Before I start reading, the National championship structure was a nightmare. We had I don't know how many discussions with GAN on the topic. The structure is documented here. But I myself had to look where I put that... Nevermind the difficulty by National championship comes from:

• The different Wikipedias are organized differently on the topic: WP:fr has an article for the national championship/country for all categories. WP:en has an article for ITT and one for Road race. Somehow the code of wikidata should allow embracing both.
• Producing this table here is very useful, as manual update of championship data is very time consuming. Because of the complex structure of the championships, queries tend to be very performance consuming (filter group, look here, look there). That's the only reason why there is 2021 national road cycling championships - women RR (Q104304845). It is a kind of cached query if you want. Psemdel (talk) 08:58, 27 February 2022 (UTC)
• " looked up by their French Wikidata label." yes very sorry for that... Did not find a better idea at that time. Can be fixed.
• Men and women are handled differently... because I take care only of women (already enough work). By men, the structure was drafted by Anthony59999 in 2016. See 2016 French National Time Trial Championships (Q27048422).
• Concerning the "road". An ITT is a race on the road, at least in French ;), it is also the way the UCI sees it https://www.uci.org/disciplines/all/2tLnZMo6WrUBplRXDxyEi7 But ok, there was a misunderstanding. I think it is only by Australia and some other English speaking countries.
• "And somehow, in Wikidata, both the English road race and the French road championships articles got mixed up. " --> yes
• As you understood, we need a kind of double inheritance (NYC and discipline). That why we used P527 and P31 in a quite similar way. Which is not clear. The problem occurs at many other places from the data for the cycling.

So about your proposal. Clearly we all agree that there is room for improvement. I just don't understand NYC will have 3 items in its P527? Then how will the module determine which one is what? For P2094, good idea (in P31, the determination if it is a class or the race parent is at 500 lines code in the module :D, see the discussion above where I also agree but we never had a consensus about the new structure to use). Psemdel (talk) 09:57, 27 February 2022 (UTC)

Thank you for the pointers to the other discussions, very useful. Regarding your question: My formulation may have been ambiguous. I meant that P527 ("has part") would systematically be given to the "upper" item, pointing to the "lower" item, and that P361 ("part of") would for the other direction. So NYC would not have any P527 (like now). Something like YC would have many P527, but all of them to NYCs, one for each country (like now). A Y-node would have P527s for both NY and YC nodes. For P361, the situation is inversed: an NYC would have three P361 (for its nation, its year, its competition).

I believe (correct me if I'm wrong) that the nationalchampionships function only ever uses the P527 relation of YC nodes, so no change would be required here. As for the bot, it also only consults the P527 relation, but it does so for N and NC nodes (the one with red labels in my drawing), in order to link to the previous years. But there it identifies the right year by label, which would still be possible. This is why I believe that little change is required to let existing code function as before.

However, if requirements change, then nodes of different types could be distinguished by qualifiers. E.g., in a Y node, the P527 statements for NY could have a country-qualifier, and those for YC a year-qualifier. Indeed, for performance reasons it could even be interesting to do this even where there is only one type of successor. Union Cycliste Internationale (Q48663) is organized like that, e.g. this query finds all UCI member federations and their countries by just looking up the UCI-item. A similar scheme could be useful for the bot. E.g. in the link_year function it could look at the qualifiers to identify the right year instead of looking up the labels (each label is an extra query unless I'm mistaken).

Edit: I only belatedly understood your comment about P31. So what you're saying is that P31 is deliberately distinguished from P361 in NYC-nodes to make things like Infobox work? (That's the raceLink function in the Lua code if I understand correctly.) In that case, what I said about the module not needing any changes would not be entirely true. Would it also work to have a P361 with a "year" qualifier to indicate the race? Or would that be undesirable?

By the way, my basic assumption was that all those items (NY, YC, NC etc) were considered desirable. When you are saying that they only exist as "cached queries", maybe my assumption is wrong, and the structure could be simplified. But on the other hand, I think they are more than just cached queries. People who are coming over from Wikipedia by clicking on the "Wikidata item" for the first time, they have only Wikidata's browsing interface to go by and cannot be expected to write SPARQL queries just to get around. Having those items improves usability, though there may be other ways.

About the road race/championship issue: Yes, I agree that ITT is a race on the road ;-) It is the inconsistent use of the English term "road race" at UCI and at Wikipedia that is causing all of the confusion. UCI calls "road races" (cyclisme sur route) a discipline of cycling comprising the race types of "Individual road race" (course en ligne), "Individual time trial" (contre-la-montre) etc. So in French everything is nicely distinguished. But Wikipedia tends to shorten "Individual road race" to "road race" (as a race type) and "road races" to "road cycling" (as a discipline). Which in turn conflicts with road cycling as a general mode of transport. Anyway, if we used P31 to indicate that some item is the N-node for its country, we can separate the issues of data storage from conflicting labels and Wikipedia articles. -- Best regards, Rz98 (talk) 13:37, 27 February 2022 (UTC)

• .
  • As Psemdel wrote above, a structure for national championships was made in the module, taking into account the existing system of articles in different Wikis.

- national championship (in all wikis there is a single article about the champion or only about RR, such as in en-wiki)

--- only RR (there are several articles in ru-wiki)

-----men RR

-----men RR U23

-----men RR U19

-----women RR

-----women RR U23

-----women RR U19

--- ITT only (articles mostly en-wiki)

-----men ITT

-----men ITT U23

-----men ITT U19

-----women ITT

-----women ITT U23

-----women ITT U19

if necessary, according to this scheme, you can add any other discipline. I have seen championship elements for criterium and uphill racing in some championships.

- separately any national championship for any year (including all possible disciplines)

As a result, when using templates based on the module in which the national championships are indicated, we can go to any of these articles. If we place all the results in one article through a module, then there will not be enough resources for processing. And so, if necessary, you can create separately necessary ones. (Ru-wiki has 2-3 championships, which, in addition to the main article, have separate articles for all RR and all ITT.)

I gradually add/create/edit championship results (winners by year) according to this scheme. To begin with, I make championships with a small number of tournaments held. Not fast, as you have to create missing racers and collect some data from the Internet.

— GAN (talk) 10:53, 27 February 2022 (UTC)

I agree that junior and U23 levels should also find their place. I only used men/women elite IRR/ITT in my example because it was already complicated enough :-) But junior and U23 races can be accomodated in the structure alongside the four elite competitions, they would simply be additional competitions in those countries where they exist (in terms of my example, the "C" attribute would have more than just 4 values). Hill racing seems to be unknown in UCI regulations, but if they are recognized at national-federation level, why not, they would be another competition. Just as long as they fall into the "discipline" (in the UCI sense) of road cycling/cyclisme sur route. Results from other disciplines like track or cyclocross ought to go into separate structures, otherwise the competitions become too unwieldy. -- Best regards, Rz98 (talk) 13:37, 27 February 2022 (UTC)

The Australian Championship is the only one that has THREE (3) articles on the en-wiki. General (Q105397574) - not linked to anyone, RR (Q2564187) - linked to all other wikis, ITT (Q3497134) - linked to similar ones on other wikis

If I understand correctly, then additional parameters are needed to improve search queries:

--- P2416 - to indicate discipline - Q51541990 or Q2266066

--- P2094 - to indicate gender and age - Q106097507 (M) / Q1451845 (W) + Q14510042 (U23) / Q21193184 (U19) + Q15991269 (amateurs) / Q1369650 (prof) or create for this field the values ​​you proposed immediately for the combination "gender + age"

There were also RRs among amateurs and professionals (GDR and USSR) in the late 80s and early 90s. And now championships are being held in a multi-day race (18 and 19) — GAN (talk) 14:30, 27 February 2022 (UTC)

First, don't hesitate to ping me, I am not so active right now. I think the best way to make everything very clear would be to create the proposed structure for an imaginary country (or a country where there is no much cycling). As written before, I agree that right now it is a mess. I just need to be sure that we can still display with the new structure without exceeding resource quotas. Psemdel (talk) 09:37, 7 March 2022 (UTC)

That sounds like a good idea. I made a start a couple of days ago with the 2022 UCI Cyclo-cross World Championships (Q104626552) to get a feeling for how it could look like - also used the rider categories there. Other than the six standard categories there is now also XE/mixed elite (for team relays) and AMA/Amateurs; those are the categories I could find in the UCI database, they seem to confound pre-1993 professionals with post-1993 Elite. I guess I'll pick a country with not so many races and create a proposal with that; will probably do so over the weekend and ping you here. -- Best regards, Rz98 (talk) 20:33, 9 March 2022 (UTC)

Property:P4958

What is the purpose of that property? Description says it is used for bicycle races in "hispanic language countries". But the property is only used for two competitions in Italy and one in the Netherlands, neither Italian nor Dutch are hispanic languages. So either the description or the usage of the property is incorrect. Comparing English and Spanish labels for Q25831754, Q27067359 and Q27104688, it looks like "metas volantes" is a Spanish translation of "intermediate sprints". Maybe the property should be changed so it can be used for "intermediate sprint classification" of bicycle races in any country. 90.227.175.218 17:26, 26 March 2024 (UTC)