Select Page

Template talk:Coord: Difference between revisions

Browse history interactively
← Previous editNext edit →
Content deleted Content added
VisualWikitext
Line 82: Line 82:
{{od|8}}{{ping|EncMstr}}, do you mean N/S and E/W must be used with DMS notation '''in the Coord template''' or do you mean in general? Because if you meant in general, you would be incorrect. For example, page 144 of the ''Explanatory Supplement to the Astronomical Almanac'' 3rd ed. (Mill Valley CA: University Science Books, 2013) gives many negative latitudes in the DMS format. [[User:Jc3s5h|Jc3s5h]] ([[User talk:Jc3s5h|talk]]) 15:20, 28 April 2015 (UTC)
{{od|8}}{{ping|EncMstr}}, do you mean N/S and E/W must be used with DMS notation '''in the Coord template''' or do you mean in general? Because if you meant in general, you would be incorrect. For example, page 144 of the ''Explanatory Supplement to the Astronomical Almanac'' 3rd ed. (Mill Valley CA: University Science Books, 2013) gives many negative latitudes in the DMS format. [[User:Jc3s5h|Jc3s5h]] ([[User talk:Jc3s5h|talk]]) 15:20, 28 April 2015 (UTC)
:Sorry, I thought the context was clear: we are on the talk page for {{tl|coord}} so I was talking about this template.
:Sorry, I thought the context was clear: we are on the talk page for {{tl|coord}} so I was talking about this template.
:OTOH, mathematically, what would -11° 22' mean? Shouldn't it be -11° -22'?
:OTOH, mathematically, what would -11° 22' mean? Shouldn't it be -11° -22'? —[[user:EncMstr|EncMstr]] ([[user talk:EncMstr|talk]]) 16:00, 28 April 2015 (UTC)


== permpoly ==
== permpoly ==

Revision as of 16:00, 28 April 2015

Microformats geo tagging: missing latitude & longitude?

In Template:Coord/link, shouldn't there be <span class="latitude"> and <span class="longitude"> wraps inside the <span class="geo">? I was testing a page with coordinates inside an Infobox Organization (hcard), and the Google Structured Data Testing Tool was complaining about missing parameters. Adding those (off-Wikipedia) fixed the problem.

Example: http://www.google.com/webmasters/tools/richsnippets?q=https%3A%2F%2Fen.wikipedia.org%2Fwiki%2FHockley_Highlands_Inn AndroidCat (talk) 16:59, 4 February 2015 (UTC)[reply]

The Bing tool seems smarter, or at least more flexible, and does pick up the coordinates. https://www.bing.com/webmaster/diagnostics/markup/validator?rflid=1 AndroidCat (talk) 17:59, 4 February 2015 (UTC)[reply]
No; see Geo (microformat)#One class. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 09:12, 26 March 2015 (UTC)[reply]
The aim of adding explicit markup is to remove the need for being smarter and needing to use heuristics to extract data. Pigsonthewing, I'm not sure this should be dismissed quite that quickly. If there is a structure containing two members, and a method to fully tag that structures and two members then it would make sense to do that. It would also be consistent with that is already generated in other situations. —Sladen (talk) 14:05, 26 March 2015 (UTC)[reply]
You're welcome to ask the microfrmats commuinity to change the spec. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 16:08, 26 March 2015 (UTC)[reply]
Pigsonthewing: The versions of the geo[1] and h-geo[2] specifications I'm looking at appear to expect <div class="geo"><span class="latitude" value="N.N">N.N</span> <span class="longitude" value="N.N">N.N</span></div>. I can't find a mention in the alleged citation references for any "single class" syntax being valid; so on that basis I've also removed[3] the section to avoid further confusion. At this stage it appears to make considerably more sense to bring the generated output inline with the specification, than attempt to bend the specification to match Wikipedia's broken mark-up in one particular instance where it the output is inconsistent! Would you be help to help with this, so that we can hopefully get AndroidCat to test and confirm the fix? —Sladen (talk) 03:55, 27 March 2015 (UTC)[reply]
Google's tool has changed since February. I have images of the results from a Wikipedia page and a related page on my own Mediawiki with a patch to Template:Coord/link. I'll have to nowiki the urls to mine due to Wikipedia's blocking rules. All differences aside, the diff seems to be the key to Google's tool seeing the geo coordinates or not.
Page: Hockley Highlands Inn Result: [4]
Page: http://umbraxenu.no-ip.biz/mediawiki/index.php/AOSH_Canada [5]
Here is the diff of the patch I made: http://umbraxenu.no-ip.biz/mediawiki/index.php?title=Template%3ACoord%2Flink&diff=41183&oldid=38377 It's late. I'll have another look after sleep and two coffees. AndroidCat (talk) 06:17, 27 March 2015 (UTC)[reply]
@Sladen: Fair enough. It seems to have changed, since I worked on it, and wrote the article back in 2007. What help do you need? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 09:45, 27 March 2015 (UTC)[reply]
Nothing. I was reporting a peculiarity I noticed when I borrowed some templates. I'm happy with the patch on my wiki. :) (If it's only needed for Google's tool and redundant elsewhere isn't a problem for me.) AndroidCat (talk) 14:35, 27 March 2015 (UTC)[reply]

WikiData

There is a vague indication in the documentation that Coord can get data from WikiData, but there is no indication about how this works. It is also a concern; since WikiData is a wiki, it is not a reliable source. Jc3s5h (talk) 13:06, 25 March 2015 (UTC)[reply]

for example, go to d:Q316080 you will see 'coordinate location' which is used by the infobox in Madridejos, Cebu to generate the coordinates. I don't see how this is more or less concerning than the other method, which would be to use the coordinates specified in the article, since Wikipedia is also not a reliable source. Frietjes (talk) 21:42, 25 March 2015 (UTC)[reply]
The use of WikiData coordinates in "Infobox settlement" is a topic for a different talk page. "Madridejos, Cebu" only has one instance of the "Coord", which supplies the coordinates within the call to the template.
Your comment suggests a fundamental misunderstanding of the Verifiability policy and the Identifying reliable sources guideline. To minimise the spread of false information, the policy is that while each Wikipedia article is unreliable, it is contrary to policy to pass unreliable information from article to article by citing unreliable sources. If an editor were to use the "Coord" template to import data from WikiData (if that is even possible, which has not been demonstrated) and then provide a citation to a reliable source, the well-supported value could be changed to an incorrect value when WikiData changes, but there would be no edit to the Wikipedia article to alert editors that something had been changed. Jc3s5h (talk) 23:31, 25 March 2015 (UTC)[reply]
oh, I see this is a fork of another thread. I was under the impression you were interested in understanding how the WikiData feature works in Template:coord, but apparently not, since you immediately dismissed my comments as off topic (hint, there are two sets of coordinates in the article). Frietjes (talk) 23:55, 25 March 2015 (UTC)[reply]
I can only find one direct invocation of the Coord template:

{{coord|11|18|08|N|123|43|45|E|display=inline|region:PH-CEB_type:landmark}}

Jc3s5h (talk) 00:10, 26 March 2015 (UTC)[reply]
where do you think it gets the coordinates in the infobox? what would happen if I added a citation for those coordinates using |coordinates_footnotes=? Frietjes (talk) 00:14, 26 March 2015 (UTC)[reply]
Every template has its own syntax. I am asking if there is a syntax available within the Coord template to bring in coordinates from Wikidata. Hinting that the coordinates in the infobox came from Wikidata doesn't help me know what the syntax would be in Coord. Jc3s5h (talk) 00:22, 26 March 2015 (UTC)[reply]
Not really. The intention, I believe, is that if you use {{coord}} with no parameters it will report the WikiData value associated with the page (if any), but this doesn't currently work properly, because the #coordinates parser function will report an error. However, one can use {{#invoke:Coordinates|coord}} with no parameters to get the WikiData value. Obviously the infobox and the other tools have ways of accessing the WikiData values. Dragons flight (talk) 00:57, 26 March 2015 (UTC)[reply]
FWIW, I developed {{ParseWDCoords}} which parses {{#property:P625}} into 8 separate elements. This was superseded by Frietjes' work. --Unbuttered parsnip (talk) mytime= Thu 17:46, wikitime= 09:46, 26 March 2015 (UTC)[reply]

Need help on Gravine Island

I'm obviously missing something. I took the coords from Google Earth right in the middle of the island. The push-pin map required some tweaking to get it close but the coords template is many miles off. What am I doing wrong? I'm not asking anyone to fix it, please show my error as I have several articles to create which will require proper coords. Thanks! JodyB talk 18:36, 27 March 2015 (UTC)[reply]

I don't know where you got your numbers but in the map location template you have:
| lat_deg    = 30.47
| lat_min    = 18
| lat_sec    = 99
| lat_dir    = N
| lon_deg    = 87.25
| lon_min    = 43
| lon_sec    = 35
| lon_dir    = W
This looks like you've taken 30.471899°N 87.254335°W and attempted to make it be 30.47°18'99"N 87.25°43'35"W which is invalid (there are only 60 seconds for example).
I think the proper coord template should look like this:
{{coord|30|47|13.7|N|87|55|45.0|W|display=title}}
Values from this template should also go in the map location template.
Or, if your source lat/long are decimal:
{{coord|DD.ddddd|N|DD.ddddd|W|display=title}}
These values can also go in map location |lat_deg=DD.ddddd and |lon_deg=DD.ddddd with the matching minutes and seconds parameter left blank.
—Trappist the monk (talk) 19:02, 27 March 2015 (UTC)[reply]
Thanks! I switched to decimal degrees and it worked perfectly. DMS was obviously in error but now it is spot on. I appreciate the quick reply. 19:23, 27 March 2015 (UTC)
Also, consider the appropriate precision. The island is about 2 miles long. Using the rule of thumb that a minute of latitude is about a nautical mile from north to south, that suggests your precision should be better than minutes. Seconds would probably be appropriate, but fractions of a second are not appropriate. Or for decimal degrees, no more than four digits to the right of the decimal point. Jc3s5h (talk) 19:32, 27 March 2015 (UTC)[reply]
@JodyB: Precision is covered at WP:OPCOORD. As regards your original values, two issues should be pointed out: first, don't use decimals except on the least-significant value - that is, if you give degrees, minutes and seconds, only the seconds can have a decimal; if you give degrees and minutes but no seconds, only the minutes can have a decimal; and if you give degrees only, that may have a decimal. Second, the valid values for minutes and seconds of angle are the same as for minutes and seconds of time: 0-59, so 99 is well outside valid range. --Redrose64 (talk) 20:27, 27 March 2015 (UTC)[reply]
@Redrose64: @Jc3s5h: Thanks guys. Is there a general preference for DMS vs. decimal degrees? JodyB talk 20:37, 27 March 2015 (UTC)[reply]
Not as far as I know. Some infoboxes only accept decimal degrees, and in such cases the {{decdeg}} template may be used to convert DMS to decimal - I used it in this edit. --Redrose64 (talk) 20:42, 27 March 2015 (UTC)[reply]

My understanding, although I've not looked deeply because I hate it, is that decimal degrees don't have directions: south and west are both negative. Although Wikipedia and GPS and similar e-mumbo jumbo are happy with decimals, I most certainly am not, being a person. Incidentally even though Wikidata holds coordinates as decimals, it emits them as dms. --Unbuttered parsnip (talk) mytime= Sat 07:33, wikitime= 23:33, 27 March 2015 (UTC)[reply]

I don't think anyone is in charge of what latitude and longitude notations are acceptable. I have never heard that cardinal directions could be used with degrees-minutes-seconds but not with decimal degrees. A counterexample to this idea is that if you enter "43.2°N73.1°W" in the Google maps search box, you get to the right spot on the map. Jc3s5h (talk) 23:49, 27 March 2015 (UTC)[reply]
@JodyB: There was an RFC a few months ago in which many agreed that decimal notation is preferable and many others agreed that neither decimal or DMS was preferable. —EncMstr (talk) 03:51, 28 April 2015 (UTC)[reply]
@Jc3s5h: With DMS notation, N/S and E/W must be used. With decimal coordinates, N/S and E/W are optional; if omitted, + is used for N and E and - for S and W (just like in Cartesian coordinates). —EncMstr (talk) 03:51, 28 April 2015 (UTC)[reply]

@EncMstr:, do you mean N/S and E/W must be used with DMS notation in the Coord template or do you mean in general? Because if you meant in general, you would be incorrect. For example, page 144 of the Explanatory Supplement to the Astronomical Almanac 3rd ed. (Mill Valley CA: University Science Books, 2013) gives many negative latitudes in the DMS format. Jc3s5h (talk) 15:20, 28 April 2015 (UTC)[reply]

Sorry, I thought the context was clear: we are on the talk page for {{coord}} so I was talking about this template.
OTOH, mathematically, what would -11° 22' mean? Shouldn't it be -11° -22'? —EncMstr (talk) 16:00, 28 April 2015 (UTC)[reply]

permpoly

I am adding geographical coordinates to the article "Peter F. Armistead, Sr., House". Wikimapia includes an entry for this landmark ([6]), and the Google Maps satellite image that it displays nicely highlights the entire area covered by this landmark.

When I look at the source code for this wikimapia page, I see "permpoly=31648770", which I assume creates the highlight.

How can I include the permpoly value when I'm adding the Coord template to the article? Knife-in-the-drawer (talk) 03:13, 12 April 2015 (UTC)[reply]

Can we make a similar template for paleocoordinates?

As we all know, Earth's continents have moved over time. In light of this, scientists have worked to determine the historical locations of different geologic formations. I was wondering if a template could be created for those articles to display their "Approximate paleo-coordinates"? I think it would be really interesting and informative. Abyssal (talk) 12:44, 20 April 2015 (UTC)[reply]

Keep the no original research policy in mind. First we would have to identify a well-accepted coordinate system for paleocoordinates; we shouldn't make up our own. Jc3s5h (talk) 13:54, 20 April 2015 (UTC)[reply]
The Paleobiology Database maintains a large database of paleocoordinates for fossil sites, geologic formations, etc. Abyssal (talk) 16:25, 20 April 2015 (UTC)[reply]
That site does not have any easy-to-find information about what geocoordinates they use. But as best I can guess from making a few maps, they provide the modern coordinates of where fossils were found, not what the coordinates would have been at the time the life forms were alive. Jc3s5h (talk) 16:46, 20 April 2015 (UTC)[reply]
Here's an example of a fossil collection site with paleocoordinates. Abyssal (talk) 18:11, 20 April 2015 (UTC)[reply]
Have a look at a (US) National Geodetic Survey (NGS) datasheet for a monument at the top of a mountain. You will notice that there is a great deal of text describing several coordinate systems used to provide the location of the monument. You might think this amount of detail is more than is necessary for the location of a fossil; the description you referred to only locates the fossil to the nearest tenth of a degree, which could be as much as 11 km. This is much looser than the 1 cm or so the NGS achieved. But on the other hand, the basic assumptions underlying modern coordinate systems become invalid, even nonsensical, when extended back billions of years. So something more than "paleocoordinates" is required to describe what system is being used. I would guess that several different systems exist, depending on what period the life forms were alive. Jc3s5h (talk) 19:29, 20 April 2015 (UTC)[reply]
I'm just talking latitude and longitude coordinates. That page you linked to has all kinds of elevation information and historical estimates that aren't relevant. We only need current estimates of paleolatitude and paleolongitude. I don't think we need paleocoordinates to be more precise than those in the Paleobiology Database given that they probably can't even be known more precisely than that anyway given the margin of error involved. Abyssal (talk) 02:11, 21 April 2015 (UTC)[reply]
For latitude and longitude you need reference points. I suppose paleoposition of north and south pole can be used for latitude, but for longitude you still need a conventional origin, as Greenwich - whose position in old times is uncertain at best. If there were an accepted system to define paleocoordinates we could start using it, but first we should tell readers what we mean with such longitude and latitude. A good starting point would be writing a well sourced article on paleocoordinates.--Pere prlpz (talk) 08:45, 21 April 2015 (UTC)[reply]
+1. What Pere prlpz said. --Redrose64 (talk) 09:49, 21 April 2015 (UTC)[reply]
Paleocoordinates do use Greenwich as far as I know. Using a pole wouldn't make any sense since earth reverses its magnetic field every couple of thousand years, so it doesn't actually provide an objective measurement. There is already some discussion of paleocoordinates in Paleomegnetism and Apparent polar wander. Abyssal (talk) 12:02, 21 April 2015 (UTC)[reply]
Greenwich is on the London Clay, all of which was laid down in the Paleogene, so the position of "Greenwich" is meaningless for anything from the Cretaceous or earlier. Poles for measuring latitude should be taken as the ends of the rotational axis, which is independent of the magnetic field. --Redrose64 (talk) 13:00, 21 April 2015 (UTC)[reply]
If you have issues with the philosophy or methodology of mainstream paleogeography this proposal is not the place to voice those concerns. Abyssal (talk) 13:46, 21 April 2015 (UTC)[reply]

@Abyssal:, you seem to have missed the point of the datasheet I referenced above. Each coordinate system on that sheet has a name, together with all necessary information to identify which implementation of the system is being used. For example, the most up-to-date latitude and longitude coordinates on that sheet are in NAD 83(2011) which is the 2011 implementation of the North American Datum of 1983 which is described at http://geodesy.noaa.gov/CORS/coords.shtml.

If you want your proposal to go anywhere you must give us the official name of the paleocoordinate system(s) you want to use and references to the publication(s) that define it. You have not provided any information about "the philosophy or methodology of mainstream paleogeography". So in essence you are hoping someone else knows about how mainstream paleogeography assigns paleocoordiantes to a place and can write directions that would allow Wikipedia editors to put those coordinates in a template with the correct identification of the system used. But there doesn't seem to be anyone involved in the discussion that knows how to do that. Jc3s5h (talk) 14:19, 21 April 2015 (UTC)[reply]

Use page status indicator for coord?

Just throwing a ball... The new page status indicator system was designed to allow limited elements (like top icons) outside content space without having to resort to CSS hacks such as absolute positioning. Top icons and coordinate templates were the main target. All top icons have been converted, but coord still uses skin-dependent absolute positioning. How would you feel about utilizing the new indicator system instead? -- [[User:Edokter]] {{talk}} 11:00, 24 April 2015 (UTC)[reply]

^.^b. –Be..anyone (talk) 12:20, 24 April 2015 (UTC)[reply]
I don't understand the details of how this would work, but I consider it important in quite a few articles to provide a citation to the source of the coordinate information. Anything that inhibits that would be a bad idea. Jc3s5h (talk) 14:32, 24 April 2015 (UTC)[reply]
As I understand it, we would fill in the {{coord}} parameters exactly as we do now. The only difference is in what the underlying module then does with the coordinate data - instead of it displaying at a fixed position on the screen, it would be passed into the new indicator system (introduced 29 October 2014 with MediaWiki 1.25wmf5, see phab:T25796). That would construct the HTML/CSS/JavaScript that will display the coords at a position which should keep it clear of the FA/GA/protection icons etc. that also sit at upper right. --Redrose64 (talk) 15:58, 24 April 2015 (UTC)[reply]
What Redrose said. It does mean that (in Vector for example), the coordinates would display next to the page title, instead of below the title, thus sharing the space (but not overlapping) with the other icons, such as the protection and featured article icons. It does require that coords should be shorter, ie. by removing the "Coordinates:" link. -- [[User:Edokter]] {{talk}} 09:59, 26 April 2015 (UTC)[reply]
You can see it in action on my user pages, excl. global user pages, where page status indicators do not work. –Be..anyone (talk) 01:56, 27 April 2015 (UTC)[reply]
At User:Be..anyone, using MonoBook skin, the coordinates are so small (font-size:9px) as to be unreadable, and indeed are barely recognisable as coordinates. I wouldn't have known what this item was, if it had not been for the hint above. Readability is not helped by the monospace font, nor by sitting directly on the grey bottom border of the page heading. Is this normal behaviour for <indicator>...</indicator> in MonoBook? --Redrose64 (talk) 08:32, 27 April 2015 (UTC)[reply]
+1 Also, I see the coordinates 'struck through' by the horizontal line which underlines the title. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 13:14, 27 April 2015 (UTC)[reply]
Yes, that's what I meant by "sitting directly on the grey bottom border of the page heading" - it has the appearance of a strikethrough; but zooming in to an excessive level shows that the grey line is behind the blue link text, not in front of it. --Redrose64 (talk) 13:25, 27 April 2015 (UTC)[reply]
xx-small was (now back to smaller) a temporary hack after an unclear issue on wikidata, where the status indicator was pushed to the right over the window edge, forcing a line break between lat+lon. I tweaked it to be at least still aligned in this situation. The position here with Monobook is odd, yes, but I want the same source everywhere, ignoring some noinclude + BiDi magic on Meta for global user pages. "Does not work only on enwiki for xx-small" is apparently a local bug. –Be..anyone (talk) 19:36, 27 April 2015 (UTC)[reply]
That is just some badly formatted text, not the actual coord template. Best way to test is to use the template directly. Try putting the following on your user page: <indicator name="coordinates-icon">{{coord|43.65|-79.38}}</indicator> -- [[User:Edokter]] {{talk}} 16:26, 27 April 2015 (UTC)[reply]
Just tested; the coordinates show, but not the globe... I get a hover box instead. Module needs work. -- [[User:Edokter]] {{talk}} 16:30, 27 April 2015 (UTC)[reply]