Wikidata talk:WikiProject Astronomy

From Wikidata
Jump to: navigation, search

Welcome to the Astronomy task force[edit]

Just a welcome to this new task force and I wish it a long life in the wikidata community. Snipre (talk) 13:23, 20 March 2013 (UTC)

Thank you very much! I hope this task force will reach to aggregate discussions concerning properties of astronomic objects as Chemistry and Molecular Biology task forces. --Paperoastro (talk) 08:50, 21 March 2013 (UTC)
I just planted a message on the Astronomy project of en-wiki. We have to advertise this project a little more so the international community knows where to look for astronomy-related topics on Wikidata. Plus we need the multiplication factor of editors checking the Wikidata-items for consistency. Can somebody notify a few other languages? --Tobias1984 (talk) 00:08, 8 February 2014 (UTC)
I tried to do a Wikidata-styled logo that we could use on Wikipedia to link to this project. What do you think about it? The link could look like this: --Tobias1984 (talk) 12:40, 8 February 2014 (UTC)
Wikidata-logo-en.svg Visit the sister-project on Wikidata. Wikidata WikiProject Astronomy.svg

Symbol support vote.svg Support. For me it is a good idea! --Paperoastro (talk) 14:36, 8 February 2014 (UTC)

I planted the link on all the other projects. Hopefully some nice people will translate the English text. Even more importantly is that data coordination becomes more global. --Tobias1984 (talk) 16:10, 8 February 2014 (UTC)

Asteroids question[edit]

Hi! What do you think about this page: 102 Miriam. It's about an asteroid and many properties that don't yet exist are needed but I think it's well construct and can serve as a model for asteroids pages. What do you think? - Sarilho1 (talk) 12:59, 24 March 2013 (UTC)

You made a good job! It is a good idea to use also the properties preceded by and followed by. They will be very useful for Wikipedia templates. I add 102 Miriam as a good example of item. --Paperoastro (talk) 15:45, 24 March 2013 (UTC)

P31 and P60[edit]

Instead of using P31, shouldn't we use P60? If so, it's better change in the table. - Sarilho1 (talk) 13:04, 24 March 2013 (UTC)

Hi Sarilho1, welcome to astronomy task force! You are right, I made a mistake in compiling the table! I corrected it and add also the "main type (GND)" property. Thank you! --Paperoastro (talk) 15:28, 24 March 2013 (UTC)


Does this property include items on Earth? - Sarilho1 (talk) 19:42, 1 April 2013 (UTC)

Yes the idea is to use this property also for the item on Earth. --Paperoastro (talk) 13:15, 4 April 2013 (UTC)
Thanks! - Sarilho1 (talk) 10:18, 5 April 2013 (UTC)

Distance units[edit]

Hello everybody !

I'm a french contributor, and there is a lot of debate, and conflict, on WP:fr about units that shall be used for distance (parsec/ly). I'm rather agnostic on this matter, and I feel that a solution for us may come from Wikidata, and centralization of this debate.

I see that there is already a property defined for distance in Wikidata, in ly. I understand this choice is made for coherence with a template field (dist_ly).

Why not (again I am agnostic), but there are some arguments to have this property in parsec units. Mainly, one can imagine to automaticaly import this property (and others) from astronomical professional databases, in which this information is always in parsec units. By the way, there should be properties also for uncertainty distance margins, which are also in parsec in these databases. A conversion in ly may gives akward numbers, with should be rounded with a loss of precision in respect to the source.

What do you think of this problem ? Best regards --Jean-Christophe BENOIST (talk) 09:43, 4 April 2013 (UTC)

Hi, Jean-Christophe. Imho I would use parsec instead of ly, because parsec is directly derived from parallax, but this is only my opinion. In Wikidata I would use the measure unit of the source, as you told. Other Wikipedias use more than one unit: English and Italian use both parsec and ly (see for example en:Andromeda Galaxy and it:Galassia di Andromeda) and Spanish one ly and kilometres (see es:Galaxia de Andrómeda). Don't worry concerning the indication of the template here: it is only an example of infobox parameter: in Wikidata there is no restriction to use one measure unit or another. In fact the number datatype will has also the measure unit, allowing conversions:
  • with special functions of the number datatype (I hope it will be possible);
  • including different values in the property "distance" of items, one for every useful measure unit;
  • changing the measure unit in Wikipedia when the value is imported.
If you and other French contributors will want to discuss this topic here, you are welcome! --Paperoastro (talk) 13:19, 6 April 2013 (UTC)

Asteroids: bot at work[edit]

Hello, please note that I asked for a bot to populate the property discoverer for several thousand asteroids. Legobot is at work now on the basis of User:Ysogo/asteroidi-scopritori. It should complete in a couple of hours. --Ysogo (talk) 20:18, 5 April 2013 (UTC)

You made a very good job! :) --Paperoastro (talk) 13:27, 6 April 2013 (UTC)

Thanks, Papero. Please be informed that I have just asked for a bot to populate "preceded by" and "followed by" for all named asteroids up to asteroid #185641. Since I'm working with a derivation of the database that I use to create articles on about asteroids, I was able to manage only named asteroids. I hope in a near future to evolve my tool to also include not-yet-named asteroids that are anyway already present on wikidata. --Ysogo (talk) 20:39, 6 April 2013 (UTC)

Redirect all discussion pages of the task force here?[edit]

I wonder whether it is not better to put a redirect to this discussion page for the dicussion pages of all the four sections in which the task force is organized, I believe it would help a lot in having an homogenous view. If we do, it now it is quite easy since we only have one other discussion page with few topics discussed so far (and we can cut & paste here). Let me know your opinion. --Ysogo (talk) 21:01, 6 April 2013 (UTC)

Chemistry task force uses all the three discussion pages, Molecular Biology task force uses redirects to one page. Also for me it is better one page only: three pages is too much dispersive. --Paperoastro (talk) 22:34, 6 April 2013 (UTC)
Let's wait few days to see any other opinion. On Saturday we merge talk pages. --Ysogo (talk) 20:13, 8 April 2013 (UTC)
I agree. --Paperoastro (talk) 07:19, 9 April 2013 (UTC)
I don't see any impediment, so I agree. - Sarilho1 (talk) 10:56, 10 April 2013 (UTC)

Done --Ysogo (talk) 08:51, 13 April 2013 (UTC)

Asteroid's infobox[edit]

In Portuguese Wikipedia, it's used an infobox for asteroids. It is almost the same as the planet infobox. It is here in Wikidata - Q7486740 - and in Portuguese Wikipedia Info/Asteroide. (Note: Predefinição is Portuguese for Template). What should we do in these situations? - Sarilho1 (talk) 16:16, 7 April 2013 (UTC)

English infoboxes is a "point of reference" for proposing new properties. If there is an infobox parameter that hasn't correspondence with a template of English Wikipedia, we can propose a new property or use a qualifier (when they will be implemented) with an existent property. --Paperoastro (talk) 17:46, 7 April 2013 (UTC)
It's almost the same, but in Portuguese there is a parameter for number. I think that it is a copy from English Wikipedia's infobox about planets. I'm not really used to the phase II of Wikidata so I was thinking that all infobox about the same subjet would be merge and store in Wikidata. Now I'm understanding better. - Sarilho1 (talk) 18:03, 7 April 2013 (UTC)
In Wikidata we store and manage informations that every Wikipedias can use for their infoboxes. You must not change Portuguese infoboxes, but use useful (for your Wikipedia) values of Wikidata item to "populate" the parameter of your infoboxes. Informations are centralized in Wikidata, but Wikipedias can decide what of them use. --Paperoastro (talk) 19:22, 7 April 2013 (UTC)
Thanks! - Sarilho1 (talk) 09:12, 8 April 2013 (UTC)


When I have maintained and created articles about minor planets on svwp, I have tried to not add more exact numbers than the sources can provide.

As an example: 102 Miriam has very nice numbers in the Orbit elements, the "uncertainty" is low. In other objects, those numbers are not that exact, because their orbit isn't studied in detail, or the orbit is affected by larger objects to such extent that a nice looking ellips never is well-established, or the "uncertainty"-is missing in the sources. - I have tried (on svwp) to show the difference between objects with low uncertainty from the objects with larger or missing uncertainty, by writing more decimals and less zeros for 102 Miriam (3.33461926 AU 498,851,940 km), than for 5261 Eureka (1.62 AU 242,000,000 km) in the infobox. I have not added this "uncertainty" directly into the infobox, since I felt that such details, will not be understood by a common reader anyway. But they do understand (at least some of them) the value of "significant figures".

Therefor the number in the "Uncertainty"-column in the source above, would be nice to add next to the orbit-elements, once the orbit-elements are added to the items. I do not know yet how this can be done, but I guess it can be provided by a qualifier. --- Lavallen (block) 18:01, 14 April 2013 (UTC)

We are lucky: developers are developing numeric datatype including a value for "uncertainty" and a measure unit (see here). So we will insert a value with the error in Wikidata and choose, for example, to display in Wikipedia only the value. The error could be used to calculate the number of significant figures to display. --Paperoastro (talk) 11:04, 15 April 2013 (UTC)
Nice! I guess we are free to use this "uncertainty" in the way we like, or not at all. Observe that the "uncertainty" in this source do not give a ±-number, but a standard-deviation. -- Lavallen (block) 11:17, 15 April 2013 (UTC)
Good observation! Probably it is better ask to developers. --Paperoastro (talk) 07:50, 18 April 2013 (UTC)
This is one of the areas where I think the data model is rather insufficient, and even wrong in some respect (m:Wikidata/Data model#Numbers describes standard deviation and calls it variance). I think it works as a first iteration, but for any serious use it is partly not specific enough and partly not flexible enough. It has been said that qualifiers should be like a catch all when something can't be stored as a single value, or that you could add more specific properties, but it seems to me like that isn't always possible. Unstable orbits of celestial bodies are a good example, another is statistical analysis where some census bureau gives mean and other gives median for some number, a third could be numbers where there are imposed some limits due to privacy concerns. Jeblad (talk) 22:33, 18 April 2013 (UTC)
I think there is mainly four kinds of uncertain orbits here: 1. Not wellstudied. 2. Dynamic, like trojans and other objects who switch orbits, they cannot be described by a smooth elipse. 3. They are affected by the gravity of larger bodies, like Jupiter. 4. Comets who change mass during orbit. -- Lavallen (block) 14:27, 19 April 2013 (UTC)
Probably the ideal solution is have various numeric datatype that include different kind of errors. I don't know if this solution is feasible. Qualifiers that describe the type of error of values could be a more simple solution, for now. For the orbits, I have a particular idea: a property that describes the kind of uncertain that Lavallen described above. What do you think? --Paperoastro (talk) 15:32, 21 April 2013 (UTC)
Sounds good! -- Lavallen (block) 16:43, 22 April 2013 (UTC)
I need to change the data model. A more current version is here: Representing values, and shows that there is an "upper" and "lower" bound, and we do not fix what it exactly means. This would be up to the given community. Sometimes qualifiers might be needed as well. --Denny (talk) 12:23, 22 April 2013 (UTC)

Semi-major axis[edit]

Is the apsis-property supposed to be used for "Semi-major axis" too? Apiapsis and periapsis is not availible everywere, esp outside the solar system, and for many asteroid-moons. I can accept to use the apsis-property in all cases, with "Semi-major axis" as a qualifier. -- Lavallen (block) 11:26, 21 April 2013 (UTC)

We will use the apsis-property if we have values for it, both if we use it as "property", or if we use it as a "qualifier property". Semi-major axis is a length; apsides is a distance (the major and the minor from the body on the focus of the orbit), but the semi-major axis is on the line of apsides. Your proposal is good, but I'm afraid that it may create confusion with infoboxes of the other Wikipedias. --Paperoastro (talk) 15:49, 21 April 2013 (UTC)
I'm thinking that the space task force can also use some of these for man-made objects. en:Template:Infobox spacecraft and en:Template:Infobox space mission have orbital data - are we using the same words to mean the same thing? Secretlondon (talk) 16:29, 28 April 2013 (UTC)
Hi Secretlondon, if there is no opposition, for me is more simple that astronomical objects and spacecrafts uses the same property for orbits. --Paperoastro (talk) 21:14, 4 May 2013 (UTC)
If they are the same thing then it is obvious that we don't need special spacecraft orbit properties.Secretlondon (talk) 21:28, 4 May 2013 (UTC)
Yes, of course! --Paperoastro (talk) 21:36, 4 May 2013 (UTC)

Constraint violations of property P60 (type of astronomical object)[edit]

It is active an automatic check of wrong values of property. In particular, the property P60 (type of astronomical object) is checked for "single value" violation: here there is a list of items where P60 has more than one value. I'd like to discuss:

  1. the possibility to have a very small number of exceptions of this violation, e.g. Ceres is both dwarf planet and asteroid.
  2. most of exceptions listed are asteroids (see 11668 Balios as example). I propose to remove the value different by Q3863 (asteroid) and use for the "wrong value" the property P361 (part of). An example can be found here. Are you agree? --Paperoastro (talk) 09:53, 11 May 2013 (UTC)

I start a list in this page with the astronomical properties checked by KrBot. I invite everyone to propose/add other properties. --Paperoastro (talk) 10:24, 11 May 2013 (UTC)

Yes, trojans and others can use P361 instead, it's a good solution. There are also a group of objects who are both comets and different kinds of minor planets at the same time, but the number is low.
How well-defined is "Asteroid" as a definition? To me "Minor planet" feels much more well-defined. I've never heard anybody name a TNO as a Asteroid, but there are many objects who are both in the inner parts and the outer parts of the solarsystem in their orbits. -- Lavallen (block) 10:46, 11 May 2013 (UTC)
"Minor planet" could be the best solution for "ontology", but I prefer to distinguish between different types of minor planet objects (e.g. natural satellites, asteroids, comets,...), because P60 (type of astronomical objects) can be used by infoboxes of Wikipedia. Concerning TNO and asteroids: you are right. It is better to distinguish both of them. In general, we should open a discussion to decide which item use for P60 property. --Paperoastro (talk) 11:35, 12 May 2013 (UTC)

check P155 vs P156[edit]

I believe we can ask for a check for P155 vs P156 (preceeded by, followed by):

  • if Qa is P155 of Qb, then Qb must be P156 of Qa.

Papero, how can we ask for such check? It looks you are familiar with this procedure. --Ysogo (talk) 18:00, 13 May 2013 (UTC)

Here there are the instructions. ;-) However the symmetric violation check only the case "Qa is Pa of Qb" and "Qb is Pa of Qa" - e.g. the property P26 (spouse) (see here)! I ask here my doubt. --Paperoastro (talk) 20:42, 13 May 2013 (UTC)
Thanks to Ivan A. Krestinin, the reciprocal constraint between P155 and P156 is now active. :-) --Paperoastro (talk) 13:14, 19 May 2013 (UTC)
Great, but it looks like we had a good idea for other task forces. Exceptions include presidents and kings from around the world, books and music LP of different artists, Olympic Games... and so on. Any way to limit it to astronomical objects? Or better, to create separate lists for items relevant to different taskforces?--Ysogo (talk) 16:56, 19 May 2013 (UTC)
Probably the solution is to check the existence of property no label (P60) with another bot (imho using the same bot could be too complicated). --Paperoastro (talk) 18:15, 19 May 2013 (UTC)

Hierarchy of the property no label (P60)[edit]

We need to choose which values ​​should have the property no label (P60), that involves all the astronomical objects. In my sandbox I propose a hierarchy that can be useful to manage infoboxes of Wikipedias. I'd like to discuss:

  1. the items involved in this classification;
  2. the relationship between them;
  3. how to show this ontology (I used a table, but I'm sure that exists a better manner to show it);
  4. where to put all (in the discussion page of the property or in a sub-page of the Astronomy task force?).

Please, be bold. ;) --Paperoastro (talk) 14:38, 16 June 2013 (UTC)

Bot to scrape Extrasolar Planets Encyclopaedia[edit]

I was wondering if anyone would be able to create a bot that would be able to copy the information about planets detected by the Kepler spacecraft from the Extrasolar Planets Encyclopaedia (link) to our list of planets discovered using the Kepler spacecraft. Rather than merely going to the list, it would be ideal if the bot could follow the link for each Kepler planet and get the full information from there, rather than merely looking at the catalog. The information in the EPE about Kepler planet is in turn copied from the Kepler discoveries catalog, which is in the public domain but is unfortunately offline at the moment (requiring us to use the copyrighted EPE. In addition to the basic information, I would like it if our bot were able to calculate surface gravity where possible based on mass/(r^2). Thanks, Wer900talk 18:08, 27 June 2013 (UTC)

Hi, Wer900. Thank you very much for your proposal. We will import certainly the data of the Extrasolar Planets Encyclopaedia with a bot, but it is better to wait for new datatypes: ulr (for references) and quantity. According to the information of this roadmap, in September we will are able to import all the available data from the Encyclopedia. See you! --Paperoastro (talk) 16:40, 9 July 2013 (UTC)

Earth (Q2)[edit]

Why Earth (Q2) is a term (Q1969448) while Mars (Q111) is a geographical object (Q618123) using property no label (P107)? I would change it, but Earth (Q2) has sources so I don't know if I should. - Sarilho1 (talk) 14:58, 6 August 2013 (UTC)

GND classification is performed by the Deutsche National Bibliothek. Following their classification, every astronomical object is geographical object (Q618123), but they decided to classify Earth (Q2) as term (Q1969448)! I know, it is strange! --Paperoastro (talk) 12:10, 7 August 2013 (UTC)
Yeah, that was my fear. :) - Sarilho1 (talk) 13:03, 7 August 2013 (UTC)
Actually, it's both. I added also astronomical object as entity type.--Ysogo (talk) 06:31, 13 August 2013 (UTC)
With the decision to not use and delete no label (P107), the problem "is solved" ;-).

Property related to Minor planet designation[edit]

In Infobox planet there is a space (mp_name) about Minor planet designation (Q817125). For example, for 102 Miriam, these code is 102. I think that such property don't exist, but I think it could be useful. Should it be proposed? - Sarilho1 (talk) 15:12, 6 August 2013 (UTC)

Is it the same for you use the property catalog code (P528)? --Paperoastro (talk) 12:12, 7 August 2013 (UTC)
Yes, I think it is. I didn't know that these property worked for several catalogs. I have just a question about the property in general: Can you put these property (with the qualifier) in an infobox? I don't know how it works so... - Sarilho1 (talk) 13:03, 7 August 2013 (UTC)
I found this draft: it is outdated but the syntax should be valid. We need a qualifier property to distinguish the different catalogs. --Paperoastro (talk) 11:56, 13 August 2013 (UTC)
There is a lot of suggested properties here about catalogs. It's better to refuse before more properties being created. - Sarilho1 (talk) 21:41, 14 August 2013 (UTC)
Yes, I put my comment there, but the proponent return on the end of August. --Paperoastro (talk) 14:55, 18 August 2013 (UTC)

Deletion of no label (P107) and astronomical objects[edit]

We ever marked all astronomical object (Q6999) with the property no label (P107) => geographical object (Q618123), but now some people are claiming that these property will be deleted and we should use instance of (P31)/subclass of (P279) instead. So, I would like to ask, I should we mark astronomical object (Q6999) from now on? - Sarilho1 (talk) 14:32, 18 August 2013 (UTC)

I have just read the discussions. For now we have no label (P60) that distinguish astronomical objects from the other items. In the future, probably P60 could be substitute by instance of (P31) with something like the hierarchy that I proposed here that start from astronomical object (Q6999). --Paperoastro (talk) 14:52, 18 August 2013 (UTC)
I'm sorry, I've forgot the title. I agree with you, I don't like to use instance of (P31) because we can use any item. Maybe we should give priority to that table to, when reached consensus, turn it official. - Sarilho1 (talk) 15:30, 18 August 2013 (UTC)

Property "not quite" proposal[edit]

Items like USA-243 (Q13392599) should not only list what kind of orbit they're in but what they are orbiting around, for those kinds of orbits where the orbit type does not implicitly link the two.

To that effect, it seems to me that there should be a property "orbits", with the qualifier "orbit type" where it is not immediately obvious what object is being orbited (and, on that note, it seems to me that there are other things that "orbit" besides astronomical objects).

Thoughts? --Izno (talk) 23:26, 19 August 2013 (UTC)

Maybe this is a moot question, given astronomical body (P397), child astronomical body (P398), companion of (P399)? Mine seems much more elegant either way though, I suppose, because it can be used for more than just astro-objects... --Izno (talk) 23:40, 19 August 2013 (UTC)
The idea to use P397, P398, P399 is natural, and I support it. Of course, we should be carefully with constraints that are set for astronomical objects.
I take this discussion to speak about my doubt concerning P397 and P398. When I proposed these properties I did not know part of (P361). Probably, as suggested by some people, the good names for P397 and P398 could be "orbiting" and "orbited by" and for the other use (i.e. stars in clusters, galaxies in groups...) we could use P361. What do you think? --Paperoastro (talk) 12:37, 20 August 2013 (UTC)

Are there any edge cases where "companion of" is not covered by either "part of" or by "orbits/astronomical body"?

As for using "orbits" instead of "astronomical body", that would change the relation of the Solar System to the Milky Way to be part of, where the Solar System is orbiting the Milky Way core, correct? (Just as an example.) Is that okay?

As for the previous discussion, I think either "orbits" or "parent body" both face the issue of "which is the actual thing that is the parent body / being orbited"? But yes, I would agree that we could do what you suggest.

On a last note unrelated to previous comments, I don't see a need for the reciprocal property of "orbited by", per the general feeling that we should build the database up rather than down. --Izno (talk) 16:12, 20 August 2013 (UTC)

Thanks for your comments. All your example are correct. Probably we can leave all as is at now: it is more simple. Concerning the two reciprocal properties you have right: "child body" is redundant and it could be difficult to manage it for high number of "child objects". When we will have queries probably it will became redundant. --Paperoastro (talk) 20:57, 30 August 2013 (UTC)

Organization of the list of properties[edit]

When I created the Astronomy task force, I thought it was a good idea to group properties as they are organized now, but they are difficult to manage. What do you think of the organization of Physics task force? --Paperoastro (talk) 16:07, 24 November 2013 (UTC)

I update the table with the list of all properties involved in astronomical objects and also the table with the properties used for the "infobox planet". My idea is to update in the same manner the other templates of astronomical objects. --Paperoastro (talk) 22:55, 21 January 2014 (UTC)

Request of deletion of the property no label (P60)[edit]

Please, see the discussion for the deletion of the property no label (P60). --Paperoastro (talk) 16:35, 8 December 2013 (UTC)

The PfD of no label (P60) is closed and the property will be deleted. Before to delete it we need:

  1. warn Russian Wikipedia (Q206855) and wait for the modification of the template ru:Шаблон:Звезда;
  2. move all the statements of P60 to instance of (P31) (see this bot request);
  3. remove the constraint Item: items with this property should also have type of astronomical object from the properties site of astronomical discovery (P65), galaxy morphological type (P223), spectral class (P215), astronomic symbol image (P367), located on astronomical body (P376),astronomical body (P397), child astronomical body (P398), companion of (P399), provisional designation (P490), orbit diagram (P491), constellation (P59), catalog code (P528), minor planet group (P196), JPL Small-Body Database identifier (P716), asteroid spectral type (P720), asteroid family (P744), type of variable star (P881), asteroid taxonomy (P1016), discovery method (P1046), redshift (P1090), apparent magnitude (P1215), astronomical filter (P1227). ✓ Done --Paperoastro (talk) 20:05, 26 May 2014 (UTC)

--Paperoastro (talk) 12:24, 26 May 2014 (UTC)

The property can be deleted. --Paperoastro (talk) 16:34, 16 August 2014 (UTC)

A double ontology for astronomical objects[edit]

After a first discussion, I make a double proposal to organize astronomical objects: a "classical" ontology using instance of (P31) and subclass of (P279) and one using part of (P361) (the last suggested by Filceolaire and Sarilho1).

The ontologies, especially the second, are not complete. At now the situation is:

I ask the community suggestions to complete the ontologies and how manage them from modification without discussion.

Please, be bold if someone want to modify the two trees below.

--Paperoastro (talk) 23:29, 7 February 2014 (UTC)

Ontology with part of (P361)[edit]

Ontology with subclass of (P279)[edit]

All the items need P31 -> Q6999?

No, they are subclass of astronomical obects, we need an item like astronomical object type for instance of (P31). TomT0m (talk)


Nice! The two trees above are really in the same ontology -- the Wikidata ontology -- they are just different hierarchies. The part of (P361) hierarchy captures part-whole relations, and the subclass of (P279) hierarchy captures subsumption relations. We absolutely want both. The hierarchies above are a reasonable early iteration.

A note above the subclass hierarchy asks whether all subclasses of astronomical object need the claim 'instance of: astronomical object". For example, does planet (Q634) need to state both "subclass of planemo" and "instance of astronomical object"? I think the answer is "no". The claim "instance of astronomical object" should be applied only to particular astronomical objects, like Mars (Q111) or Chelyabinsk meteor (Q4661508). Applying "instance of astronomical object" to items like planet (Q634) would require awkward techniques for basic tasks like retrieving all things all astronomical object tokens, e.g. physical things beyond Earth that have a unique location in space and time. In other words, applying "instance of astronomical object" to things like "planet" would complicate queries to retrieve lists of particular things that could be, for example, visited by a space probe or observed with a telescope. Emw (talk) 19:23, 8 February 2014 (UTC)

Thank you very much for your comments Emw! I'm right with you to maintain both hierarchy ;-) . If I understood correctly, an item may not have P31 if there is P279 in a hierarchy, or is it better add as P31 another value as TomT0m suggested above? I have other two question to manage these hierarchies:
  1. astronomical object is a case, but we will manage other kinds of hierarchies (books, geographic places, events, fictional characters...). We need a standard place to put and maintain them. Does we need to open a general discussion?
  2. is it possible exclude some terms from constraints? For example, in general is too much generic define an object instance of (P31) -> nebula (Q42372), but it is better use the subclasses.
--Paperoastro (talk) 12:36, 12 February 2014 (UTC)
On point of your first question, an item can have P31 if it also has P279, but it's not clear to me what the best way we should go about doing such things. I think for now we should stay away from that unless there is an obvious instance of (P31) claim to make, such as at Udaloy-class destroyer (Q1052123), which uses already-natural language. --Izno (talk) 14:58, 12 February 2014 (UTC)
Good job Paperoastro, took me a while to understand your second question, but I finally got it. I agree we should have a way to express that a class is a little generic and that we should use one of its subclasses. I see several ways to do this
  • an analysis of the subclass tree which could warn than the class is not a leaf of the tree
  • using a class of nebula types : mark the different nebula subclasses that are actually used as instance of (P31) <type of Nebula> nebula (Q42372), create a property type of to make a claim <type of Nebula> type of Udaloy-class destroyer (Q1052123) , then we can have a constraint or a gadget which suggest the more relevant subtypes that exist … TomT0m (talk) 16:46, 12 February 2014 (UTC)
@TomT0m: your first proposal is more general and "clean" than the second one, but I don't know if it technically feasible. The second is simple to manage, and we can use an item as <type of astronomical object> (as suggested by Izno) to "mark" the usable items of the whole hierarchy. If there is consensus about it, it could be the a good solution. --Paperoastro (talk) 13:52, 13 February 2014 (UTC)

As a point not part of the above, we should double check the two ontologies against each other e.g. what are the part of claims for each of those subclasses? Do those part of claims make sense? --Izno (talk) 16:07, 12 February 2014 (UTC)

@Izno: good observation! I'm uncertain to use the same number of items in both of hierarchies or not. For now in the "part of" hierarchy I didn't put subclasses items of a certain element (e.g. I don't put subclasses of interstellar cloud (Q1054444) or star cluster (Q168845)), because, for me, it is redundant, but I don't know if this solution is "ontologically" correct.
--Paperoastro (talk) 13:52, 13 February 2014 (UTC)

A possible instrument to monitor hierarchies[edit]

Waiting for queries, I made a script in LUA to monitor hierarchies; below there is an example using the hierarchy of astronomical objects made with part of. I hope it could be useful. --Paperoastro (talk) 23:07, 19 February 2014 (UTC)

Item Property Expected value Found value

Lua error in Module:CheckValue at line 38: attempt to call field 'formatEntityId' (a nil value). Lua error in Module:CheckValue at line 38: attempt to call field 'formatEntityId' (a nil value). Lua error in Module:CheckValue at line 38: attempt to call field 'formatEntityId' (a nil value). Lua error in Module:CheckValue at line 38: attempt to call field 'formatEntityId' (a nil value).

Hi, in the same spirit it occured to me that module:Trees code can with moderate rewrite and eytend tocheck if there is a loop if a part of or a subclbss hierarchy, any interest here ? TomT0m (talk) 08:01, 20 February 2014 (UTC)

Hi, TomT0m, I gave a look to the module Tree and I saw that you used the second algorithm written here. If I understand correctly, you can use the first algorithm listed there, it is true?. --Paperoastro (talk) 12:55, 22 February 2014 (UTC)
Paperoastro Hi, actually I used a variant of the second algorithm which instead of a list to store the order assigns a number to a node such that if a node would be before another in the list the number is greater (or shorter if I'm wrong /o\).
I could for sure have used the first algorithm, but the second is based on a Depth-first search (Q816319) which can be used to detect loops as well (and I do here)
The things are a little bit complicated than that, as the algorithm I coded does several stuffs at the same time (and should be removed from the Tree module to be moved on a Graph one): It computes a Spanning tree (Q831672) of the input graphs, which is shown with the tree templates, with annotated nodes to add the missing edges as HTML anchors and links, detects the links who makes the input graph either a tree (Q223655), a directed acyclic graph (Q1195339) if there is no loop but a node have several path from the root to him, or a directed graph (Q1137726). The order is used in order it ensures that if the HTML internal link is one that would make the Spanning tree (Q831672) a general graph the link direction in the HTML page, the link should go up into the page, and if the link is one that would make the graph a directed acyclic graph (Q1195339) it should go down into the page (not sure that works though ...).
So actually this algorithm already detects loops, and DAGs if the graph is supposed to be a simple tree. I just have to add a DAG and a Graph template which respectively adds the categories if some DAG or a loop have been detected when the tree template have been called and so on. TomT0m (talk) 13:51, 22 February 2014 (UTC)
Hi, TomT0m, I apologize to answer you only now. I read and re-read your comment above, but for me now it is too hard learning the logic of the trees! So, I'm sorry, I cannot help you. --Paperoastro (talk) 14:45, 11 March 2014 (UTC)

New subpage for ontologies[edit]

Hi, I have created a new subpage of the WikiProject Astronomy for managing the two ontologies with subclass of (P279) and part of (P361). At now I added the two radix items and one element: supercluster (Q27521). Some problems:

  1. Is it correct the statement astronomical object (Q6999) <part of (P361)> astronomy (Q333) (a scientific branch)?
  2. Has sense the statement astronomical object (Q6999) <part of (P361)> universe (Q1)?
    Please discuss here.
  3. supercluster (Q27521) and other objects have the statement subclass of (P279) -> gravitationally bound system (Q15053549). Is it useful in Wikidata this subclass? In my opinion it is better classify supercluster (Q27521) and other astronomical objects (see tree) directly as astronomical object (Q6999). In addition, also star (Q523), planet (Q634), asteroid (Q3863) are, by definition, gravitationally bound bodies. It can generate confusion...

--Paperoastro (talk) 22:45, 31 May 2014 (UTC)

seems not correct. A science is a study and a collection of validates models and ideas, not an astronomical object. part of (P361) and
< X > has part (P527) miga < Y >
are more easily usable for concrete objects, such as
< door > part of (P361) miga < house >
I think we can view a science as a class of knowledge, then we could have subclass of (P279), with a property study of (I proposed but was not accepted, I think), or subject of we could make the link beetween the science and the object it studies.
For the third point, I don't know, seems to not do harm, it is probably sourcable and an interesting fact. Could be a Wikipedia article (or a part of it) ^^ TomT0m (talk) 09:12, 1 June 2014 (UTC)
Thanks for your comments, especially for the use of part of (P361). Concerning the third point, I prefer a "symmetric" hierarchy so I need an item for objects that are "not gravitationally bounded" or similar. --Paperoastro (talk) 20:38, 2 June 2014 (UTC)

At now (21 July 2014) the classification of astronomical object is more confused (see the tree) and it is made following different criterion: "type of objects" (i.e. stars, planets...), "gravity bounds" (i.e. gravitationally bound system), "wavelength" (i.e. astronomical radio sources, infrared sources, x-ray sources), evolutionary phases... In addition there are items that should not be in this classifications (e.g. part of HR diagrams, artificial satellites(!), lists). If there is no opposition, I will start to clean and simplify the classification. My idea is to use at the first step the "type of object" criterion (the more simple) and then we could decide how to proceed. --Paperoastro (talk) 20:52, 21 July 2014 (UTC)

A hierarchy for minor planet group (P196)[edit]

In the talk page of the property minor planet group (P196), I propose a hierarchy of the item involved. Please, comment and/or suggest modifications. --Paperoastro (talk) 21:54, 23 February 2014 (UTC)

A hierarchy for type of variable star (P881)[edit]

In the talk page of the property type of variable star (P881), I'm building a hierarchy for variable stars using the documentation of General Catalogue of Variable Stars (Q222662). Any kind of help and/or suggestion are welcomed. --Paperoastro (talk) 10:12, 10 March 2014 (UTC)

Get rid of those horrid specialized typing properties :) Each time we create one, Reasonator has to add code I guess. For nearly no advantage. TomT0m (talk) 19:21, 10 March 2014 (UTC)
A replacement : put a
< star class > instance of (P31) miga < variable star type >
on the classes used in that hierarchy. TomT0m (talk) 19:24, 10 March 2014 (UTC)
I'm following suggestions that Emw gave here in the discussion of the deletion of no label (P132): with few "specialized" properties like P881, we can avoid multiple classification with P31. --Paperoastro (talk) 11:00, 12 March 2014 (UTC)

apparent magnitude (P1215) and filter[edit]

I have just created the property apparent magnitude (P1215), one of the most important for astronomical objects. Before to ask a bot to add apparent magnitudes, I'd like to create the qualifier wavelength band (here the proposal). I'd like suggestion to hierarchy. For example:

B band (Q6746395) subclass of (P279) -> UBV photometric system (Q129833)

could be correct? --Paperoastro (talk) 23:24, 23 March 2014 (UTC)

planetoid (Q398938) (View with Reasonator) / planemo (Q400144) (View with Reasonator)[edit]

Hi, may someone (like Paperoastro for example) help me sorting out the differences beetween these two items ? the language links seems almost complementary.

Is there a reson why they are not merged ? I wanted to sort things a little following a Wikidata game question about no label (Q3844581) (View with Reasonator), a formation (cavus (Q358877) (View with Reasonator)) on Triton (Q3359) (View with Reasonator), which lead me to claim that
< cavus (Q358877) (View with Reasonator) > part of (P361) miga < planetoids / planemo / both ??? >
. TomT0m (talk) 18:39, 29 June 2014 (UTC)
Hi TomT0m, the nomenclature of sub-stellar objects is a little confusing. The two terms are not synonyms:
  • planemo (Q400144) was proposed as definition for objects massive enough to be rounded by its own gravity and not massive enough to cause thermonuclear fusion; so this definition includes planets, dwarf planets, and "spherical" satellites. The terms was proposed in 2003 (at now I do not reach to find a source!), but it is not so used in literature (see this link). IMHO I would not use this definition.
  • planetoid (Q398938) is an historical synonym of asteroids (so excludes planets and includes "not spherical bodies"): see the introduction of Minor planet (and reference therein).
For Triton (Q3359) and similar, I suggest natural satellite (Q2537). --Paperoastro (talk) 22:35, 29 June 2014 (UTC)
Planemo seems a good and useful definition though, works for me. Do you think "Cavus" applies only for planemo or also for ateroids ? TomT0m (talk) 06:58, 30 June 2014 (UTC)
At now in literature "Cavus" is used only for some geological formation on Mars and Triton, so you can apply this only for planemo. --Paperoastro (talk) 10:15, 30 June 2014 (UTC)

Conflicts between P60 and P31[edit]

SamoaBot has moved statements from the deprecated P60 to P31, but for 295 items there is some conflicts between statements already present in P31 and the ones in P60. Before to delete P60, we need to resolve them we need to copy the statement to P31 as is and, if we want, resolve the conflict. Follow the list.

--Paperoastro (talk) 11:48, 27 July 2014 (UTC)

That is not true. The claims can be copied to P31 and P60 can be deleted. The manual review is independent from that. Tamawashi (talk) 15:33, 27 July 2014 (UTC)
@Tamawashi: I modified the explanation following your suggestions. --Paperoastro (talk) 16:44, 27 July 2014 (UTC)
@Paperoastro: Grazie mille! Tamawashi (talk) 17:25, 27 July 2014 (UTC)

✓ Done --Paperoastro (talk) 16:32, 16 August 2014 (UTC)

apparent & absolute magnitude[edit]

We now have the property apparent magnitude (P1215) with dimensionless number as datatype, but the corresponding property 'absolute magnitude' is still pending, awaiting for number with dimension datatype. It's an aspect of how you look at it if magnitudes are an unit or an unitless number – but I think we should be consistent between both properties.

I suggest to also create absolute magnitude now, with dimensionless number as datatype. For the following reasons:

  • apparent magnitude (P1215) is already widely in use
  • It might take a long time until number with dimension is available
  • Magnitudes can't be written as a product of potentials of SI-Units, thus it is unknown if Wikidata will support it and it is questionable if they are real units.

When it once is possible to use magnitudes as units, we can discuss this again and eventually migrate with using a bot. -- MichaelSchoenitzer (talk) 11:23, 28 August 2014 (UTC)

you are right: for the same arguments, apparent magnitude (P1215) was created. absolute magnitude (P1457) is here! --Paperoastro (talk) 19:34, 28 August 2014 (UTC)

Launch of WikiProject Wikidata for research[edit]

Hi, this is to let you know that we've launched WikiProject Wikidata for research in order to stimulate a closer interaction between Wikidata and research, both on a technical and a community level. As a first activity, we are drafting a research proposal on the matter (cf. blog post). It would be great if you would see room for interaction! Thanks, --Daniel Mietchen (talk) 01:45, 9 December 2014 (UTC)

Adding NASA Exoplanet Archive as a Reference to the Planetbox template[edit]

Hello everyone! My name is Marcy and I work for the NASA Exoplanet Archive. The archive offers a good amount of data on confirmed planets and we would like Wikipedia to link to us as a reference where appropriate, specifically in the Planetbox template. Currently there are links to SIMBAD and in that template, but it's not clear how we can add ourselves (or through which channels we have to work through to be added). The Reference "data" link in the Planetbox template would go to the planet's Overview page in the Exoplanet Archive, which is a summarized compilation of the confirmed planet's aliases, planetary and stellar parameters, and links to published literature.

For example:

1. Go to the Wikipedia article for Kepler-22 b.

2. On the right side of the screen there is an infobox. At the bottom of the info box is the References section listing SIMBAD and the Exoplanets.EU, each with a link to the respective archive's data on that object.

3. We would like to add an entry for NASA Exoplanet Archive with the data link going to the archive's overview page; in this example it would be

Please let me know if you have any questions, and how we should proceed. Thank you!

P31 for Minor planets[edit]

Minor planets today often have the statement instance of (P31) asteroid (Q3863) no matter where in the solar system the minor planets can be found. That looks wrong to me, but what are we going to change them to? One alternative would be to change all P31 asteroid (Q3863) to minor planet (Q1022867), but I am not sure that would be a very stable solution with so many potential editors.

We have minor planet group (P196) to specify what groups they fit, and another property for families, so we do not have to be very specific. I suggest instance of (P31) distant minor planet (Q5282923) for all TNO, Centaurs and Neptune/Uranus Trojans (and Saturn Trojans if/when we ever find any). My main objection is that "distant minor planet" does not look very widely used. And I do not know if the Jupiter Trojans can be called distant or not. There will always be object which fits both as asteroids and as distant minors, like Hidalgo, but that problem will always be there somewhere, even if we use "minor planet" for all objects. -- Innocent bystander (talk) 08:50, 1 March 2015 (UTC)

You can use multiple instance of (P31) statements if you want. So if a particular object fits under two branches of the "astronomical object" class hierarchy, put it in two. (Mind you, my understanding of an asteroid is something a fair bit different from that of a minor planet, so it sounds more like there are a certain set of minor planets that either fit in both classifications, or fit in one or the other, and if they fit in one or other the statement in question should be refined/crosschecked with reliable sources.) --Izno (talk) 17:24, 25 March 2015 (UTC)
The statements are today mainly based on Wikipedia articles, or more accurately the botowners interpretations of the use of categories and templates in Wikipedia. I guess there can also be some inconsistencies between how the word "asteroid" and "minor planet" can be used in different languages. -- Innocent bystander (talk) 08:43, 28 March 2015 (UTC)


Hi, I'm new to this. I was interested in adding data about solar and lunar eclipses. However it seems there are not many relevant data properties. My efforts so far can be seen at Q6312134. Any advice would be appreciated. I was dismayed that I couldn't even enter times properly ... ? Ideally all the information at en:List of 21st-century lunar eclipses would be present. Thanks MSGJ (talk) 14:49, 25 March 2015 (UTC)

Did you check Wikidata:List of properties/Geographical_feature#Astronomy and Wikidata:List of properties/Generic for fit for use? Also, after checking, it might help if you could explicitly identify which of the table header items at the list you think should be added. Just from taking a look at your start, I might suggest that we need to look into explicit properties for those images. --Izno (talk) 17:49, 25 March 2015 (UTC)
I did check all those, and no others seem to be applicable. Even the times have now been removed from Q6312134 so this looks like a non-starter at the moment. Shall I propose all the necessary properties? MSGJ (talk) 12:23, 26 March 2015 (UTC)
Might as well. Did you check any of the others? (As a side note, total penumbral lunar eclipse (Q7828129) exists for marginally more detailed 'type' statement.) --Izno (talk) 02:29, 28 March 2015 (UTC)
Well, I proposed them a week ago and not a single person has commented yet MSGJ (talk) 22:33, 7 April 2015 (UTC)
We move slowly here sometimes (too quickly in others). You can try pinging some of the same people as have already made a contribution to another proposal above. Alternatively, you might do some en:WP:canvassing on en.wp. --Izno (talk) 13:49, 10 April 2015 (UTC)

Mapping of property based on list of observatory codes[edit]

In the past days I have been adding all observatory codes to Wikidata items. The usage of these codes is mapped on User:Romaine/Mapping/List of observatory codes based on the primary source of List Of Observatory Codes. This makes it possible to search easily for the item belonging to a code. Romaine (talk) 07:08, 27 March 2015 (UTC)

A related question is what we should do with the robot-discoverers? I have started to add missing discoverer or inventor (P61) to items related to these lists. Often the site of discovery also is the discoverer. Should we use the same item for the discoverer as for the observatory? In 1802 Zhang Heng (Q143987) Purple Mountain Observatory (Q714352) have been added to both P65 and P61, but I am not fully sure that is a good idea. Should we instead create a separate item for "the robot-observatory of Purple Mountain Observatory"? -- Innocent bystander (talk) 10:58, 28 March 2015 (UTC)

Use of Unicode character (P487)[edit]


On Wikidata:WikiProject Astronomy/Properties, Unicode character (P487) is listed to be add on astronomical objects which is a good idea but it doesn't seems to be the rightest way to me as it is redundant.

I suggest to use only notation (P913) to link an astronomical object to the item on its astronomical symbol (notation (P913) is already use but not documented on Wikidata:WikiProject Astronomy/Properties, is it on purpose or just neglected?) and to use Unicode character (P487) only on this item on its astronomical symbol.

Example: Mercury (Q308) -> notation (P913) = Mercury symbol (Q3594850) and Mercury symbol (Q3594850) -> Unicode character (P487) = "☿".

Doesn't it seems better to you? Can I remove Unicode character (P487) and replace it by notation (P913) on Wikidata:WikiProject Astronomy/Properties? (and add a note to explain that Unicode character (P487) can be use on the symbol item)

Cdlt, VIGNERON (talk) 17:25, 27 July 2015 (UTC)

  • I believe your suggestion is a better way to handle this. Artem Korzhimanov (talk) 12:53, 28 July 2015 (UTC)
But I am not aware of that we have any items for the Vesta-symbol etc yet. -- Innocent bystander (talk) 14:02, 28 July 2015 (UTC)
Thank you for your answers Artem Korzhimanov and Innocent bystander.
Innocent bystander : 4 Vesta (Q3030) symbol has no item right now but I think there is no problem to create this item (and since this symbol has an history and multiple variations, a article on Wikipedia could be useful too).
Cdlt, VIGNERON (talk) 09:55, 29 July 2015 (UTC)
As far as I know, the number of asteroids with own "symbols" is limited and only a few of them have a Unicode-version of the file. So it is fully possible to do even with a limited effort. -- Innocent bystander (talk) 10:01, 29 July 2015 (UTC)
Yes it is. Besides asteroids, there is planets and the 12/13 zodiacal constellations but it's still a limited number of items (30 at the most). Cdlt, VIGNERON (talk) 10:23, 29 July 2015 (UTC)
FYI, 1. I removed Unicode character (P487) from planets and constellations. I'll deal with the minor bodies later. 2. I change the documentation consequently : Special:Diff/247277729. Cdlt, VIGNERON (talk) 13:59, 2 September 2015 (UTC)

An other related question : should astronomic symbol image (P367) (created in 2013 but currently unused) goes on the object or on the symbol? The constraint says on the object but to be logical with this proposal, I think it should go on the symbol. Ping @Paperoastro, Ricordisamoa: for comments. Cdlt, VIGNERON (talk) 07:56, 30 July 2015 (UTC)

P367 has mediafile as type. Well, it looks like it is used to me. It is probably possible to migrate this to the symbol-item, and maybe even replace it with P18? If I remember correctly Moon (Q405) was once flooded with files. It is maybe better to create one item for each phase (full, new, etc)? -- Innocent bystander (talk) 08:26, 30 July 2015 (UTC)
Innocent bystander :ooups, sorry! I looked at Special:WhatLinksHere/Property talk:P367 instead of Special:WhatLinksHere/Property:P367
Using astronomic symbol image on an astronomic symbol may be a bit redundant and indeed maybe we should delete delete this property and use P18 on astronomic symbol instead.
Cdlt, VIGNERON (talk) 08:00, 31 July 2015 (UTC)
P367 was meant for astronomical objects at the time, but its use cases may have changed meanwhile, or it may have become completely useless. For what it's worth, I believe Commons files should be linked to Wikidata items, not vice versa. --Ricordisamoa 08:44, 30 July 2015 (UTC)
@Ricordisamoa: It looks like that "Commons files should be linked to Wikidata items" will not happen, since the file-namespace will not be included into Wikidata. Maybe other solutions comes with new technology, but that is to far far away to be seen today. On the other hand, there are discussions that P18 normally only should have one statement per item. The idea behind that thought is difficult to accomplish with the "Commons files should be linked to Wikidata items"-solution. -- Innocent bystander (talk) 16:14, 31 July 2015 (UTC)
@Innocent bystander: Sorry for being unclear, I meant: "Commons files should state which item(s) they depict, Wikidata items shouldn't care about which file(s) they're depicted by". --Ricordisamoa 16:39, 31 July 2015 (UTC)
That was exactly how I read you! :) Not a bad idea, but it is a little to far far away today. -- Innocent bystander (talk) 16:49, 31 July 2015 (UTC)
When structured datas for commons will be up this wont be a problem. Amyway a few images for an item should not harm, after all how it lioks likr is a property of the entity. Not very structured, but ... author  TomT0m / talk page 16:46, 31 July 2015 (UTC)

Celestial coordinates[edit]

Here is a reminder of my recent question elsewhere left unanswered about this matter. I wanted to introduce the celestial coordinates of Hubble Deep field (Q741255) as you show the geoloc of a city or place on earth. This property seems not to exist ( ? ) I trust it would be much more accurate (scientific) than using a (property)-constellation. Is this the right place where I should ask  ? Thanks in advance. Cordialement, et Hop ! Kikuyu3 (talk) 17:43, 6 August 2015 (UTC) PS : I will copy this request on Wikidata:Property proposal/Natural science, too.

@Kikuyu3: Probably something dumb as I don't know the field, but would universe (Q1) would work as a reference "globe" ? author  TomT0m / talk page 17:49, 6 August 2015 (UTC)
No, but a good idea. --Izno (talk) 21:22, 6 August 2015 (UTC)
We can't do this yet because we don't have the number with units property type available quite yet. It's coming very soon! --Izno (talk) 21:22, 6 August 2015 (UTC)

Quantity properties[edit]

A lot of new quantity properties need attention and inclusion on this Wikiproject Page. Please also check on Wikiproject Physics and Chemistry for properties that are useful to both projects. orbital inclination (P2045), M sin i (P2051) and luminosity (P2060) were recently created. --Tobias1984 (talk) 11:57, 13 September 2015 (UTC)

Since we do not have a solution to the epoch-problem yet, many proposed properties are still stalled. -- Innocent bystander (talk) 13:02, 13 September 2015 (UTC)