GeoNames Home | Postal Codes | Download / Webservice | About 

GeoNames Forum
  [Search] Search   [Recent Topics] Recent Topics   [Groups] Back to home page 
[Register] Register / 
[Login] Login 
wrong data for Hungary  XML
Forum Index -> Administrative Divisions
Author Message
jotta



Joined: 16/10/2007 23:35:39
Messages: 2
Offline

Hi there,

I'd like to report some important bugs regarding the Hungarian part of the geonames database.

We've found wrong duplicates in Hungary on the highest level: all the cities being "seat of a first-order administrative division" are also recorded as "first-order administrative division" which is not true: they're simply the seat of counties (and ofcourse most counties have different names than the seats).

That's why we couldn't yet build a correct tree structure from the data available, as there are false divisions (counties) and cannot either remove these records using the online editing facility, as the "fips" user owns them.

Let me show an example:

Szolnok, the city - correct:
http://www.geonames.org/715126/szolnok.html

Szolnok, the wrong duplicate:
http://www.geonames.org/715125/szolnok-megyei-varos.html

In its name it says "megyei varos" which means "city of a county" (sometimes a verbose alternative is used, "megyei jogu varos" - "city with county rights"). Both terms are only used in an administrative manner (I mean they're not used on maps or in the road signs).

The division Jasz-Nagykun-Szolnok is present (also correct):
http://www.geonames.org/719637/jasz-nagykun-szolnok-megye.html
(megye means county)

The same applies for these cities and false records:
http://www.geonames.org/721472/debrecen.html
http://www.geonames.org/721467/debrecen-megyei-varos.html
http://www.geonames.org/721239/eger.html
http://www.geonames.org/721227/eger-megyei-varos.html
http://www.geonames.org/3046526/pecs.html
http://www.geonames.org/3046510/pecs-megyei-varos.html
...and so on.

We'd like to ask removing these false records from the geonames database. Could you let us know who to contact?

Thanks in advance,

jotta.
marc



Joined: 08/12/2005 07:39:47
Messages: 4412
Offline

Hi Jotta

This is pretty difficult and I am glad you want to help us out with it.

First we have to distinguish between Cities (feature class P) and administrative divisions (feature class A). This means we certainly have "Szolnok" the city, which is also the seat of the administrative division and therefore gets the feature class P and the feature code PPLA. Furthermore we have the administrative division "Jasz-Nagykun-Szolnok", which has the feature class A and the feature code ADM1.
So far there are no problems, but what is "Szolnok Megyei Város"? Gwillim Law from Statoids.com writes : "Hungary is divided into nineteen megyék (sing. megye; counties), one föváros (capital city), and 22 megyei jogu város (urban counties, county boroughs, cities of county right)." [1]. Both administrative divisions have a fips and an iso code [2] and [3]. So I think we should go with them and keep it as ADM1.

What do you think?

Regards

Marc

[1] http://statoids.com/uhu.html
[2] http://www.geonames.org/HU/administrative-division-hungary.html
[3] http://en.wikipedia.org/wiki/ISO_3166-2:HU

[WWW]
jotta



Joined: 16/10/2007 23:35:39
Messages: 2
Offline

Hi Marc,

I'm trying to understand the coding and categorization mechanisms mentioned (just learning the geonames terminology, and as I can see there's so much more

According to the links you've provided, only the Wikipedia ISO_3166-2:HU is the one being the most clear and correct: there are

- 19 counties (megyek)
- and 23 cities with county rights (megyei jogu varosok)

The "city with county rights" is only a categorization, it describes that the city is the administrative seat of the county. As you can see, this 'label' is not even attached to the name of the cities anywhere (nor in Wikipedia or elsewhere), so there's no reason to use it. Besides, this information is perfectly recorded by P and PPLA for Szolnok itself.
(However, the cities itself ofcource do use this describing name on their websites, brochures, and so on, just to emphasize their importance).

(Note: as you can see there are more cities w.c.r. than counties: it must be for some historical and/or administrative reasons - some districts were joined into one county, and in some cases it was easier to keep two seats for the county).

Meanwhile I've found another piece of information that might be useful: even the statoids.com table is correct, but you have to take care of the 'type' column: 'm' means megye (county) and 'v' means varos (city).
The "only" thing that's incorrect that all these entities are treated divisions (as they're listed together in a table entitled 'primary subdivisions'), which is not true (how could a city be a subdivision at the same level as its container county?).

What I think right now is that it must have been some kind of misinterpretation of terms when the source databases were created and/or compiled, and they can be safely removed.

I can also try to find someone to ensure the above (eg. a professional cartographer), in case you'd like to.

All the best,

Jotta
levanti



Joined: 01/11/2007 09:19:15
Messages: 2
Location: Hungary
Offline

Hi,

As far as I can see there are several problems with the Hungarian administration levels here at Geonames. I try to sum them up.

The first one, mentioned by Jotta, is that the ISO 3166-2 is not very much used in Hungary. (Actually I've never seen it used.) The county system is more frequently used by the general public, it is described at
http://en.wikipedia.org/wiki/Counties_of_Hungary

And since we joined the EU recently, official organizations started to use the NUTS division, which is also based on the county system, described at
http://en.wikipedia.org/wiki/NUTS:HU

The ISO system has a conceptual problem, also pointed out by Jotta. The counties are misleading in this division, as they denote the county without its seat, while the seats are stored at both city and admin level.

In my opinion we have to decide which is more important: consistence or general usage.
[WWW]
levanti



Joined: 01/11/2007 09:19:15
Messages: 2
Location: Hungary
Offline

I would like to mention some mistakes which should be corrected:

http://www.geonames.org/search.html?q=adm1&country=HU

1) Inconsistence of long and short forms (+spelling)
In geonames, in some cases the counties are stored as Tolna Megye, Heves Megye, Nógrád Megye, ... while in some cases they are only referred to as Zala, Komárom-Esztergom, Fejér... without the "megye" tag.
In the official admin description they are called Tolna megye, Heves megye, Nógrád megye, Zala megye, ... (note: spelling is not with capital M!), while the short form is used on maps (Tolna, Nógrád, Komárom-Esztergom, Vas, ...)
The use of short forms is tricky, because there is one county, namely Veszprém, which bears the name of its seat, so this way it is hard to distinguish which object you are referring to.

2) Update
There is one new city with county-level rights. In November 2005 "Érd" in Pest county has become such a city.
Source: http://en.wikipedia.org/wiki/%C3%89rd#History

http://www.geonames.org/HU/administrative-division-hungary.html

3) The name of the subdivisions are shown without the accents
4) The capitals or seats of some entities are shown incorrectly
The correct seats are as follows:
Csongrád -- Szeged
Fejér -- Székesfehérvár
Győr-Moson-Sopron -- Győr
Zala -- Zalaegerszeg
[WWW]
rumcajsz


[Avatar]

Joined: 19/03/2008 22:35:02
Messages: 1
Location: Hungary
Offline

Hungarian Administrative Divisions are stil wrong.
The problem solution are in the previous message.
marc



Joined: 08/12/2005 07:39:47
Messages: 4412
Offline

I think most points should be fixed by now. Please feel free to help us fix the remaining ones.

Marc

[WWW]
chrisknight



Joined: 09/02/2014 20:45:36
Messages: 5
Offline

Hi
We just downloaded and checked some of the data you provide and on the first quality check we found some issues when we downloaded the GB.zip file.....which we assume was for Great Britain.

On the file we found the following

1. Towns in Cyprus showed up in the file and were categorised as GB locations.
2. Time Zones also did not make sense.

Please advise if there is indeed mistakes in the GB.zip file or if we are just misreading the provided data.

Kind Regards
Chris
marc



Joined: 08/12/2005 07:39:47
Messages: 4412
Offline

Hi Chris

What does your posting have to do with Hungary?
The UK do have possessions on the Island of Cyprus. Whether this is a mistake this you will have to discuss with other people. GeoNames only models the political facts, not whether they make sense.

Regards

Marc

[WWW]
chrisknight



Joined: 09/02/2014 20:45:36
Messages: 5
Offline

wrong section, sorry.....found th equery though a link on google....

regardless, cyprus is part governed by Turkey and Greece, nothing to do with UK since late 1800's...

just keep finding odd data everytime i start to use it.

regards
chris



chrisknight



Joined: 09/02/2014 20:45:36
Messages: 5
Offline

posted in wrong thread, can someone please move it to the relevant thread.

thanks
chris
marc



Joined: 08/12/2005 07:39:47
Messages: 4412
Offline

chrisknight wrote:

regardless, cyprus is part governed by Turkey and Greece, nothing to do with UK since late 1800's...
 


http://en.wikipedia.org/wiki/Akrotiri_and_Dhekelia

[WWW]
 
Forum Index -> Administrative Divisions
Go to:   
Powered by JForum 2.1.5 © JForum Team