Talk:List of cities proper by population/Archive 6

How to implement the decision to merge/redirect World's largest municipalities by population?
The article World's largest municipalities by population, a POV fork of this article that was created by BsBsBs, was nominated for deletion in September 2010. On 7 October, 2010, a decision was reached to "merge or redirect" (see decision here). It has been 10 months since then and I believe there has been enough time to merge any useful content (if there was any), and it should be time to redirect the article. As I don't have any experience with implementing an AfD and wasn't involved in the initial dispute and discussion that led to the decision, I'd like to solicit everyone's opinion on how we should proceed. Zanhe (talk) 04:14, 5 August 2011 (UTC)

Karachi is the LARGEST city in the world; CITED
It has been cited and people should stop arguing.

Also, people from China keep updating this page to show Shanghai and other cities as having such big populations when that data is not authentic. Someone needs to put a stop to this. People have mentioned in the past as you can see below. — Preceding unsigned comment added by Hjaveri (talk • contribs) 06:15, 5 August 2011 (UTC)

Delhi
Surprised no one noticed this, but it looks like Census numbers for Delhi have been released. They have the territory at 16,753,265, something like three million more people than Mumbia. --Criticalthinker (talk) 11:42, 5 August 2011 (UTC)


 * Thanks, it's now updated. Zanhe (talk) 23:02, 5 August 2011 (UTC)
 * The 16 million figure refers to the entire Delhi NCT, which additionally includes New Delhi, Delhi Cantt., and the non-urban parts of Delhi MC. The 2001 census figure for what is the "city proper" of Delhi excludes all these and should probably be excluded when putting the 2010 data. --Polaron | Talk 03:13, 6 August 2011 (UTC)
 * Have we yet really been able to find out the land area of the Municipal Corporation of Delhi? I remember this being very confusing with me arguing that it was just slightly less than the area of total territory (meaing that it took up the vast majority of the territory), and then someone else insisting using an "urbanized area" within the corporation as the "city proper". --Criticalthinker (talk) 03:46, 6 August 2011 (UTC)
 * The Delhi NCT is almost equal to the Delhi MC. New Delhi is the government enclave smack dab in the center of the Delhi MC, it's an area with mainly government buildings and some very nice houses which only the elite can afford, so its population is tiny, only about 200K.  The Delhi Cantt is the military enclave that's also entirely surrounded by the Delhi MC, and has a even tinier population of only 100K.  The NCT is quite small for a city the size of Delhi and has very little rural population.  In fact, the new commercial district of Gurgaon has now expanded into the neighboring Haryana State.  So for all practical purposes I believe we should treat the whole NCT as Delhi (which is basically what census2011.co.in does: it lists Delhi instead of NCT).  The populations of New Delhi and Delhi Cantt are just a rounding error.  Zanhe (talk) 06:06, 6 August 2011 (UTC)
 * In the 2001 tabulation of cities, there were 50 or so "census towns" that are listed as separate urban entities that are part of the NCT but not of the Delhi Municipal Corporation proper. This resulted in having only roughly 71% of the NCT population to be in the municipal corporation. Using the whole of NCT is probably ok until the new city-level data is released, which may not be for a few more years. --Polaron | Talk 13:06, 6 August 2011 (UTC)
 * We've been over this. "Census towns" are statistical divisions, not municipal divisions.  If any numebr is to be listed for Delhi, eventually, it will be the municipal corporation number that covers over 1,300 sq km of the territory.  For this page, the "census towns" couldn't possibly be more irrelevant.  It'd be like concerning ourselves with an "urbanized" area population of every city, which there is already a seperate page on. --Criticalthinker (talk) 09:47, 7 August 2011 (UTC)
 * This shouldn't be so hard, unless we allow it to be made hard. The number listed now appears to be for the "State" of Delhi a.k.a the National Capital Territory of Delhi. If Delhi is correctly informed, then the NCT "comprises nine districts, 27 tehsils, 59 census towns, 300 villages and three statutory towns – the Municipal Corporation of Delhi (MCD); the New Delhi Municipal Committee (NDMC); and the Delhi Cantonment Board (DCB)." The editor who changed this also changed the previous definition of Municipal Corporation to a National Capital Territory which is a pipe to List of districts of Delhi. I am no expert of Delhi, but the current entry does not appear to fall under the City proper definition of this list. "Almost" is not good enough. If we want the population of the MCT, then we need the population within the boundaries of the MCT.   BsBsBs (talk) 11:18, 7 August 2011 (UTC)
 * Whenever applicable - that is to say that whenever there is a local government covering a city proper or the majority of a city proper - the population of that local goverment/municipality should be used. In this case, there is actually local government within Delhi's capital territory.  In fact, it appears that there are three local governments.  Not sure if the Indian Census has yet released information for municipal corporations, but when it does (if it hasn't already), the city proper, in this case, would be the Municipal Corporation of Delhi.  It won't make too much of a difference, as the MCD covers the vast amount of the territory, but it'll definitely be a more accurate listing.  Hope that made sense.  It's not always that simple, as a lot of counties don't have local governments, or at least local governments that generally correspond with the general definition of a "city proper" (i.e. China), but in the case of Delhi, it is that simple, because a local government does exist beneath the territory. --Criticalthinker (talk) 10:08, 14 August 2011 (UTC)

Just in case someone gets tired of the Karachi war ...
I just found an unassailable source for Beijing. Geohive says that the City proper of Beijing is "Dongcheng and Xicheng = 92 sq.km." Adding up the population of Dongcheng and Xicheng gives a grand total of 2,162,000 - guaranteed WP:OR free! Routine calculations (as long as they are not performed on another source ...) are allowed. How about it? This way, Beijing sinks below the apparently agreed-upon 3 million mark - and can be removed! As for the City Mayors list: The Mayors really should know how many people live in their city. Apparently, they don't. BsBsBs (talk) 21:07, 5 August 2011 (UTC)

Tianjin & Guangzhou
It looks like things are starting to be properly sorted out, but the populations of Tianjin and Guangzhou seem to be sticky out, still. It seems for consistency, and like we've done for Beijing and Shanghai, it'd make sense to only add together the central and suburban districts for Tianjin and only the central and maybe "new districts" for Guangzhou. Both cover far too much hinterland, still, for every district in each of these cities/municipalities to be counted as the city proper. Tianjin's number, for instance, shouldn't include those districts clearly defined as "rural" and Guangzhou's number shouldn't include the outlying county-level cities, obviously. --Criticalthinker (talk) 00:33, 6 August 2011 (UTC)


 * I don't completely agree with all of Geohive's numbers, but Guangzhou's population looks reasonable. No it doesn't include the county-level cities, but includes all of the central and new districts.  Of the "new districts", Panyu, the biggest and most densely populated, should definitely be included, Huadu is where the airport is, so it also makes sense to include it.  Nansha and Luogang are tiny and inconsequential, so overall it looks fine.
 * I agree Tianjin includes too much hinterland. The Geohive number includes all districts, but I would prefer to have Baodi and Wuqing left out.  These two districts contribute 1.7 million or 16% of Tianjin's 11 million, so I still consider the result acceptable.  On the other hand, I think Shanghai's number doesn't include enough of its population.  Its "outer suburbs" are actually also quite urban with high population density, and all of them are already or soon to be connected to the metro system.  But we're not here to do original research, so overall I find Geohive's numbers quite acceptable.  Zanhe (talk) 06:34, 6 August 2011 (UTC)

Removal of unsourced definition (sorry, long)
I have taken the liberty of removing the following recently added statement from the intro of the article: "For cities in China, because administrative boundaries usually encompass large areas of rural territory, the urban agglomeration concept is used instead. Chinese urban agglomerations are based on the National Bureau of Statistics criteria for defining urban areas, which include an average population density of at least 1,500 per km² or the contiguity of built-up areas."

Reason: Using "the urban agglomeration concept" for Chinese cities is an arbitrary definition that is not grounded in a pertinent reference. No sources are given to support this. The statement that "Chinese urban agglomerations are based on the National Bureau of Statistics criteria for defining urban areas" is not supported by the reference provided. I am aware that this matter can become nauseating, but before jumping to conclusions, one should read the following carefully. Those who claim that this is another case of "verbal diarrhea" should realize that most of this has already been said, but apparently it needs to be reiterated, because it fell on deaf ears.

Citing sources recommends giving a page number when citing books. None was given except a rather vague "383-412," which corresponds to the complete page range of a book excerpt found on-line. I therefore had to conduct a search of the document. I found two instances where "1,500 per sq km" are mentioned:


 * Chan says on page 387: "The shaded areas are NBS-defined urban areas, according to criteria principally reflecting physical features and de facto population density—more specifically, an average population density of at least 1,500 per sq km or contiguity of the built-up area. The NBS-defined urban areas therefore are rarely in total congruence with the administratively defined urban areas (city districts). " (Emphasis mine.)


 * Chan says on page 392: "For the 2000 Census, the NBS has established relatively reasonable criteria for defining urban areas that are acceptable to many observers (Chan and Hu, 2003; Zhou and Ma, 2005). The criteria include an average population density of at least 1,500 per sq. km or the contiguity of built-up areas, as noted earlier." (Emphasis mine.)

As clearly evident, Chan simply claims that the NBS defines an urban area as an area that has a population density of at least 1,500 per sq km or is a contiguously built-up area. Chan does NOT say that this defines the city proper. Chan does NOT even say that the district must have an average population density of at least 1,500 per sq km to be considered urban. As a matter of fact, Chan explicitly states that "the NBS-defined urban areas ... are rarely in total congruence with the administratively defined urban areas (city districts)."

"City proper" is not characterized by pure "urban areas". This conflicts with the definition of city proper as "a locality defined according to legal or political boundaries" (see intro). Contrary to what has been claimed here repeatedly, Chan does not say that a Chinese city proper is defined by districts above 1,500 pop / sq. km.

As a matter of fact, Chan explicitly states: "The fact that this “city proper”–level data (as shown in Table 3 above) is not readily available or published officially in China seems to suggest that de facto city proper data, such as those derived from the Census, are still not used by mainland Chinese officials and researchers." In simpler words, Chan claims that Chinese officials do not use the city proper concept.

As we will see later, Chan's claim that the NBS defines an urban area as an area that has a population density of at least 1,500 per sq km is erroneous. It either does not, or not any longer. However, I am aware that "the threshold for inclusion in Wikipedia is verifiability, not truth". Those who want, can use his definition for "urban area," but not for "city proper."

Therefore, the Chan paper cannot be used as a source that proves that the population of a Chinese city proper can be derived by adding up the populations of districts above 1,500 pop / sq. km. Likewise, the source does not support the wild claim that "Chinese urban agglomerations are based on the National Bureau of Statistics criteria for defining urban areas, which include an average population density of at least 1,500 per km² or the contiguity of built-up areas."

Even if the reference would support the claim (which it does not), an independent tabulation of districts above 1,500 pop / sq. km. to arrive at a city proper count would not be allowed. As exhaustively explained in WP:OR, this constitutes original research and would conflict with unambiguously stated Wikipedia rules.

The stated intent of the Chan paper is to "to unravel some of the complexities of Chinese city population statistics at the city level, particularly with regard to large cities." It is a paper that attempts to explain a "highly complex system of urban definitions used in Mainland China — a system that appears to be the world’s most complicated and confusing." It is an excellent paper if one wants to understand the challenges and pitfalls of Chinese population studies and data. The paper is of no help to determine the current population of a Chinese city. Doing so would be highly amateurish, irresponsible, and would only add to the already considerable confusion.

Furthermore, arbitrarily using the agglomeration in place of "city proper" contradicts widely held principles in demographics. Bloom et al say in the 'Urban Settlement working paper for the UN World Institute for Development Economics Research: "For statistical purposes, where more specific definitions are needed, three concepts are generally used to define urban areas and populations: the ‘city proper’, the ‘urban agglomeration’, and the ‘metropolitan area’."

Bloom et al explain the differences between these three concepts as follows:


 * "The city proper is determined by legal and administrative criteria, and typically comprises only those geographical areas that are part of a legally defined, and often historically-established administrative unit."


 * "However, many urban areas have grown far beyond the limits of the city proper, necessitating other measures. An urban agglomeration is the ‘de facto population contained within the contours of a contiguous territory inhabited at urban density levels without regard to administrative boundaries (UN 2006: glossary). Urban agglomerations are thus determined by density: the agglomeration ends where the density of settlement drops below some critical threshold."


 * "A still more comprehensive concept is the metropolitan area. This concept includes both urban agglomerations and any ‘surrounding areas of lower settlement density that are also under the direct influence of the city’ (UN 2006: glossary). Populations in rural settlements can thus be counted as urban, as long as they fall under the direct political or economic influence of a city."

Confusing these separate concepts is a common mistake made by amateurs. Bloom et al expressly mention that "inconsistent definitions of the term ‘urban’ across time and countries imply significant measurement errors in the data."

Data should not be massaged to fit a certain perspective. Bloom et al specifically warn that "a proper reclassification of historical data can be an arduous, or even impossible, task."

Let's move on.

One editor had taken the time to go through the discussion archives. When doing so, one comes across a lengthy discussion in which it is explained that there is something called the "United Nations Demographic Yearbook" (UNDY.) The most recent one is that of 2008. The UNDY or "DYB" is compiled using questionnaires dispatched annually to more than 230 national statistical offices around the world. For the purpose of this discussion, the statistical offices are asked "how many people live in your cities proper?" and "how many people live in your urban agglomerations?" This is accompanied by a glossary that says:


 * "City proper is defined as a locality with legally fixed boundaries and an administratively recognized urban status, usually characterized by some form of local government."
 * "Urban agglomeration has been defined as comprising the city or town proper and also the suburban fringe or densely settled territory lying outside of, but adjacent to, the city boundaries." (Note that a strict reading of this rule would exclude an urban agglomeration that is wholly within the city boundaries.)

The United Nation also hands out a questionnaire in which respondents can refine, and if they want redefine the terms. It asks:


 * The definition of a city proper for this census is:
 * The definition of urban agglomeration for this census is:

The answers received then go into the footnotes of the tabulated data. The table of interest to us is the "Population of capital cities and cities of 100 000 or more inhabitants." The 2008 table is here. Please note carefully that this table provides separate entries for "City proper" and "Urban agglomeration." They should not be used interchangeably, unless allowed by a footnote. The footnotes are on page 330.

On careful inspection of the results in the table and of the footnotes, a few points become evident:


 * Most respondent countries accept the definitions provided by the UN as is.
 * Some respondent countries expand the definitions to account for special cases (Tokyo for instance).
 * Some respondent countries say: "Sorry, I can't give you the information for 'city proper', use 'agglomeration instead'" (Australia, for example).
 * Some respondents give other criteria, such as "City refers to densely built-up residential areas with a resident population of more than 3000 persons," (Lithuania) or "A city is a territorial authority which is a distinct entity, is predominantly urban in character, has a minimum population of 50,000 and is a major centre of activity within its parent region." (New Zealand North Shore.)

Interestingly, China accepts the definition "City proper is defined as a locality with legally fixed boundaries and an administratively recognized urban status, usually characterized by some form of local government" without comment for the UNDY 2008, as it did in previous editions. There is no footnote (page 281). Without such a footnote, the definitions provided by the UN are accepted by default.

As late as for the most recent United Nations Demographic Yearbook, the National Bureau of Statistics of China represented vis-a-vis the United Nations that as far as the bureau is concerned, China accepts "City proper is defined as a locality with legally fixed boundaries and an administratively recognized urban status, usually characterized by some form of local government" as the definition of "city proper."

The Chinese city proper data supplied seem to be those for the complete city or municipality and not for any subset. The data supplied are those of the 2000 census, and only of the de jure population - whether the listed data should be used is a completely different question. What imminently becomes clear is that as recently as 2008, the NBS did NOT use the alleged 1,500 pop / sqkm definition, neither for City proper nor for Urban Agglomeration. In the context of the UNDY, China's statistical bureau is completely urban agglomeration agnostic: It does not provide urban agglomeration data. Furthermore, by consistently delivering "City Proper" data to the UN, the NBS disproved Chan's claim that Chinese officials do not use the city proper concept.

One example: In the 2008 UNDY, China's National Bureau of Statistics reports a total population for the Beijing City Proper of 11,509,595. The data are marked 1 XI 2000 CDJC, meaning Census, de jure, complete tabulation, as of November 1 2000. Beijing's statistical bureau says that in 2000 "the registered permanent population totaled 11.075 million at the end of the year." Close enough with the usual Chinese precision. The Chinese government desired that the registered population of the whole Municipality of Beijing was reflected in the UN's city proper data - and not some subset. It also needs to be noted that the 1 XI 2000 CDJC population of the City Proper of Chongqing is listed in the 2008 UNDY as 9,691,901, while the Chongqing statistical yearbook gives the year 2000 de jure population of CQ as 30,910,090. The 9.7 million number is nowhere to be found in the CQ statistical yearbook, and remains a mystery.

Most importantly, UNDY shows that '''NO country engages in a philosophical discourse about the meaning of "city proper". There is a huge worldwide official consensus by more than 230 national statistical offices that "city proper" means what the UN says''' (except for some noted exceptions.) We can rest assured that these statistical offices come with the necessary qualifications. The United Nations is a place where ideologies routinely clash. However, none of the 230 national statistical offices raise the objection that "City proper is defined as a locality with legally fixed boundaries and an administratively recognized urban status, usually characterized by some form of local government" could be pure ideology.

As arcane, as misunderstood, as misused and as abused the term "city proper" may be, as absurd as it may sound to some, City proper could very well be the best referenced term in Wikipedia, and the one with the highest degree of consensus: The leading demographic institutions of 230 countries of this world agree year by year by written ballot that "City proper is defined as a locality with legally fixed boundaries and an administratively recognized urban status, usually characterized by some form of local government."

If the United Nations can agree on something, then so should we. And if you have read up to here, I will award you a barnstar of your choice. Thank you. BsBsBs (talk) 10:07, 6 August 2011 (UTC)

How to stop BsBsBs's continuously disruptive behavior?
This guy will just never stop flogging the dead horse. As soon as he loses one debate, he starts a new one. He's now starting a new edit war on Shanghai, see Talk:Shanghai and Talk:Shanghai. Zanhe (talk) 00:39, 15 August 2011 (UTC)

Hong Kong in China?
In this article, the country for Hong Kong is given as "China". But in List of urban areas by population and List of metropolitan areas by population it's given as "Hong Kong" (actually, for the latter, both Hong Kong and PRC China since it's partly in both).

I'd think we should be consistent. I looked into it some, and I think we usually give the country of entities in Hong Kong as "Hong Kong" rather than "China", except for this article. (Also List of metro systems, which doesn't have a country field but rather a list divided by countries includes the Hong Kong Metro in the China section). At Talk:List of tallest buildings in the world the same subject is under discussion.

Is there any objection to changing the country for Hong Kong from "China" to "Hong Kong"? Alternatively we could change List of urban areas by population and List of metropolitan areas by population, but we should be consistent I think. Herostratus (talk) 03:39, 14 September 2011 (UTC)


 * This topic comes up often, see Talk:List of metro systems and Talk:List of world's busiest container ports. The prevailing consensus is a compromise: to list China as the country but use Hong Kong's flag.  Zanhe (talk) 00:21, 15 September 2011 (UTC)


 * It is? I haven't seen that (China but with Hong Kong flag) in any article, including this one, and I checked quite a few. Herostratus (talk) 02:17, 15 September 2011 (UTC)


 * See the intro and infobox for Hong Kong. Also World's busiest airports by passenger traffic, and List of world's busiest container ports, just a few that I know of.  When disagreements arise this is usually the consensus result.  You're right it's not always consistent, on many pages not enough people care so it can go either way.  Zanhe (talk) 04:14, 15 September 2011 (UTC)

Melbourne and Sydney are not on the list.
The article for Sydney states that its population is 4,575,532 and that the population of Melbourne is 4,077,036. That means that Sydney should fit in at number 42 and Melbourne should be, after adjustment, number 50. I would fix it but I don't know how to hyperlink and edit tables. — Preceding unsigned comment added by Rpsrox (talk • contribs) 10:18, 26 November 2011 (UTC)
 * These are not city propers as this article defines them. Those are metropolitan areas.  They may refer to them as "cities" in Australia, but they are each in fact made up of multiple local governments. --Criticalthinker (talk) 14:17, 26 November 2011 (UTC)
 * Who put them on the list? They shouldn't be on the list.  They could make lists of metropolitan or urban areas, but these are not "city propers" under any definition of the word.  Sydney and Melbourne do, indeed, have city propers...they just happen to be tiny geographic areas that wouldn't make this list. Whoever put them on needs to remove them. --Criticalthinker (talk) 03:50, 16 January 2012 (UTC)

Paris
Where is Paris? How can I add it? --serhio talk 16:35, 13 October 2011 (UTC)
 * It's city proper isn't large enough to make the cut-off. --Criticalthinker (talk) 00:21, 14 October 2011 (UTC)

How large do a city have to be to make the list? What about Luanda? It got a population of 5 millions (I believe it's proper, as it got 8,5 metro) — Preceding unsigned comment added by Hokrollo (talk • contribs) 17:20, 23 March 2012 (UTC)

The "city proper" / "municipality" argument
It's been a while since this was brought up, but I think that it deserves another look: What is a "city proper"? Is it the same thing as a "municipality"? Or is it some more ambiguous concept?

Here are the arguments for both sides:

Municipality: The best approach is to define a city as the area governed by a certain government body, within a set of prescribed geographic limits. It's consistent and reliable; nobody can argue about the boundaries of an administrative division. The city proper definition doesn't mean anything, because there's no standardized definition of "city proper" is, only shaky ones that depend on culture, or on the data source. The resulting list has numbers that are entirely arbitrary.

City proper: The best approach is to define a city as a certain standard area that corresponds with the idea of "city" as we know it -- an urban center, without the suburbs*. It's meaningful and intuitive; it lines up with people's perceptions that a city has its own unique identity. The municipality definition doesn't mean anything, because there are a lot of municipalities that contain suburbs, or even rural areas. The resulting list makes some cities seem much larger, and some much smaller, than they are in real life as collections of tall buildings and such.

* (Don't argue with me about my use of the phrase "city proper" to mean this. I didn't come to debate the definition of words. I came to debate what a list of the largest cities in the world should look like, and this is the term I am using to refer to one side of the debate so that I can talk about it.)

So now we have two articles, this one, and the one on municipalities, each of them claiming in their own unique way (one with title, one with content) to be the "real" city-proper article. I think that both ways of listing cities have their merits. The municipality definition is relatively straightforward, and it's very easy to make a list with. However, it's not very fruitful, because the list you get doesn't really say much about the cities in question, except the size of the areas over which their governments have jurisdiction.

I realize that this is a long and sometimes quite formatted post, but the question I really want to ask is: what are we going to do about this? (More ranting to come, maybe) Someone the Person (talk) 01:34, 5 June 2012 (UTC)
 * Any list that shows Chongqing at a population over 30 million is not a list I'm ever going to support, and that hasn't changed. "City propers" have their problems, but the problems presented by "municipalities" - particularly by the Chinese definition, which is so much different than the meaning in most other parts of the world - are so much greater.  I think with a very few exceptions, the list currently works, as is.  It's an acceptable compromise and balance. --Criticalthinker (talk) 02:39, 5 June 2012 (UTC)

Dongguan
Admittedly, I'm not very familiar with Dongguan, but shouldn't the city proper, in this case, be just the four "districts" instead of the entire prefecture including the "towns"? Is Dongguan developed relatively evenly across its near 1,000 square mile area? --Criticalthinker (talk) 03:49, 12 June 2012 (UTC)
 * Dongguan is one of the biggest manufacturing centres of the world. Most of those "towns" are actually huge clusters of factories whose residents are mainly migrant workers. Judging from Google Satellite, the entire prefecture-level city is a contiguous urban sprawl except for the mountainous area that separates it from the neighbouring Shenzhen. -Zanhe (talk) 05:25, 12 June 2012 (UTC)
 * But aren't the "towns" essentially suburbs of the urban core? Just because the development is contiguous does not mean it is even. (It might be worth looking up population densities, perhaps using, to see how much of a difference there is between the subdistricts and the towns. I will do this soon if nobody else gets to it first.) Someone the Person (talk) 01:45, 6 August 2012 (UTC)

Continent
Does the continent represent the one the city is located on, or the country it belongs to? For example, Ankara / Turkey is located 100% on Asia yet it says "Europe & Asia". For me this doesn't make sense as this is a list of cities. 213.172.101.198 (talk) 11:34, 29 October 2012 (UTC)


 * I've removed Europe for Ankara. Elockid  ( Talk ) 17:35, 29 October 2012 (UTC)

To be added
BUENOS AIRES, this city is in everylist as a metropolitan area, it is a continuos from the city boundaries to the city, it works as well as new york. — Preceding unsigned comment added by 190.247.43.72 (talk) 06:16, 20 March 2013 (UTC)

Cities that should be added: Sydney, Australia (pop. c. 4.5*106); Melbourne, Australia (pop. c. 4.1*106); Addis Ababa, Ethiopia (pop. c. 3.3*106) — Preceding unsigned comment added by 140.160.176.138 (talk) 23:02, 22 May 2012 (UTC)
 * Sydney and Melbourne have been discussed very much. They're not municipalities, but then again, now that we have a separate list of municipalities, we shouldn't have any qualms about adding them, should we? Why don't you? Someone the Person (talk) 00:27, 5 June 2012 (UTC)

This list is missing Paris — Preceding unsigned comment added by 108.56.246.224 (talk) 17:09, 27 October 2012 (UTC)

This list is also missing Metro Manila, Philippines. — Preceding unsigned comment added by 112.206.20.203 (talk) 17:02, 11 December 2012 (UTC)

The list is missing Caracas, Venezuela (population of over 4.5m people) Eddypc07 (talk) 04:15, 21 April 2013 (UTC)

Cities that need to be added

 * China:

Hangzhou, city center pop. ~3.3 M, urban area ~5.7M, admin. area ~8.7M;

Nanjing, urban area pop. 7.1 M, admin. area 8.1M;

Chengdu, City Core + Suburban districts within city limit pop. 6.8 M;

Xi'an, city proper (CP) pop. 3.3 M;

Harbin, CP pop. 4.4 M;

Changsha, CP pop. 3.1 M;

Zhengzhou, urban districts pop. 4.2 M;

Wuxi, admin area pop. 6.4 M; districts over 1200 pop/km2 only: 5.1 M;

Kunming, CP pop. 3.2 M;

Hefei, Built up area (only districts over 2700 pop/km2) pop. 3.35 M;

Ningbo, pop. 3.5 M;

Quanzhou, 4 densest districts combined pop. 3.6 M, according to Quanzhou; — Preceding unsigned comment added by Solomonfromfinland (talk • contribs) 14:24, 11 December 2012 (UTC)

Foshan, city proper + inner suburbs pop. 6.05 M;

Xiamen, all districts pop. 3.3 M;


 * Australia: Sydney; Melbourne (i feel it is fair to include these cities because the population of the City of Melbourne or the City of Sydney is unreasonably small);


 * Algeria: Algiers, pop. 3.3 M; — Preceding unsigned comment added by Solomonfromfinland (talk • contribs) 12:22, 16 November 2012 (UTC)


 * Pakistan: Faisalabad, city district pop. 4.17 M, according to Wikipedia article on said city, and 2,793,721 in 2009 "Demographics"; — Preceding unsigned comment added by Solomonfromfinland (talk • contribs) 13:53, 11 December 2012 (UTC)


 * South Africa: Ekurhuleni Metropolitan Municipality, pop. 3,178,470 — Preceding unsigned comment added by Solomonfromfinland (talk • contribs) 18:26, 13 January 2013 (UTC)


 * Turkey: Izmir, Metro. Municip. pop. 3,366,947 in 2011 “TÜRKİYE İSTATİSTİK KURUMU: ADRESE DAYALI NÜFUS KAYIT SİSTEMİ (ADNKS) VERİ TABANI: Büyükşehir belediyeleri ve bağlı belediyelerin nüfusları - 2011”, meaning “TURKISH STATISTICAL INSTITUTE: ADDRESS BASED POPULATION REGISTRATION SYSTEM (ABPRS) DATA BASE: Metropolitan municipalities and municipalities depending on their population - 2011” in Turkish


 * Venezuela: Caracas, pop. 6 M.--Solomonfromfinland (talk) 10:05, 11 November 2012 (UTC)

Karachi
According to the Wiki page on Karachi (http://en.wikipedia.org/wiki/Karachi) it is as of 2012 the worlds largest city by population with approximately 21 million people living there. Or if you look at this article which is the source to the information on Karachi being the worlds most populated city in the world: http://www.citymayors.com/statistics/largest-cities-mayors-1.html — Preceding unsigned comment added by 90.227.190.204 (talk) 01:28, 19 December 2012 (UTC)


 * You need a non-Wikipedia source to support your data (see WP:RELIABLE). Also, CityMayors is not a source that I would consider reliable.  Elockid  ( Talk ) 00:23, 20 December 2012 (UTC)


 * This source is presenting some figures that they claim to be from the 2011 Population Census in Pakistan. Karachi is given a population figure of more than 21 million there. However, I haven't seen any authoritative source presenting any results from the Census, and in addition to that, this article tells that preliminary figures from the Housing Census shows some "statistical anomalies", especially in Sindh where Karachi is situated. I wouldn't trust any "2011 Census results" from Pakistan, until it's presented by a more authoritative source.--Pjred (talk) 14:38, 26 December 2012 (UTC)

Tehran
How did the population of Tehran go from 8-million-something to over 12-million so quickly? Did they recently add new districts to the municipality? If so, than the area measurement needs to be updated. --Criticalthinker (talk) 05:13, 31 December 2012 (UTC)
 * Indeed. how they reach by 8million to 12million people? 95.114.114.32 (talk) 12:34, 5 January 2013 (UTC)
 * Yes, that's four million people added to the city, supposedly, in less than a decade. That kind of relocation to a city isn't even happening in the fastest growing cities of the world, anymore.  So, unless Tehran annexed surrounding suburbs, I'm highly skeptical of this number.  If they did, than the land area needs to be corrected.  Who changed this, BTW? --Criticalthinker (talk) 02:51, 6 January 2013 (UTC)
 * IMO, I'm more inclined to think this is vandalism related.  Elockid  ( Talk ) 03:00, 6 January 2013 (UTC)
 * The change occurred here.  Elockid  ( Talk ) 03:28, 6 January 2013 (UTC)
 * Thanks for the person that corrected this. I'd have done it, but I wasn't really sure. --Criticalthinker (talk) 02:53, 4 April 2013 (UTC)

Chinese cities
Shouldn't the Chinese cities have their municipal populations and areas listed? That means Chongqing has the top spot with those expansive borders, but the municipalities are the cities proper, so they fit the article's definition. Dralwik|Have a Chat 02:44, 6 March 2013 (UTC)
 * Nope. And, this is something we've been through ad naseum.  The page is pretty clear about what it is and isn't.  It's a page about city propers, which while an elastic term, not so elastic as to include the entirety of something like Chongqing. --Criticalthinker (talk) 04:13, 6 March 2013 (UTC)
 * Another did disagree with this which spurred on the creation of World's largest municipalities by population. Elockid  ( Talk ) 04:17, 6 March 2013 (UTC)
 * Right, that was my point. This is already a decided issue, decided to such an extent that it lead to the creation of another page that compares apples-and-oranges. --Criticalthinker (talk) 11:16, 7 March 2013 (UTC)
 * For the record: "City proper" is not an elastic term. It is defined as the "area within city limits." It is this list that stretches or compresses that term beyond the breaking point, in direct contravention of the stated inclusion criteria of this list. World's largest municipalities by population does not compare apples and oranges, it strives to list the population within those administrative boundaries without making up arbitrary criteria, and without redrawing boundaries. I understand what this list was trying to do, but it used the wrong term. With that, I bow out. BsBsBs (talk) 12:06, 7 March 2013 (UTC)
 * I see you came back a few days later not only to correct spelling, but to add a further point. It's really time for you to move on.  As far as this page is concerned, it's a decided issue.  Sorry. --Criticalthinker (talk) 09:44, 11 March 2013 (UTC)

BsBsBs's fork World's largest municipalities by population was nominated for deletion in October 2010. It was decided that the article should be merged or redirected. It's been over two years and I don't think anyone has found any content that is worth merging here. I think it's time that we implement the redirect. -Zanhe (talk) 22:42, 20 March 2013 (UTC)

Beijing
I changed the stated pop. density for Beijing from 7400 to 8560 to fit the stated area and pop.--Solomonfromfinland (talk) 07:12, 7 June 2013 (UTC)

Surabaya
— Preceding unsigned comment added by 201.15.160.185 (talk) 19:52, 21 July 2013 (UTC)

Bangkok and Jakarta
I increased the populations of Bangkok and Jakarta by finding newer pop. figures, cited by the cities' respective Wikipedia articles. I moved Bangkok up in rank accordingly, and increased the cities population densities accordingly. Okay?--Solomonfromfinland (talk) 14:34, 24 July 2013 (UTC)

Karachi
I edited the karachi numbers to reflect the numbers shown in the link. Is that fine? — Preceding unsigned comment added by 115.248.58.203 (talk) 16:48, 4 September 2013 (UTC)


 * Yes, absolutely. Elockid  ( Talk ) 18:27, 4 September 2013 (UTC)

Mistake in the ordering
The population for the 39th spot, Santiago, is listed as 5,012,973, but the population for the 40th spot, Saint Petersburg, is listed as 5,023,313. I have relisted Saint Petersburg as 39th and Santiago as 40th.Blippy1998 (talk) 22:45, 6 July 2013 (UTC)

Part of this list is incorrect, Sydney has population of 4,667,283 and Melbourne has a population of 4,246,345 and neither of them are on the list. — Preceding unsigned comment added by 2001:8003:440D:A901:223:12FF:FE24:D7A5 (talk) 05:07, 10 October 2013 (UTC)

Sydney and Melbourne, Australia
The large, multicultural cities of Sydney and Melbourne are not included on this list which is a big oversight. Sydney has a population of nearly 4.6 million which is larger than most cities of the USA. Melbourne has a population of over 4 million. Also, as at 2013 I understand that the new megacity of CHONGQING is now one of the MOST POPULATED cities in the world at over 30 Million people. — Preceding unsigned comment added by 122.106.122.121 (talk) 00:51, 14 November 2013 (UTC)

There is a problem with comparing Melbourne, Sydney and Chongqing with cities in the US and elsewhere that is also systemic to this entire list. The "multicultural" definition and the Chinese regional definitions are not "cities proper" by global standards of the defintion. They are in fact the definitions of Metropolitan Areas or administrative regions (in the case of Chongqing). For a true comparison with European and North American cities you would have to use the MSA definition for the US, CMA for Canada and the NEW Urban Centre definition for the EU or use the official municipal population for the actual city, which in Melbourne's case would be 93,625. See City of Melbourne. Using your "multicultural" definition, the US has 10 cities larger than Sydney and 15 larger than Melbourne. 1.229.130.160 (talk) 06:14, 18 December 2013 (UTC) With an area of over 31,000 square miles, Chongqing is NOT a city. — Preceding unsigned comment added by 77.100.85.130 (talk) 22:18, 20 November 2013 (UTC)

Failure to follow Article Definition
This list is defined by its name as city proper by population. This means it should be the population of the city within its actual city boundaries following the global standard definition for "city proper." All cities should be measured by that definition, so then why is the definition column included that then allows contributors to use whatever definition they choose? The only valid entries on the list are the ones with "city proper" in the definitions column! Not "Special Adminstrative Regions" or "Statistical Areas," There are other lists for that! A metropolitan or Greater "anything" is NOT a city proper. If a city does not have a defined single adminstrative unit, it is not a single city, but really just a bunch of smaller cities sharing one name! If a city does not fit the definition or doesn't have population data gathered that follows the recognized global standard definition, it doesn't belong on this list! A city like Lagos with basically no single local government does not belong on a list of "cities proper," but should instead be compared only on lists with broader definitions like "Metropolitan Areas by population" or "Agglomerations by population." 1.229.130.160 (talk) 05:50, 18 December 2013 (UTC)

Since people are refusing to follow the UN recognized standard definition for city proper, I made changes to the list discription and highlighted the deviations from the definitions to match the data everyone continues to insist on using. If this is Unsat, then please purge this article of all the false data and only include data that follows the UN and WIKI's own definition for "city proper."1.229.130.160 (talk) 07:07, 18 December 2013 (UTC)

Annoying
This is very annoying. I feel the list at this point includes every city of over 3 million in the world, except for several in China, including Xi'an, Harbin, Nanjing, Foshan, Hefei. I wanted to use the English Wikipedia articles on these cities to find area and population info for them, to add them to this list, but the English Wikipedia articles don't cite sources for population. I tried the Chinese Wikipedia articles (I can use Google translate), but they don't properly cite websites, just some things described in words only.--Solomonfromfinland (talk) 06:30, 23 December 2013 (UTC)

Chittagong
Should Chittagong, Bangladesh, be added? Some sources state a population of over 3 million, but is that actually the city proper? Which should gbe defined as the City Corporation, by analogy with Dhaka. “City (town): Chittagong: map, population, location” suggests 3,920,222. “BANGLADESH: Chittagong” suggests 4,009,423 in the Statistical Metropolitan Area (not soemthing which I consider a city proper, especially since most of the districts are listed as having low population densities, not something I think of as a city. Also, I couldn't find a figure for the area of the city proper from an off-Wikipedia source.--Solomonfromfinland (talk) 04:03, 25 December 2013 (UTC)

Cite
Over the last few days I have added the following Chinese cities (in order of decreasing population): Xi’an, Foshan, Shantou, Chengdu, Nanjing, Harbin, Hangzhou, Quanzhou, Hefei, Changsha. Unfortunately, for many of them I couldn't get decent sources (including the whole prefecture-level city is generally not acceptable to me because it is so sparsely populated that it will include large rural areas; not my concept of a city); so I cited within Wikipedia. Can someone please provide decent, off-Wikipedia sources?--Solomonfromfinland (talk) 04:54, 28 December 2013 (UTC)
 * I added three more Chinese cities that need the same: Zhengzhou, Wenzhou, Xuzhou.--Solomonfromfinland (talk) 09:07, 28 December 2013 (UTC)

Question about the number of different world city lists
See post at. If you have any comments please post there. Eldumpo (talk) 09:45, 28 December 2013 (UTC)

Surabaya
Should Surabaya be added?--Solomonfromfinland (talk) 10:12, 28 December 2013 (UTC)

Faisalabad
Should Faisalabad, Pakistan be added? According to “Faisalabad Population Statistics | Faisalabad Pakistan”, Faisalabad had in 2012 a population of 3,547,446, which is not that unreasonable if Faisalabad means Faisalabad City Tehsil, which according to  “Geohive: Pakistan: General Information”, had a 1998 population of 2,140,346 (look under “Expanded administrative units” → “Punjab” → “Faisalabad Division”). To get from 2,140,346 to 3,547,446 people in 14 years (1998-2010), you need a population growth rate of (3,547,446/2,140,346)1/14 = 1.0367 → 3.67% per year; Faisalabad (rather, the article's source, “The 2004 Baseline Survey on Millennium Development Goals in AACs: Chapter 6: Faisalabad, Pakistan”) suggests a growth rate of 4.6% per year as of 1981.

I was concerned that List of most populous metropolitan areas in Pakistan didn't list Faisalabad as having over 3 million people; but that the metro area had only 2,880,675 people as of 2010. However, the Wikipedia article lists the 4th largest metro area of Pakistan as Rawalpindi, pop. 1,991,656. However, the source, “Pakistan	mise à jour: 10 septembre 2013 République islamique du Pakistan” (in French) supports a considerably higher population for Rawalpindi, 3,499,028. Also, while said source does support the figure of 2,880,675 people for Faisalabad, that's, according to the source, from 2007 rather than 2010, so there is severe inconsistency between the source and the Wikipedia article. Perhaps I should instead use the other source's suggested pop. for Faisalabad; 3,547,446. The source is more recent and not misused by Wikipedia. Btw, the fact that the 2012 figure stated by is so much higher than the 2007 figure stated by, even though the former is a “city” (a term which I prefer to mean a city proper, possibly restricted so as to have a very high population density), while the latter is a “metropolitan area”; doesn't bother me because “urban area” (the term used on the French website, “Aires urbaines”) is vaguely defined and may mean simply city proper.--Solomonfromfinland (talk) 11:20, 28 December 2013 (UTC)


 * The following source do not support the idea that Faisalabad has over 3 million people: “Faisalabad 1 Web Portal”;  “Pakistan Demographics Profile 2013”. However  “The Express Tribune: Delimitation scheme: Faisalabad divided into 346 union councils” says that “The Faisalabad district has been divided into 346 union councils in the new delimitation scheme – 157 in the urban area and 189 in rural areas.” This suggests that the urban area within city district limits must have well over 1.57 million people.


 * According to “Geohive: Pakistan: General Information”, the whole Faisalabad city district had a population of 5,429,547 over 5,856 km2 in 1998.


 * According to “The 2004 Baseline Survey on Millennium Development Goals in AACs: Chapter 6: Faisalabad, Pakistan”, Faisalabad is predicted to reach 3.4 million population in 2015.--Solomonfromfinland (talk) 11:41, 28 December 2013 (UTC)


 * According to “Local body politics: LHC cancels increase in union councils for Rawal, Potohar towns” (November 14, 2013), the delimitation scheme was declared unconstitutional.--Solomonfromfinland (talk) 12:03, 28 December 2013 (UTC)

Tehran
Why is Tehran listed twice ? 24.135.49.98 (talk) 10:58, 25 December 2013 (UTC)

http://esa.un.org  /  it is not a credible source!!! / the organization only make predictions--MIRAIL.MIHAI (talk) 18:17, 13 January 2014 (UTC)

Metropolitan area of Manilla should not be listed here
Manilla's wiki page states that the city has got 1,652,171 inhabitants. But some people like to insert Manilla in this list claiming that Manilla has almost 15 million inhabitants. But to achieve this number these editors sum the population of several other cities that surround Manilla, what constitutes a Metropolitan Area, not a city proper. MarcosPassos (talk) 19:32, 10 January 2014 (UTC)


 * I restored Metro Manila, on the grounds that there is strong precedent for national capitals in particular as being defined as being more than just a municipality - Greater London is actually a county, unlike the municipalities in a county, which are often thought of as city-equivalents. (The City of London is tiny.) Same with Santiago - a whole province, unlike other “cities” in Chile which are actually municipalities. Or the City of Sydney, which is unreasonably small for a definition of Sydney. For an area as crowded as Metro Manila, it seems that counting it as a city will better do justice to the fact that it is one of the most inhabited places on the planet.


 * Btw, I used Lagos State as a definition of Lagos that extends the population to 21 M. Okay? Previously, the city had been defined as a “Statistical Area” of about 999 km2 and about 7.9 M people. I felt the “state” definition was acceptable because the resulting expanded Statistical Area was sufficiently crowded. (There is no single municipality called Lagos.) The old population figure was from 2006, so I felt it had to be revised, especially since it's such a rapidly growing city, and the 2006 census is believed to have been very shoddy. If you prefer to stick to the old Statistical Area, here's a population figure that might be acceptable: 11.2 million; see “This Is Africa's New Biggest City: Lagos, Nigeria, Population 21 Million”. THe 11.2 M refers, presumably, to the Statistical Area; but since it is not clear what size area the 11.2 M figure refers to, it may be better to count the 21 M figures, which in all likelihood refer to the whole state.--Solomonfromfinland (talk) 22:28, 12 January 2014 (UTC)

Metro Manila with 633 km2 (not city) -> Istanbul 5343 km2 (is city), Melbourne 2,250 km2 - 1650 density (is city), Cape Town 2454 km2 - 1424!!!!! density (is city). Manila with 633 km2 is larger than London and Berlin together with 2464 km2. Just do not want as a district with 38 km2 (43000 density) to have 10 million people?--MIRAIL.MIHAI (talk) 18:08, 13 January 2014 (UTC)


 * Solomon, you said that you "restored Metro Manila, on the grounds that there is strong precedent for national capitals in particular as being defined as being more than just a municipality". Fine. But can you tell me why the hell do we have a List of metropolitan areas by population? If people simply can't accept that list is a list of cities proper by population, then I think we should merge this list with the one at List of metropolitan areas by population or remove one of them from wikipedia, since it's just pointless to have two identical lists. MarcosPassos (talk) 20:17, 14 January 2014 (UTC)


 * Metro Manila is a discrete administrative unit, as are most of the “cities proper” on this list; therefore it could arguably be considered a city; on the grounds of its extreme density; even its least dense municipality, Muntinlupa, has ρ > 9000 km-2. In fact, it is a national capital region; see its. And this list often does define cities proper broadly; it gives Capetown's pop. as 3.4; whereas the Lexicon Universal Encyclopedia (1988) edition that I read listed its pop. as only 215,000; clearly the definition of the city was expanded greatly. Also, alot of the cities on this list are national capital districts, dejure or defacto; e.g. Jakarta; Moscow (the only other Federal City in Russia is St. Pete); Mexico City; Bogota; Hong Kong (often considered to consist of several cities such as Kowloon, but given its high density, analyzing it as one city is probably more appropriate); Singapore (country); Dar es Salaam (Region); Berlin (state). Calling Metro Manila a city proper is probably legitimate on the grounds that it is not a whole metro area in that there are surrounding regions that are part of the same built-up area, that are part of the same metro area in the sense that New York City suburbs in New Jersey or Connecticut are part of the NYC metro area. Metro areas are usually not distinct administrative units; cities proper usually are (not always, in the  case of the “urban areas” in China and Australia, which I find to be the most acceptable definition of city in this case; not one that gave a “city” that was not too small or too sparsely populated)).


 * Why was the 21 M population for Lagos reverted? While the 21 M claim was an extraordinarily high value, and extraordinary claims require extraordinary evidence, I had four sources for the Lagos population. Also, given that the figure in all likelihood applied to the whole of Lagos State (≈3600 km2), the result was an acceptably high population density of 5800 km-2. Furthermore, the old figure was badly obsolete, given the city's rapid population growth and the unreliability of the 2006 census. Therefore, the city pop. needs to updated periodically. Photographs suggest Lagos is almost as crowed and unsanitary as Dhaka. Lagos as a state has the further advantage of providing a clear-cut statistical area.--Solomonfromfinland (talk) 19:04, 15 January 2014 (UTC)

“not one that gave a “city” that was not too small or too sparsely populated))”, “should be not one that gave a “city” that was not too small or too sparsely populated))”--Solomonfromfinland (talk) 19:08, 15 January 2014 (UTC)

“not one that gave a “city” that was not too small or too sparsely populated))”, should be “not one that gave a “city” that was not too small or too sparsely populated)” is what I mean to say. (I made punctuation errors.)--Solomonfromfinland (talk) 19:10, 15 January 2014 (UTC)

I meant, “not one that gave a “city” that was not too small or too sparsely populated))”, should be “not one that gave a “city” that was too small or too sparsely populated)”. (Note that I originally had an inappropriate double negative.)--Solomonfromfinland (talk) 19:22, 15 January 2014 (UTC)

Urban Area
Who added what I count as three "urban areas?" I'm sorry if these don't show up on this list, because there is no easy defintion of city proper, but "urban areas" don't belong here. There is another page for measurements of built up area irrespective of administrative boundaries. I hate to say it, but the Australian cities in particular probably can't make this list just because how local government seems to be structured, there. If you measure by urban area, that's not a city proper, and even if you decided to simply add up all of the local government areas in which the urbanized area occupies, then you're essentially measuring the total metropolitan area, and there is another page for that. --Criticalthinker (talk) 03:28, 29 December 2013 (UTC)
 * City proper has lost of definitions, as can be seen from this list. City of Sydney and City of Melbourne were not, to me, acceptable definitions of said prominent cities; they resulted in too small a population. Nor were Greater Sydney and Greater Melbourne acceptable definitions; they were too low-density; not my concept of city.


 * For instance, Finland has 9 municipalities of over 100,000 people, but I only regard it as having 8 cities of over 100,000. For the municipalities with under 700 people/km2 (Tampere, Oulu, Jyväskylä, Kuopio), I only consider the core urban area as a “city”; however, according to Finnish Wikipedia, in all of these four municipalities except Kuopio (population about 82,000 in the core urban area), the core urban area has over 100,000 people. Fro the municipalities with over 700 people/km2 (Helsinki, Espoo, Vantaa, Turku, Lahti), I will count the whole municipality as a “city”.


 * Therefore urban area was the most acceptable definition of said Australian cities. Besides, about the same definition is used in defining the Chinese cities, including; the set of all “core districts” is probably not a distinct unit administratively; but is used as a definition of the city.--Solomonfromfinland (talk) 23:00, 12 January 2014 (UTC)


 * Btw, I strongly agree that the population figure for Xi'an needs an off-Wikipedia source. I tried to find one cited by English Wikipedia, but I couldn't, because it was very poorly referenced, with the population not clearly attributed to any specific source. In general, I have been forced to do the same for other Chinese cities because of poor referencing. Can someone find good references? Same problem; I want to add Qalyubia, Egypt (pop. 4.2 M?); and Peshawar, Pakistan (pop. 4.6 M?, if you count the city district, with an area of 1257 km2), but I can't find sources for population because the English Wikipedia articles on said cities are so poorly source – dead links, mostly; or no inline citation; or you go to a website that you have to figure out several links to go to before you get to something that actually supports the relevant factoid on Wikipedia.--Solomonfromfinland (talk) 23:16, 12 January 2014 (UTC)


 * I added Qalyubia, with a CAPMAS reference for population. CAPMAS didn't state an area, so I cited a difference source for that. Okay?--Solomonfromfinland (talk) 20:44, 22 January 2014 (UTC)

Izmir
At only 322 people per km2, Izmir is not a city. Please use a narrower definition that includes only crowded districts, and then include Izmir only if it still has over 3M population. Also, the figure is not referenced.--Solomonfromfinland (talk) 05:55, 27 January 2014 (UTC)


 * I removed Izmir. “Greater Izmir has” has just over 3.5 million people; with the following municipalities: Aliağa, Balçova, Bayındır, Bayraklı, Bornova, Buca, Çiğli, Foça, Gaziemir, Güzelbahçe, Karabağlar, Karşıyaka, Kemalpaşa, Konak, Menderes, Menemen, Narlıdere, Seferihisar, Selçuk, Torbalı, Urla.


 * See “TURKISH STATISTICAL INSTITUTE: ADDRESS BASED POPULATION REGISTRATION SYSTEM (ABPRS) DATABASE: Population of province/district centers and towns/villages by districts - 2012: İzmir” for populations. However, the total area of these districts is just over 5300 km2; too low density to be a city IMO. See  “İl ve İlçe Yüzölçümleri” for areas. If you want İzmir to be a “city”, I suggest using the following 11 core districts: Balçova, Bayraklı, Bornova, Buca, Çiğli, Gaziemir, Güzelbahçe, Karabağlar, Karşıyaka, Konak, Narlıdere. That results in a “city” with an area of about 816 km2 and a pop. of 2,816,632.


 * Okay?--Solomonfromfinland (talk) 10:07, 31 January 2014 (UTC)


 * For more information see File:İzmir districts.png and İzmir ilinin ilçeleri.--Solomonfromfinland (talk) 10:44, 31 January 2014 (UTC)

Metropolitan Lagos
The Lagos State is made up of 20 districts. The Metropolitan Lagos is made up of 16 districts: Agege; Ajeromi-Ifelodun; Alimosho; Amuwo-Odofin; Apapa; Eti-Osa; Ifako-Ijaye; Ikeja; Kosofe; Lagos Island; Lagos Mainland; Mushin; Ojo; Oshodi-Isolo; Shomolu; Surulere. The reliability of the presented census results is not very high. The authentic Census, clearly shows that the number recorded was 17552942. The number of males is 9115041, while that of females is 8437901. Many Nigerians, and informed commentators on demographics have loudly and vehemently condemned and rejected the under- reporting of the population figures "allotted" to Lagos State, for the 2006 Census. Authorities of Lagos State have attacked the results of the 2006 census, accusing the Nigerian National Population Commission of having undercounted the population of the state. This accusation is denied by the National Population Commission. In 2007, Gov.Bola Tinubu said it was a shame that Nigerians still lacked accurate and reliable data for planning and public administration in the 21st century, adding that the Federal Government could afford a total re-counting exercise. He said: “We can afford a recount throughout the country; we want a recount in Lagos State. This is the height of corruption: falsifying census figures just to please some people. No amount of adjustment done to the [provisional] figures can make them right.The [provisional] figure is totally rejected ”. The National Population Commission reports a census population of 8048430 for Metro Lagos and 9113605 for Lagos State in 2006. Eight years after the last National Census in the country, the National Census Tribunal has declared null and void the headcount in 14 Local Government Areas of the State ordering the Nigerian Population Commission (NPC) to conduct a fresh headcount in the areas to ascertain the true population figure. The Lagos State Government had, on behalf of 19 out of 20 old LGAs, which covers 40 out of the new 57 Councils in the State, sought the intervention of the National Census Tribunal to nullify the 2006 Census figure of 9113605 allotted to the State by the NPC based on petitions written by the Local Governments alleging irregularities which resulted in the exclusion of many residents from the exercise. The official figures issued by The Lagos State Government for 2006 are: 16060303 for Metro Lagos (999.6 square kilometres/1171 square kilometres with water (lagoon)) and 17552942 for Lagos State (3568.61 square kilometres).--MIRAIL.MIHAI (talk) 10:51, 6 February 2014 (UTC)
 * While I have seen a number of sources supporting said statements, I recommend sourcing the statements yourself, in your post.--Solomonfromfinland (talk) 04:34, 11 February 2014 (UTC)

Managing the Edits like herding Cats
Clearly the subject of ranking global cities can be highly contentious. People's national and civic pride often get involved and sometimes politics and hatred, too. I am happy to see from this talk page that most people here have shown a willingness to put personal feelings aside for the goal of creating a more accurate and factual page. The sometimes lengthy explanations while wordy, are appreciated in explaining why some edits are redacted and shows understanding that this subject can be very personal for some. But, verified and accurate content is afterall the ultimate goal of Wiki and I am happy to see several folks have decided to reign in the loose fact tossing on this page. Great job! 1.229.130.160 (talk) 05:00, 11 February 2014 (UTC)

Baghdad
I provided a new, larger pop. figure for Baghdad, that was more recent and could be directly cross-referenced, improving verifiability. Okay?--Solomonfromfinland (talk) 15:02, 15 February 2014 (UTC)
 * By directly cross-referenced I mean that the link was directly to a website that stated said population. IMO, having references that lead as directly as possible to the relevant factoid is an important part of Verifiability.--Solomonfromfinland (talk) 09:33, 16 February 2014 (UTC)

Sydney
The pop. for Sydney is excessive. It is that of the whole Statistical Division, which at ~12,000 km2, is not a city. “Australian Bureau of Statistics: 3218.0 - Regional Population Growth, Australia, 2011-12: NEW SOUTH WALES” says that Greater Sydney, as of June 2012, had 4.67 million people at a density of 380 km-2. Probably the people who complied the Urban Area figures that were cited, were duped.--Solomonfromfinland (talk) 10:30, 16 February 2014 (UTC)


 * “complied” should be “compiled”.--Solomonfromfinland (talk) 10:30, 16 February 2014 (UTC)

Alex
Who removed Alexandria? Was it because the density withing city limits was too low? If so, then give the combined population of core districts.--Solomonfromfinland (talk) 00:11, 17 February 2014 (UTC)

Shanghai - Lagos - Istanbul - Karachi
I have seen many reviews, but... many disputes: Shanghai -> City Proper (2605.79 km/2); Population - 17836133 (2010) / Suburbs (5155 km/2); Population - 22315426 (2010) / Region (6345.5 km/2); Population - 23019148 (2010) - 23.71 mil (2012) /// Lagos -> City Proper (999.6 km/2); Population - 16060303 (2006) / State (3568.61 km/2); Population - 17552942 (2006) - 21 mil (2012) /// Istanbul -> City Proper (2189.79 km/2); Population - 13907015 (2013) / Province (5343.02 km/2) - 14160467 (2013) /// Karachi -> City Proper (734 km/2); Population - 13969284 (2006) / Metropolitan (3527 km/2) - 21142625 (2011). Must be the "City Proper"... without the suburbs or vast region with several cities together. If we take the region (suburbs), in this situation are many cities. Examples: Delhi (1483 km/2 - 16787941 (2011)); Manila (633.11 km/2 - 11855975 (2011)); Jakarta (1490.52 km/2 - 17720485 (2010); Mumbai (1135.11 km/2 - 18394912 (2011)); Dhaka (1268.7 km/2 - 15077992 (2011)); Kolkata (1027.4 km/2 - 14057991 (2011); Cairo (1709 km/2 - 15628325 (2006); Seoul (2353 km/2 - 20414618 (2012))... etc.--MIRAIL.MIHAI (talk) 23:32, 17 February 2014 (UTC)

Continent column
I think the final column should be removed given that country is the column before and the entries are not obscure countries. It just makes the table bigger than necessary, for no real gain. Any thoughts? Eldumpo (talk) 23:40, 29 January 2014 (UTC)
 * I agree. The table is wide enough already, and the column adds little useful information. -Zanhe (talk) 06:38, 31 January 2014 (UTC)
 * I agree too. Thank you for the revision.--Solomonfromfinland (talk) 12:39, 18 February 2014 (UTC)

Hong Kong
Yes, Hong Kong is a country...--2.245.142.24 (talk) 00:19, 19 February 2014 (UTC)

Fais
For the area and population of Faisalabad, I cited a file that I found online. However, the system didn't hyperlink to the file, and the file was replaced by a soruce which didn't support the stated population figure. What do i do, so that the original file is linked to?--Solomonfromfinland (talk) 07:52, 20 February 2014 (UTC)
 * Luckily, I still have the file, so I can re-reference it. I also have the URL.--Solomonfromfinland (talk) 08:18, 20 February 2014 (UTC)

Delhi
It is incumbent upon the user who changed the area measurement of Delhi (the Municipal Corporation of Delhi) to show that the municipal corporation is the size he says it is. All other information points to it being what it was formerly listed. --Criticalthinker (talk) 08:00, 20 January 2014 (UTC)
 * If we use the 1,397 sq.km, we also have to adjust the population figure. Delhi has 11,007,835 inh. within the urban core (excluding separate census towns and villages within the Delhi Municipal Corporation) as defined by The Registrar General & Census Commissioner, and it covers about 431 sq.km as listed. 1,397 sq.km is for the whole Delhi Municipal Corporation, which includes all of the NCT, excluding just New Delhi Municipal Council and Delhi Cantonment Board, and the total population is about 16,4 million. So, the question is if we accept the urban core as defined by The Registrar General & Census Commissioner as the "city proper", or if we shall use the wider definition for the Delhi Municipal Corporation.--Pjred (talk) 12:09, 20 January 2014 (UTC)
 * Admittedly, I haven't been to the Indiain census agency's website in awhile, but I was almost sure the 11 million number was for the entire Municipal Corporation. BTW, the municipal corporation is what we should use for "city proper" just like we do for every Indian city, unless there is something incredibly and structurally different about the municipal corporation of Delhi as compared to the other municipal corporations in the county.  Could you guys post a source to some information to clear this up? --Criticalthinker (talk) 04:00, 21 January 2014 (UTC)
 * A problem is that The Registrar General & Census Commissioner don't present population figures for the whole Delhi Municipal Corporation. Instead, they have defined what they consider to be the "city" (or urban core), and in addition to that they have defined many separate urban census towns and rural villages within the municipal corporation. I think we covered this topic quite well in a an old thread.--Pjred (talk) 08:19, 21 January 2014 (UTC)
 * I see no reason why we can't use the 1397 km2 definition. Likely then the population would be over 16 million.--Solomonfromfinland (talk) 08:21, 20 February 2014 (UTC)

Size of list / population cut-off
I added a note saying that the list is for cities of 3m+ population, as that seems to be the case at the moment. However, is 3 million too low, as the list is already quite lengthy and will only get bigger as city populations rise. What cut-off do external lists use. I notice Geohive only lists the Top 20. Therefore, what about reducing this list to cities of at least 5 million? Eldumpo (talk) 09:15, 2 February 2014 (UTC)
 * Probably not. I don't think a list with 86 cities is too long. There are plenty of lists of cities on Wikipedia that are longer than that, such as U. S. (289 cities as of 2012 estimates; not counting 8 CDPs); Canada (100 cities); Brazil (115 cities); Russia (357); Italy (149). This list could be made alot easier to expand if we introduce automatic ranking, like they have on Sweden. With such an innovation we could even expand the list down to 2 million. Increasing the cutoff to 4 or 5 million will lead to the loss of too much information.--Solomonfromfinland (talk) 07:45, 3 February 2014 (UTC)
 * In fact, German Wikipedia lists cities with pops. down to 1 million. The lsit is shoddy however; it is inconsistent about city proper versus metro/urban area; and it is outdated, from 2005.--Solomonfromfinland (talk) 15:18, 20 February 2014 (UTC)

Talk:World's largest municipalities by population
There is a discussion at the above article regarding merging sourced facts from that article into this one. Please post there if you have comments, in order to keep the discussion in one place. Eldumpo (talk) 16:10, 26 January 2014 (UTC)
 * Said two articles should not be merged. Not all cities on this list are municipalities. The Chinese cities usually aren't; low-density parts of a municipality don't count. Similarly, Lagos, Sydney, Melbourne, Giza and Lima are statistical areas; while London and Santiago are administrative divisions higher than municipalities.--Solomonfromfinland (talk) 15:28, 20 February 2014 (UTC)

Addis
I think the pop. for Addis Abeba, at 3,103 thousand is underestimated.Accorsing to the Wikipedia article, “It is the largest city in Ethiopia, with a population of 3,384,569 according to the 2007 population census with annual growth rate of 3.8%. This number has been increased from the originally published 2,738,248 figure and appears to be still largely underestimated.” If we take the 3,384,569 figure for 2007, and the growth rate of 3.8%, at face value, the city population should have been 3,384,569 × 1.0386 = 4,233,382.--Solomonfromfinland (talk) 16:52, 20 February 2014 (UTC)

Guangzhou
I was thinking of using a narrower definition of Guangzhou that would exclude districts of density ρ < 1500 km-2, resulting in a city with pop = 9,865,702; area = 2,346 km2; ρ = 4206 km-2. Or isn't the area of Guangzhou bigger because of the industrial districts?--Solomonfromfinland (talk) 11:45, 21 February 2014 (UTC)
 * See Guangzhou and List of administrative divisions of Guangzhou.--Solomonfromfinland (talk) 11:48, 21 February 2014 (UTC)
 * I'm not even sure if you have to get that arbitrary and complicated. To me, it seems pretty clear that the best way to do this would be to could cut off the two county-level cities, which are cities in their own right, and leaven the rest of the "metropolitan area" to describe the city proper as displayed on the map on that page.  If you really wanted to fight to take Huadu District out, I guess you could make a compelling case.  But, you'd actually have to make the case beyond its low density.  Is it connected to the districts which compose Guangzhou's contiguous urban area? --Criticalthinker (talk) 07:53, 4 March 2014 (UTC)
 * Just looked at an aerial map of Huadu District, and it does appear it's far enough away from the core districts that you could make a case for excluding it from the "city proper" definition we're trying to build here. It appears that Xinhua Subdistrict within Huadu District is not contiguous in a way that would justify it as being part of the definition we're using, here.  So, to reiterate, I'd chop off the two county-level cities and Huadu District.  In fact, I'd even cut off Nansha District as it also doesn't appear to be part of the contiguous urban area of Guangzhou, either.  It's probably even less part of the "city proper" than Huadu District.  What do you think?


 * Actually, I think the current definition of Guangzhou here used is fine. After all, the two Industrial Parks are probably just parts of the city that are mainly industrial not residential, like in many American cities.--Solomonfromfinland (talk) 03:21, 10 March 2014 (UTC)


 * Then why did you ask? I disagree.  Go look at aerials.  The industrial district to the south is located across channels and body of waters and seem like a seperate locality entirely.  I think you were right to question adding all districts for Guangzhou because there are some that certainly don't fit what we are defining as "city proper", here. --Criticalthinker (talk) 04:18, 10 March 2014 (UTC)

Athens Prefecture
Even if Athens Prefecture still existed, it seems an awfully abritrary choice for this list as a "city proper." I think this page, quite frankly, has gotten way out of hand, with any large urban area being forced into the measurement of a "city proper" whether that concept exists in these countries or not. The old list wasn't perfect, but it was far more consistent in its measurements. What used to be a measurement of a single local government/municipality within a larger urban area - or more rare but still correct, a regional government with control over a clearly defined smaller local governments (London) - has become a list where we're measuring anything from those original narrow-yet-consistent confines to things like "urban area" measurements irrespective of administrative boundaries, measurements of metropolitan areas using multiple administrative divisions with no regional government, and in some cases, measurements of entire regions including huge swaths of rural or wild lands. This case of Athens - with a measurement of an entity that doesn't even exist, anymore - just further highlights how arbitrary the measurement for "city proper" has become. Sometimes, being too active destroys the spirit of a wiki article. This is an example of that. --Criticalthinker (talk) 14:53, 15 March 2014 (UTC)
 * Yes, Athens Prefecture doesn't exist anymore. Athens might as well be deleted. Nonetheless, there should be, for purposes of this list, city propers for countries around the world. Some city propers here are sets of core districts, like most of the Chinese cities; and Tokyo, Istanbul, Ankara, Giza, Santiago, and Lagos.--Solomonfromfinland (talk) 05:02, 20 March 2014 (UTC)
 * I want to be clear so that words are not put into my mouth. I, too, want a fairly representative list.  It is my opinion, however, that the problem with the messing around of this page is not that we have too few cities, but that we may have gotten to the point of having too many in the spirit of inclusiveness.  It's my opinion that we should have a tighter definition than we've developed as of late.  We should place more a premium on accuracy than making a run-away list with what's become an inconsistent definition of the concept of a city proper from region to region.  Again, just as an example, we should probably never have on this page urban area defitions, since these aren't city propers by any defintion, and since there is not a page for that.  We should also discourage inclusion of a grouping of administrative districts where no larger regional government exists and/or where the country has no set concept of "city proper".  I think I could compromise on nations with "cities" which while maybe without a regional government have a governmental statistical definition of "city proper".  So, maybe Australia's "city propers" could make this list, even though the concept in this nation better corresponds with the ideas of metropolitan areas in similar countries.


 * Anyway, this is just my opinion, but I wanted it to be heard as I think we've quickly gone beyond the spirit of the original page in such a way that we're not overlapping with other lists of the world's largest metropolitan areas and urban areas. --Criticalthinker (talk) 12:48, 20 March 2014 (UTC)

By city propers I meant a working definition of CP.--Solomonfromfinland (talk) 05:03, 20 March 2014 (UTC)

Update to top 100 cities proper and add Chicago?
Surprised to not see Chicago in this list? For a city of just 234 sq miles, density of 11,864/ sq mi and 2.7 million people, seems like it should be there even if we have to expand list to top 100. 02:59, 22 March 2014 (UTC)02:59, 22 March 2014 (UTC) — Preceding unsigned comment added by College Watch (talk • contribs)
 * I recommend lowering the cutoff to 2 million. Then we would include Houston, Paris, Rome, Toronto... A specific cutoff pop. is easier than a top N scheme, as you only have to delete cities if their pop. drops below threshold, not when a city is added. I have, on a Word document, some code for automatic ranking, imported from Swedish Wikipedia; I hope to install it on this page. The code is:


 * {| class="sortable wikitable"



namn !! Finskt namn !! class="unsortable" | Samiskt namn !! Folkmängd !! Regionförvaltningsverk !! Landskap !! Sjukvårdsdistrikt
 * ! Svenskt




 * | Ackas || Akaa || || align=right| || Västra och Inre Finland || Birkaland || Birkaland


 * Will the code work here? It will be very valuable because manually adjusting all the ranks is already time-consuming and will get much worse if we lower the cutoff to 2M.--Solomonfromfinland (talk) 23:30, 22 March 2014 (UTC)

Russia
The list is an attempt to include only "cities proper"; however, for Moscow and St. Petersburg the populations are those of the federal cities, which are constitutionally the subjects of the Federation and include territories (and populations) other than those of the "city proper". At this point this whole page seems to be an original research anyway, so I'm not going to make any changes myself, but on the off-chance it can be saved, the population figures for Moscow and St. Petersburg will definitely need to be changed.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); March 14, 2014 ; 17:08 (UTC) With Moscow, even before the merger the law on administrative-territorial division recognized several rural localities which were not a part of Moscow proper. After the 2012 enlargement, the law has been further amended to clarify that there are other inhabited localities on the territory of the federal city of Moscow other than Moscow proper. Statistical documents such as those you cite are really not an appropriate source for determining the status of the city and/or its parts. Statistics are collected and reported in a manner most convenient and useful to the entities utilizing them; while they do generally follow the administrative-territorial and municipal conventions, they have been known to deviate from them, sometimes on purpose.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); March 14, 2014 ; 18:05 (UTC)
 * Saint Petersburg federal city actually includes 1 (one) city only, all former towns were merged to the Saint Petersburg city proper, former towns are named 'town' traditionally only. Moscow case is not so clear as 5 towns and 200 villages were transferred to the federal city of Moscow. Current statistics divides the federal city population to 11,971,664 urban and 136,593 rural, but official list of cities proper table Tabl-29-30-31-32-13.xls lists city of Moscow population data including federal city population total. Bogomolov.PL (talk) 17:34, 14 March 2014 (UTC)
 * St. Petersburg is actually a fuzzy case, since the entities now predominantly known as "municipal towns and settlements" have never been explicitly merged into St. Petersburg proper (at least I've never seen any document stating that). They merely are treated as a part of it within the federal subject framework (which is not what "cities proper" in this list are supposed to be about). Here, you can treat the city proper and the federal subject as one for convenience, but that would emphasize the original research nature of this list even more.
 * But Russian Statistical Agency clearly states all former towns dissolving and merging to St. Petersburg. We are using the reliable sources, isn't it? Statistical documents are really an appropriate source for determining the status of the city and/or its parts as regional divisions of this Agency are monitoring every local law changing locality status.
 * The same Agency before Moscow expansion noted lack of rural population and presence at the federal city territory only 1 (one) locality - Moscow City proper. Bogomolov.PL (talk) 21:34, 14 March 2014 (UTC)
 * A source may be reliable, but not necessarily applicable. Rosstat may decide which entities to "merge" for statistical reporting purposes, but it has no say in whether the entities are actually merged. There is no applicable source that says that the entities had ever been explicitly merged into St. Petersburg; for that specific local legislation should have been passed. And while the regional divisions of Rosstat are supposed to monitor every local law monitoring the status of the localities, we all know that in reality they do a really lousy job of it (OKATO mess to wit). Until the actual local laws, or a reliable secondary source explicitly referencing such laws, is found, I'm afraid taking Rosstat's word for it is a bit naive.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); March 17, 2014 ; 14:34 (UTC)
 * Ezhiki, I agree in a sense. A lot of this page does have pretty arbitrary definitions of what "city proper" means.  Moscow's current listing is not even the most egregious example.  As has been discussed on this very page, a lot of prefecture-level and provincial-level cities, as well as Municipalities for China are listed in their entirety as "city propers" despite including administrative areas in which the urban settlements have not spread contiguously.  As I said in my post just below this one, some of these listings are for entire metropolitan areas and some even for entire regions with huge swaths of undeveloped lands.  Moscow's listing is at least realistic in the sense that there is a supreme governing body for the entire territory.  Some of the other listings are just statistical areas with no supreme, regional government whatsoever (i.e. Lagos, Sydeny, Melbourne, etc...) and some are territories so large that it's laughable to call them even "cities" let alone have them under the concept of "city propers".


 * In the case of Moscow, since this page is at least supposed to be about administrative and/or municipal divisons into which an contiguous urban settlement extends, what you'd need to do is include all of the administrative okrugs in which the locality of Moscow proper urban area extends. If that still includes an area too large to constitute a "city proper", then you break this down further and include all of the district/raions in which the locality of Moscow proper's urban area extends.  In the first sense, it would seem to me that the idea of Moscow "city proper" would at least have cut off Troitsky and Zelenograd administrative okrugs, entirely, and then parts of Western.  The Troitsky, because I don't imagine the urban area extends that far, and then Zelenograd and parts of Western since they are not even contiguous to the rest of the federal subject.  If done in the second sense, that's too many districts to add together and/or subtract for me to do, right now, but I would imagine it'd leave off every district in Troitsky and many in Novomoskovsky.  If you can come up with a better definition when you get the time, you can make your case, here.  I think you could make a very convincing arguement for a redefitions of Moscow's city proper.  As for St. Petersburg, I am far less familiar with that federal city's layout. --Criticalthinker (talk) 10:15, 16 March 2014 (UTC)


 * The Russian version of this article several days ago was deleted as it was impossible to remove OR from it. Even here we see understanding a 'city proper' as an contiguous urban settlement extends, not a municipality limits. We have current Demographia. World Urban Areas & Population Projections. 10th Annual Edition March 2014 where an 'urban footprint' is a city definition. Bogomolov.PL (talk) 17:36, 16 March 2014 (UTC)
 * Unless something's change, there is already another wiki page on the list of the world's largest "urban areas", and it was always agreed that this page wouldn't duplicated that list, which was of a spatial measurement. The only compromise that was ever made was that when an primate administrative division was deemed too large, we'd use a collection of subordinate lower-tier divisions to define "city proper".  But, the point is that we never compromised to such a degree that administrative divisions/municipal divisions weren't the basic building block for the concept of "city propers".  This was never supposed to overlap the "urban area" page, because otherwise, there would be no need for both of these pages because they'd be redundant.  Administrative and municipal divisions HAVE to be the basis for the concept of "city propers" on this page.  Perhaps, I misunderstood you.  --Criticalthinker (talk) 13:28, 18 March 2014 (UTC)
 * Their definition of an "urban footprint" is the lighted area that can be observed from an airplane (or satellite) on a clear night. That's all good, but, out of curiosity, do you know how they determine the actual population within that "lighted area", if the local statistics are reported for the administrative units which the said "lighted areas" may or may not correspond to? I glanced through the document, but I have either missed this part or it's not there, and it's kind of important to know if that definition is to be used.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); March 17, 2014 ; 14:34 (UTC)
 * They are understanding the complicity of the task, the values estimated from municipal statistics may vary from year to year, but direction (common principles for every city) is positive, I guess.
 * 'Unlighted area' seems to be almost uninhabited, I see. Bogomolov.PL (talk) 20:11, 17 March 2014 (UTC)
 * I think the two fed. cities can stay as is, except for possible deletion of rural districts. Their high population density (> 3000 km-2 for both) qualifies them as cities.--Solomonfromfinland (talk) 04:57, 20 March 2014 (UTC)
 * They can stay as is, for now. But, if there are districts of these federal cities in which the urban area does not extend, a legitimate argument could be made to remove those districts.  I think you may be slightly confused as to what the concept of "city proper" has been on this page for quite some time.  I think I described it above quite well.  What we've always been looking for - when applicable - is for a mostly urbanized administrative/municipal district or set of administrative/municipal districts, particularly when we're talking about very large over-arching regional governments.  When large swaths of rural or wilderness area can be cut out of these definitions, that is always preferable.  Fortunately, in the case of both Moscow and St. Petersburg, these "city" governments, which are really regional governments, are further divided into smaller municipal and administrative districts.  So, all it will take to most accurately find their "city proper" measurements is someone familiar with the area and willing to do the calculations.  Hopefully, someone like Ezhiki, when he gets the time, can drill down and figure this out for us. --Criticalthinker (talk) 12:40, 20 March 2014 (UTC)
 * Moscow and St. Pete cover some rural or semi-rural areas. I am ambivalent on whether these should be removed, given the cities' high pop. densities (higher than most major U.S. or Canadian cities). If the cities were much less dense than they are, I'd definitely favor removing the rural districts. (For instance, such as with Chin. cities.)--Solomonfromfinland (talk) 23:48, 22 March 2014 (UTC)
 * At Moscow we have several administrative units officially named 'rural', we even know the population of the urban part of the city 11,971,664 and these 'rural' units population 136,593 . Bogomolov.PL (talk) 19:14, 20 March 2014 (UTC)
 * I would not simply take their word for it. Again, if the contiguous urban settlement of Moscow extends into an administrative/municipal district, it should be included.  If the urban settlement of Moscow does not extend into one of these administrative/municipal districts, it should be excluded.  I guess what I'm saying is that there may be some "urban" districts that are seperate settlements apart from Moscow proper, but still within the federal city boundaries.  And there may be some "rural" districts within the federal city, which include part of contiguous Moscow proper, so you can just go by their formal name.


 * This needs to be done visually with someone familiar with the area, or willing to go through Google or Bing aerials to see. Given how many administrative/municipal districts this have, it may be a bit time consuming. --Criticalthinker (talk) 03:37, 21 March 2014 (UTC)
 * To verify rural areas presence we don't need Google or Bing mappers: every our decision will be a sort of original research, I guess. But statistically rural areas population we need exclude from the city proper total. Bogomolov.PL (talk) 05:37, 21 March 2014 (UTC)
 * I'll keep reiterating until you get it, but if the city proper extends into one of these "rural" districts, it doesn't matter how much of it is undeveloped, it would be included. A city proper is defined by its urban area being contiguous.  The only way we wouldn't include a district is if the city proper urban area doesn't extend into it, whether it is fully rural or includes another larger settlement that is not connected to the city proper. I don't think this is up for debate. --Criticalthinker (talk) 06:12, 21 March 2014 (UTC)
 * It's a matter of definitions: it is no cities with real urban extend corresponding the official city limits. So the 'city proper' definition may be a NYC (in city limits) but not a Combined Statistical Area. But an other option is to name 'city proper' a sort of 'real city' when we are adding developed areas and subtracting undeveloped spaces (An urban area ("built-up urban area," urbanized area or urban agglomeration) is a continuously built up land mass of urban development that is within a labor market (metropolitan area or metropolitan region). An urban area contains no rural land (all land in the world is either urban or rural). But this definition in facr merges dozens of cities together in common 'real city' ('urban area'). Does it mean 'city proper'? Bogomolov.PL (talk) 06:58, 21 March 2014 (UTC)

This article is a mess
I already removed a few entries which don't correspond to "city proper". There are also a number of entries such as Lahore and Kinshasa that are using figures that don't even correspond to a "city proper" but rather an "urban area" which is entirely different from a "city proper". I'm going to try and clean-up an article. Elockid  ( Talk ) 15:18, 25 March 2014 (UTC)

City Proper
There are people who put data at random and destroys the spirit of a wiki article. On a site where everyone can change, the work is in vain. Some cities are deleted without logic, other cities are made with a population unrealistic... The list will not be good, never entered to thank everyone. It is my opinion, too many people with different opinions, destroys the spirit of fairness. In many rows, the objectivism people is decisive. To make the city bigger, often exaggerate. The work is in vain, because each change data after his good pleasure....--MIRAIL.MIHAI (talk) 16:53, 25 March 2014 (UTC)

Karachi
Source with 12.9 million people is not credible. It's only a projection over a period of 10 years. Population of Karachi grows very very quickly (580000 people/year - city proper; 800000 people/year - municipal). In 2008 it is impossible to have only 12.9 million people. In 2008 Karachi have at least 15 million people (over 17 million people 2013). The city's site is written for 2006 population: 13969284. Must be a governmental source, an official source, a source of accredited by Pakistan Government, but in any case, organizations from other countries that do 10-year projections.--MIRAIL.MIHAI (talk) 18:08, 13 January 2014 (UTC)
 * I agree, that figure of 12,991,000 is not reliable because it is only a projection.--Solomonfromfinland (talk) 17:25, 22 January 2014 (UTC)
 * I agree, the 13969284 sounds better; a better source and a source that is in better agreement with the fact that it's such a rapidly growing city, and what other sources state. Populations of over 20 million have been stated, but they are probably for the whole metro area. Lagos is also very rapidly growing, which is why I felt the figure of 14,920,000 was better.--Solomonfromfinland (talk) 20:55, 22 January 2014 (UTC)

Once again I say: Source with 12.9 million people and 13.1 million people are not credible. The sources are not credible from many points of view. First are the screenings on many years. In much the developed, as well as developing world, population growth is slowing. Not so in Pakistan according to reported preliminary results of the 2011 Pakistan census. Here population is growing much faster than had been projected. Pakistan's population stood at 197.4 million people in 2011, an increase of 63 million people from the last census in 1998 (~5 million/year). The new population is 20 million more than had been forecast in United Nations documents. Population of Karachi grows very quickly (minimum 550000 people/year - city proper; 800000 people/year - municipal). The cities of Karachi and Lagos are the only cities that grow with over 500000 people/year. In 2008 it is impossible to have only 12.9 million people or 13.1 million people!. The Karachi Metropolitan (3527 km/2) has an estimated population of 21142625 million people (2011 - preliminary census) and 23.5 million people as of April 2013 (estimated). The Karachi city proper (734 km/2) has an estimated population of 17 million people (2011 - minimum) and 18 million people for 2013. According to official data, in 2006 the city of Karachi have 13969284 million people. If we make a comparison with 1998 and 2011, it really is true and consistent with reality!!!. According to an estimate, 90% are migrants from different backgrounds and the population is estimated to be growing at about ~5% per year, mainly as a result of internal rural-urban migration, including an estimated 50,000 migrant workers coming to the city every month from different parts of Pakistan. In the official site archive is a local census in 2006.--MIRAIL.MIHAI (talk) 20:19, 15 February 2014 (UTC)


 * I found the following sources:


 * “Pakistan Today: Karachi’s population explosion far greater than experts’ calculations” (Dec. 6, 2011), which suggests a pop. of 18 million. What makes this particularly credible is that it states Karachi's area to be 3,530 km2, which is about what is generally stated, such as on.


 * “The Express Tribune: Population explosion: Put an embargo on industrialisation in Karachi” (October 6, 2013) (Cited by Karachi) suggests, “The city has an estimated population of 23.5 million as of April 2013, and a density of nearly 6,000 people per square kilometre. It is the third largest city in the world by population within city limits and the 11th largest urban agglomeration.” This suggests a city covering about 4000 km2, close to the supposed area of 3,530. Also,the “third largest city in the world by population within city limits” is in agreement with the fact that Shanghai and Chongqing municipalities are believed to have slightly larger populations. (Chongqing, however, is too low density to count as a city.)--Solomonfromfinland (talk) 01:40, 16 February 2014 (UTC)


 * Where does the 734 square kilometer number come from for the land area of the city district of Karachi? I can find no mention of this figure anywhere on the Karachi page. --Criticalthinker (talk) 07:03, 19 February 2014 (UTC)


 * I agree. We need a better citation, one that leads directly to the stated figures. I also agree that we might as well keep using the broader definition of Karachi, at 3527 km2.--Solomonfromfinland (talk) 06:13, 20 February 2014 (UTC)

The Karachi Metropolitan Corporation is divided into eighteen towns and six cantonments, grouped into six larger districts: Karachi South, Karachi East, Karachi Central, Karachi West, Malir, Cantonments. City Proper: Lyari Town, Saddar Town, Jamshed Town, New Karachi Town, North Nazimabad Town, Gulberg Town, Liaquatabad Town, Site Town, Baldia Town, Orangi Town, Gulshan Town, Malir Town, Shah Faisal Town, Landhi Town, Korangi Town, Karachi Cantonment, Clifton Cantonment, Korangi Creek Cantonment, Faisal Cantonment, Malir Cantonment, Manora Cantonment, Kemari Town (Bhutta Viltage, Sultanabad, Keamari, Baba Bhit, Machar Colony, Shershah), Bin Qasim Town (Ibrahim Haideri, Rehri, Cattle Colony, Quaidabad, Landhi), Gadap Town (Murat Memon, Gujjro, Yousuf Goth, Maymarabad). Total area of the Karachi Metropolitan is 3527 square kilometres. Kemari Town - 365.18 km/2 (rural localities); Gadap Town - Kirthar National Park: 848.25 km/2, 1144.26 km/2 (rural localities); Bin Qasim Town - 435.52 km/2 (rural localities). The total area of the rural localities outside the town is 2793 km/2 -> result: the city proper 734 square kilometres. Official government data relates only to 3527 square kilometres, but 80% of the area is wilderness or very little populated. The case is similar to Cairo (3085 km/2 - 80% dessert, wilderness). Each government provides data about the city no matter how big is the territory. This section of the site is trying to distinguish between cities, not regions - conurbations - vast suburbs - state etc.--MIRAIL.MIHAI (talk) 19:09, 24 February 2014 (UTC)
 * Does the population figure, then, correspond to the 734 square kilometers or the entire territory? I think the only thing that we should make sure of when we post or change numbers is that the populations and land areas given correspond to one another. --Criticalthinker (talk) 11:28, 25 February 2014 (UTC)


 * I have been wondering that as well.--Solomonfromfinland (talk) 21:17, 5 April 2014 (UTC)

Lahore's
population is totally out of date. It is from 1998.--Solomonfromfinland (talk) 21:16, 5 April 2014 (UTC)

Chennai
expanded its city limits from 176 to 426 km2. The pop here listed reflects the old city limits. On Largest cities in India, someone listed its new pop as 8.981 M, which I copied to this article, but it was not verifiable, so it was reverted. Could someone please verify Chennai's new pop?--Solomonfromfinland (talk) 22:15, 5 April 2014 (UTC)

Lima
I'd like to propose coming up with a more accurate definition for Lima's "city proper". While the provincial government calls itself a metropolitan government, apparently, only 30 of the 43 districts are what we'd define as urban and thus as a city proper (List of districts of Lima). Some of the remaining districts are actually almost totally rural from what I understand. Anyone want to look into this, or agree that perhaps only including the inner 30 districts would be the most accurate listing of Lima's city proper? --Criticalthinker (talk) 06:49, 10 April 2014 (UTC)

Beijing
Who messed with the population of Beijing? The note on the land area of the listing clearly notes that for "city proper" we're using the population for the two historic core districts, the four inner-city districts, and six suburban districts. The population listed covers the entire municipality, including the four outer districts that we've agreed not to count as being part of the city proper. Whoever changed the population, change it back. --Criticalthinker (talk) 07:08, 20 April 2014 (UTC)

Lagos
Based on the sources provided i'm not sure where the 16 million number comes from. In the sources 17.5 million is given. Also, on the page Lagos, there is a link to a 2012 source that estimates its population at 21 million. Should this number be used? Or should 17.5 million be used? Jacob102699 (talk) 23:56, 28 April 2014 (UTC)

Chicago
What rank would Chicago ,Illinois be if it was on this list? Also a side note i think that the cutoff Rank is a little too specific. The cutoff number should be A Arbitary number like rank 100 or 150 ,200 ect ,ect. 99.45.130.77 (talk) 18:47, 20 May 2014 (UTC)