Template talk:Infobox settlement/Archive 9

Please improve documentation

For example, what are the allowed values for the various "size" parameters? Is it a number, or a number followed by px? Thanks. -84user (talk) 23:26, 20 July 2008 (UTC)

Population estimates

I started a proposal to create a bot to update US city articles with the 2007 population estimates that is tangentially related to this infobox (the bot would be updating the population_as_of, population_total and population_footnes lines in the template accordingly) and a discussion regarding the topic on the Village Pump (located here) suggested that population estimates may not be sufficiently official for population figures. A suggestion was also made to modify this template to include separate lines for population figures and estimates. I got no response from the WP:CITY project on this, so figured I'd see if anyone has perhaps got this template watchlisted and would care to chime in as to whether it's preferable to modify this template as suggested, or just go ahead and consider the annual estimates as sufficiently "official" to use as population figures. Shereth 18:32, 22 July 2008 (UTC)

Adding an "estimate" field is something that should be explored. Many uses of the template will give the value in the total population field and place 2008 estimate in the "population_as_of" field. —MJCdetroit ^(yak) 16:16, 25 July 2008 (UTC)

I've placed some new fields associated with population estimate in the sandbox (in this version: here). You can review the results in the testcases page —MJCdetroit ^(yak) 16:28, 4 August 2008 (UTC)

Metric

I'm trying to use metric units for the article on Plymouth, but when I try to insert the units parameter it doesn't change it. This might be because it's in the United Kingdom, but I'm really not sure. There is also a discussion about the use of metric or imperial going on at Wikipedia talk:WikiProject UK geography/How to write about settlements#Climate and consistency. bsrboy (talk) 14:51, 25 July 2008 (UTC)

The UK defaults to imperial. —MJCdetroit ^(yak) 16:12, 25 July 2008 (UTC)

How do you take the default off? bsrboy (talk) 17:02, 25 July 2008 (UTC)

Rewrite the code, which would/may affect the present look of the all the UK infoboxes. —MJCdetroit ^(yak) 17:21, 25 July 2008 (UTC)

If you really wanted to do it, I guess adding a no-break space to the country parameter would do the trick. But is there any reason why Plymouth's infobox should look different from the standard for its country?--Kotniski (talk) 00:28, 26 July 2008 (UTC)

On previewing it, it comes up with the following message in the infobox "Expression error: Unrecognised punctuation character "&"". bsrboy (talk) 00:55, 26 July 2008 (UTC)

The nbsp has to be added to the value of the parameter "subdivision_name". Then it should work (but I still don't see the justification).--Kotniski (talk) 09:27, 26 July 2008 (UTC)

This is too annoying. I'm just going to use imperial throughout instead, except climate. bsrboy (talk) 12:44, 26 July 2008 (UTC)

Proposal: delete nickname field

We've had some controversy and discussion at Talk:Albuquerque, New Mexico regarding nicknames for the city. There's been a persistent problem where some editors add nicknames that are not well-known and are challenged by other editors. It doesn't help that the editors supplying the new nicknames are often IP editors or are later discovered to be sockpuppets of banned editors. For the time being we have solved the problem by not using any nicknames.

My feeling is that city nicknames are not encyclopedic. They are also difficult to verify; that is, it's hard to find reliable sources that say in so many words that X is a nickname for city Y, even if the nickname is well know (as Duke City is for Albuquerque).

Therefore I propose that we delete the nickname field from this template. --Uncia (talk) 16:10, 27 July 2008 (UTC)

Some places do have verifiable nicknames; most probably don't. We don't want to delete a field which is useful in articles in the first category, just because it occasionally gets misused for those in the second. (There are other ways of dealing with this problem; and in any case deleting the field doesn't prevent people from inserting the information in slightly different ways.)--Kotniski (talk) 16:36, 27 July 2008 (UTC)

proposing coords parameter

Would anyone object if an additional coords parameter were accepted by this template? I suggest doing this the same way as {{Infobox Protected area}} (which I recently changed) where the series of 8 coord parts (lat_degrees, lat_minutes, lat_seconds, lat_direction, long_degrees, long_minutes, long_seconds, long_direction) can be replaced by, for example:

 {{Infobox Settlement
  ...
 | coords = {{coord|45|12|34|N|123|45|56|W}}
  ...

This has several advantages.

• Latitude and longitude may be expressed, optionally, in decimal {{coord|45.678|-123.456}}
• The {{{1}}} parameter may be used to avoid the need to have two (or more) sets of coordinates in an article.
• The {{{1}}} parameter may be used to name the point on the external maps differently than the article.
• Minutes and seconds need not be specified if not known, or not appropriate. {{coord|45|N|123|W|name=Somewhere in Oregon}}
• Direct access to the coordinate qualifiers for specifying map scale, region, type, source data, etc. See Wikipedia:WikiProject Geographical coordinates#Geo tag.

Example: {{coord|45.678|-123.456|type:lake_region:US-OR_scale:10000|display=inline,title|name=Somewhere in Oregon}} Naturally, backward support requires continuing to accept the plethora of parameters. I propose that only if {{{1}}} is given a value, the other parameters are ignored. —EncMstr (talk) 16:02, 28 July 2008 (UTC)

This seems redundant to what we already have. And keep in mind that through subtemplates, we already use {{coord}}. —MJCdetroit ^(yak) 12:46, 30 July 2008 (UTC)

While the template uses coord, it doesn't give access to its many capabilities: How would one specify the coordinate (45.123, -123.456)? (Conversion to DMS is required.) How about setting a map scale of 1:150000? Or, in an article named somewhere (qualifier1, qualifier2) make the point display on the map as somewhere (without the qualifiers)? —EncMstr (talk) 14:45, 30 July 2008 (UTC)

Decimal is not a problem: see Pie Town, New Mexico.

If you wish to have these features then the present subtemplates and/or the code need to be upgraded. I'm afraid that adding a second coord parameter will only confuse people. The subtemplate used here is {{Geobox coor}} and the code use here is:

(Removed code)

—MJCdetroit ^(yak) 16:26, 30 July 2008 (UTC).

In fact if you use the little known parameter of |coordinates_type=, I think you'll be able to do all that you want. See my last edit to Pie town. I'll update the doc page later. —MJCdetroit ^(yak) 17:27, 30 July 2008 (UTC)

I've updated the doc page, but we will have to deprecate the field named "coor_type", which has a slightly different function. Since that field seems to indicate exactly what/where the coords are pointing to, should we rename it something like "coor_info" or "coor_pinpoint"? Kotniski, what do you think? Do you have any better names? —MJCdetroit ^(yak) 16:53, 31 July 2008 (UTC)

"coor_pinpoint" sounds cool. I don't know if anyone else has used this parameter, but there are lots (thousands?) of instances on Polish municipality pages, so it will require botwork to make the conversion.--Kotniski (talk) 17:18, 31 July 2008 (UTC)

"coor_type is now "coor_pinpoint". I have a bot sweep through and change them at some point. —MJCdetroit ^(yak) 02:30, 2 August 2008 (UTC)

From my experience with other geo/info-box templates, coordinates_type= allows specifying the "coordinates" parameters (region:, type:, scale:, and source:, ...) but does not provide for specifying the Coord template's "template" paramters (display=inline,title, format=dms, name= ...) The "difficulty" arises from the latter being separated by "|" pipe/vertical-bar charaters, which creates syntax problems in trying to pass them. A workaround for name= is to append "&title=" to the end of the other coordinates parameters. But the Infobox Settlement template should provide additional coordinates_ fields for each of the Coord template parameters (display, format, ...) so that they can be controlled or specified). LeheckaG (talk) 12:59, 5 August 2008 (UTC)

(outdent) I got it to display in the title in the sandbox. Please see the Mexico city infobox in the Testcase page. Seemed small to me. But that is very doable. —MJCdetroit ^(yak) 16:40, 5 August 2008 (UTC)

An article's "title" bar coordinates are primarily for "computer programs" use rather than readability by people (so "small" is okay, you might look at some other template to see what they do with font/size). Coordinates within an Infobox are the ones which require easy readability by people.

Lately there has been a performance issue where coordinates without a "region:" specified result in ToolServer's geohack.php calling region.php to attempt to do a lookup to determine an appropriate region (for mapping services to know which sub-set of maps). When too many concurrent blank/empty region: occur it results in the "Fastcgi Protocol Error" message getting displayed instead of the GeoTemplate map selection page. So an ISO 3166-1 alpha-2 region: code should be supplied whenever it known. For instance region:US for the United States or region:US-XX where XX is the ISO 3166-2 code for a State or Territory (they are generally the same as 2-letter postal abbreviations in Canada and the U.S.) So normally, North American city coordinates would look like:

• {{Coord|lat_d|lat_m|lat_s|N|lon_d|lon_m|lon_s|W|region:US-XX_type:city|name=name_of_city_if_different_than_article_title}}

I have not looked into Geobox coor(d) to see whether it has "standard" names for parameters; I recommend following whatever naming conventions {{Geobox}} and {{Coord}} use for parameters. For instance, Geobox sets the inline,title parameter by default and coordinates_no_title= turns title off.

LeheckaG (talk) 18:57, 5 August 2008 (UTC)

Request: area rank

Hello, I would like to request an additional field where I could input rankings by area (for example, note that administrative division ranks 5th by area in this or that country). Right now all area fields are auto-converted from km to mi. Renata (talk) 05:50, 30 July 2008 (UTC)

We discussed this once and concluded that such a field would either a) add too much clutter, or b) be potentially misleading, depending on how much information was shown. But you can use the area_footnotes field for such purposes (despite its name, it doesn't really produce footnotes).--Kotniski (talk) 08:08, 30 July 2008 (UTC)

GNIS identifier

I added a parameter to the template for Geographic Names Information System identifiers, and was subsequently reverted. Apparently, there's some concern as to whether or not this ought to be here. GNIS identifiers have already been added to the majority of US city articles via use of the "Blank" parameter. Since GNIS id's are also static, and provide a unique identifying number to geographic features in use by several agencies (such as the Census Bureau) having this field included in the infobox is immeasurably useful to automated tools such as bots. In the sake of transparency, my motivation was that I am running a bot that will use GNIS identifiers to circumvent problems with ambiguous names for the automatic updating of demographic information.

Anyway - I felt it was a fairly innocuous change, it only shows up in the infobox when the parameter is there (meaning there is no net effect on non-US articles). Given that there is no harm in adding this parameter and the utility of having a unique and easily accessible identifier for several thousands of articles I figured it would be uncontroversial. If anyone feels that this additional parameter is in any way detrimental or harmful to either the project or the readers, please explain. Shereth 23:18, 8 August 2008 (UTC)

Your addition seems sensible to me, but I recently proposed a parameter which could provide a way to decrease the number of parameters by 9 or so. That was opposed as "too complicated". —EncMstr (talk) 23:49, 8 August 2008 (UTC)

E, Did you actually read all of that #proposing coords parameter? It doesn't appear to be so because at no point did anyone oppose your proposal because it was "too complicated". The concern was that it was redundant to what we already have and it would not reduce anything because of how the location maps work. We have found ways to accommodate most of your request and were waiting for you to comment.

As for GNIS, my concern is that is it is too country specific (i.e., only the U.S. articles would use it). If we allow GNIS to have its own parameter, what's to say the Canadian, Mexican, Chinese, and all those other countries wouldn't argue for some country specific parameters or other U.S. parameters like the FIPS code? And adding extra parameters can slow down the performance of all the articles that use this template (75,000+) not just the U.S. ones. That was the reason for the "blank" parameters. In respect to a bot, why couldn't the bot be programed to look for "|blank1_name = [[Geographic Names Information System|GNIS]] feature ID" anyway? —MJCdetroit ^(yak) 15:26, 9 August 2008 (UTC)

Because "|blank1_name = " is nonstandard, and can be used for any information field. Yes, it would be fairly simple to program a bot to look for the data in that field but then we are assuming that article editors are going to put it there, as opposed to "|blank_name = ", or somewhere else. The idea is to have a reserved field for the data. FIPS55, the standard for places, is deprecated. Anyway, isn't worrying about what might happen in the future the wrong way to go about thing? It just sounds too much to me like you are willing to discard the known benefit for the sake of possible problems down the road. Besides, if the primary concern is being too country-specific it could be renamed to something generic like "|geo_id = " and then it can be used for GNIS, Geographical Names Board of Canada identifiers, or whatever Russian, Chinese, Mexican or whatever unique identifying system applies. Shereth 16:18, 9 August 2008 (UTC)

I would like to make several points:

• "Codes" should provide readers with a "one-click link" to an appropriate authoritative/normative/primary reference page, and not be a cryptic one with no link or reference and therefore only understood by a few (the GR templates should be BANNED). For instance, any mention of a GNIS code should be: [{{Gnis3|gnis_feature_id}} gnis_feature_id] so that a reader can see what the code is, but also be able to click on it to go to the appropriate reference page. The same should be true of "Census FIPS 55" codes, etc. (but sadly, it is currently not).
• There are other GNIS systems besides the (United States) USGS one, for instance British Columbia, Canada has one. So whatever fields and codes provided should indicate which and be "navigable" to the appropriate system. i.e. the field name should clearly navigate to the appropriate national or regional general system, while the code itself should unambiguously navigate to the appropriate specific reference page.
• There is more than one type of GNIS "Feature Class" (for a "populated place"): Census, Civil, Locale, Populated Place, ... For coordinates and elevation, generally "Populated Place" or "Locale" should be used - it is where the USGS plots a name on maps. The Census and Civil codes should ONLY be used in-line within the context of discussing where the centroid of a U.S. Census statistical area or where the geographic center of incorporated boundaries are. The generally should NOT be used in place of a (geography) coordinates, location, ... for "Populated Place" or "Locale" in ANY infobox. Census or Civil code references are appropriate in a population section of an infobox or article, but not a general geography/location section.
• My specific issue is having "Census borgs" update articles with nonsensical geographic information: often city or CDP coordinates "in the water" or "on top of a mountain". While the U.S. Census Bureau "might" be good at counting people, they apparently "failed" their geography lessons in several cases. Please leave geography updates to either other (typically local) Wiki projects which use USGS, NOAA or other authoritative geography and not "made up" US census stuff.
• Census data should be centralized behind templates transcluded into articles, and there should not be bots running around and "willy nilly" editing articles - in particular, bots SHOULD NOT BE deleting or moving information into UNIMPLEMENTED fields - extremely "poor behavior" for a bot.

All that "ranting" said ... there should be a standardized reference id field to a government reference should for geography and there should likewise be one for population. While many governments might not have GIS (Geographic Information Systems) or other publicly-accessible database systems, there should be standardized fields for when they do. Such "code id" fields should be the primary index or key for retrieving such information, and any code or id provided should not just be "text" but rather a reference link to such information. i.e. there should be a one-click link to geography, population, climate, ... reference data provided. So, there should be a "general" GNIS or GIS or ... field and it should be usable across regions, "bots" should not be "inventing it", but rather it should be invented by proposal or consensus, then bots should follow "implemented" behavior rather than "inventing" things and "breaking" article pages by "deleting" or moving information. LeheckaG (talk) 05:34, 12 August 2008 (UTC)

Before you go shaking me down for "bad behavior", how about you stop making assumptions? I did not have any intent to run a bot that "invents" new behavior - I added (what I felt was) an uncontroversial parameter to the infobox and was using that. As soon as another editor expressed a concern (and reverted the addition) I immediately ceased operation of the bot. I take offense at the accusation of running about "willy nilly" and operating a bot with "poor behavior". Also, if you had bothered to do a little bit of background checking you would know that the bot is only changing places with a location type of "populated place" - it is not touching CDPs or any other type of location, and therefore it is not using what is, as you describe, "nonsense geography". As for your first suggestion, I was not aware of the {{gnis3}} template and I do appreciate your mentioning its utility in this situation - but the rest of your rant comes across as rather rude, condescending and ill-informed. Next time familiarize yourself with the situation better before you begin accusing other editors of exhibiting poor behavior and ignorance. Shereth 13:54, 12 August 2008 (UTC)

(outdent) Shereth, I'd favor some type of generic fields like "geo_id_type" and "geo_id_info". That way, it could be used by many countries—not just the U.S. It should go at the bottom near the postal code and area code fields. Also, the table on the doc page would need to be updated with a description of these new fields. —MJCdetroit ^(yak) 12:02, 12 August 2008 (UTC)

Okay, something like that would be extremely helpful, and I have absolutely no issues with a generic identifier field. Shereth 13:54, 12 August 2008 (UTC)

Yes, I understand from looking at the histories what the sequence of events was (that you started something which the template revert subsequently "broke" beyond your immediate control). "Bad behavior" is also having a bot which summarizes that it is updating "2007 census" but actually makes additional changes: rearranging GNIS fields. When I saw "2007 census" and "bot" on the watchlist, I "ignored" it; Later to find out that it shuffled the GNIS fields. The "willy nilly" was in reference to ArkyBot's edits' which "broke" some Infobox pages which contained a reference link rather than simply an ID number, so it was not in reference to your bots' editing "mistakes" and not you. I suspect, but I am not sure that it occurred on a few pages which otherwise had "harmless" typos (either added or missing "parenthetic" punctuation) which the bots' edits changed into more serious Wikitext syntax errors resulting in pages which visually did not display properly due to the subsequent Infobox "errors". I am not sure about the bots' "Populated Place" assertion: one of the unified-home rule city-boroughs, guessing either Sitka or Skagway, had a Census or Civil GNIS code and not the Populated Place one - which could have occurred after your bot passed by; subsequently I inserted the appropriate Populated Place reference link before the other code - i.e. the blank1 GNIS field now cites both codes. LeheckaG (talk) 14:47, 12 August 2008 (UTC)

The incorrect summary is my fault; I had implemented an updated summary but apparently forgot to save that revision, so I do apologize for that. In any case, for your reference the source I used for the GNIS parameter is the "Populated Places" gazetteer available from [1], so it should only be using populated places and not Census or Civil GNIS codes. Anyway, I am quite prepared to have the bot go back through the articles it has edited (Alaska and Arizona) and either revert the GNIS codes to the blank field, or to a new field, depending on the result of the discussion here. As always, there are potential problems that I cannot forsee in the coding (such as unexpected "parenthetic" punctuation) and I do appreciate when instances of unexpected behavior, "breaking" the infobox, come to my attention so I can make sure it doesn't happen subsequently. Shereth 15:09, 12 August 2008 (UTC)

Why should the "Populated Place" coordinates be used instead of the "Civil" coordinated for a city? To me, it makes much more sense to identify a city by its "Civil" center (centroid) rather than a seemingly arbitrary "Populated Place" (uncentered) coordinates. 68.41.144.239 (talk) 23:06, 26 August 2008 (UTC)

Any coordinates need to be used with "common sense", the USGS is generally the official source for most U.S. maps (NOAA geodetic survey, US Army Corps of Engineers, US Coast Guard, FAA, FCC, each having their own specialized maps), so USGS "Populated Place" coordinates are generally a reasonable representation of a population on a map. While the U.S. Census Bureau produces maps to attempt to explain where population counts and demographics come from, census borders are primarily supposed to follow election districting borders (which are often determined by "party politics"). The closest approximations to U.S. Census borders can be found at http://www.census.gov/geo/www/cob/, and they have several disclaimers regarding accuracy which along with some other maps can be found at http://www.census.gov/geo/www/maps/CP_MapProducts.htm. The "problem" with both municipal boundary and census tract geographic boundary centroids is that they often can be separate from where the "mean center of population is". Where population population densities are higher, and "evenly distributed", and boundaries are relatively rectangular, then the geographic boundary centroids work out somewhat better. "Problems" occur when the population densities are lower, or unevenly distributed, or irregular-shaped boundaries/tracts, then the centroids often end up in "odd" places away from the actual populated areas. In Alaska, the census areas and municipal boundaries are both relatively large, so the geographic area centroids often end up away from where the populated cities actually are ("in the water", or "on a mountain", ...). In Ohio, Municipal and census boundaries are "often" odd-shaped having been "carved out" by annexations and dissolutions of various municipal corporations (cities, townships, counties), the method the U.S. Census Bureau uses to compute the geographic centroid works okay for relatively smaller "rectangular" or square areas, and it does not work as well with the "more common" irregular tract shapes formed by cities, townships, counties, and census tracts. With GNIS Populated Places, Civil, Census, or Locale, ... coordinates, one should verify that the coordinates map to a "reasonable" location on a map, and should match what one is describing.

When one is using Census-derived data, a "better" answer is to review the actual census tract boundaries rather than "blindly" re-publishing census coordinates. LeheckaG (talk) 23:56, 26 August 2008 (UTC)

Multiple Pushpins?

Is it possible to have a map with multiple pushpins, each with their own coordinates? And furthermore, can the color of some be changed? As in having, say, 5 red pushpins and 3 blue ones on a map? Is that possible, and if yes, how? — N-true (talk) 10:55, 15 August 2008 (UTC)

Not within the infobox, but Template:location map+ does what you want. See Hertfordshire#Urban areas for an example. Kanguole (talk) 11:25, 15 August 2008 (UTC)

Using Location map+ and Location map~ you can also "dock" it to the bottom of an Infobox by creating a table (which would contain the Infobox and the Location map) and setting the same CSS classes on the table as the Infobox has (primarily class=Infobox). LeheckaG (talk) 19:02, 15 August 2008 (UTC)