GeoNames Home | Postal Codes | Download / Webservice | About 

GeoNames Forum
  [Search] Search   [Recent Topics] Recent Topics   [Groups] Back to home page 
[Register] Register / 
[Login] Login 
Historical places and place names  XML
Forum Index -> Discussion of GeoNames Toponyms
Author Message
norheimd



Joined: 29/01/2009 09:28:42
Messages: 1
Offline

I find myself sometimes in the need of referring to historical places and areas with a URI. e.g. the Soviet Union, Poland before WW2, Luleå - a city that was moved 10 km due to get a better port etc.

Do any of you have ideas on the best way to do this? There is a feature called "historical site" under "spot-building-farm". Should this feature be used for this? I cannot find a single place where this feature is used though. Maybe a new feature class is more called for?

- David
marc



Joined: 08/12/2005 07:39:47
Messages: 3759
Offline

David

GeoNames does net include a timeline. To do it properly I think a single feature code would not be sufficient. For instance you also want to know what feature code the historic entry was and the time would also be required.

For some French city names used during the French revolution we have added the alternate names with a pseudo language code that includes the year of it's use.

Best

Marc

[WWW]
codepoet



Joined: 23/05/2008 21:37:34
Messages: 26
Offline

Hello,

I've noticed that there are some historical alternate names for various prominent cities, e.g. "Philadelphia" for Amman, Jordan, and "Islamabad" for Chittagong, Bangladesh. I'd like to mark these alternate names as "historical" in the gazetteer. What's the most appropriate way to do so? If necessary I can set the language code as Marc did above to some value indicating the approximate date of usage.

Best,
Mike
marc



Joined: 08/12/2005 07:39:47
Messages: 3759
Offline

Mike

I think it would be best to start an spreadsheet with all the information required so that we see what it needs and whether we have sufficient data to justify an addition to the database, the user interface and the dumps.

The language code have some consistency checks. It only accepts two and three character iso language codes and the pseudo codes post,icao,iata,fr_1793,abbr.

Best

Marc

[WWW]
dcstrawson


[Avatar]

Joined: 15/06/2009 16:31:24
Messages: 3
Offline

Hi -

I have a similar problem and have come across this thread.

For a project I'm working on, I need to display the English variant of placenames. For a number of places we have 'preferred' English - so I'm using that, however for others we have an English version but not 'preferred' so I've picked one of the English versions.

Now, that more or less works and I was pretty happy, then I came across Zimbabwe. My code now wants to call Zambia and Zimbabwe Northern Rhodesia and Southern Rhodesia. I'd say that no one (now) would call them that, so they are good candidates for historical names.

I will go in a moment and check 'Preferred' for the appropriate English names for these in Geonames, and will change my code to assume that the 'default' name for a place is used if there is no prefered alternate name for English, however it would be good if we could capture the English names that are no longer valid. If there were fields to do this in the alternate names structure I would try and update as I go.

Could there be an 'isObsolete' flag to cover these?

Daniel
dcstrawson


[Avatar]

Joined: 15/06/2009 16:31:24
Messages: 3
Offline

So, looking some more I decided to change my code so that if a place's alternative name isn't the preferred english name, it picks the default Name for the place instead.

This will fix Zimbabwe however if I then go to - say - Munich, it doesn't pick Munich over Munchen.

As I'm changing the data is there a definite way that I should think about these fields? If I ever meet an alternate name that is obviously english but doesn't have a country code should I fix this?
marc



Joined: 08/12/2005 07:39:47
Messages: 3759
Offline

I see you have added the language code 'en' to 'Munich', thanks.

There actually is an attribute 'isFormerName' on the database but it is not yet available on the user interface.

Marc

[WWW]
dcstrawson


[Avatar]

Joined: 15/06/2009 16:31:24
Messages: 3
Offline

Hi Marc - or the alternative names download.

That's great, do you intended to expose this information?

marc



Joined: 08/12/2005 07:39:47
Messages: 3759
Offline

There is an attribute in the db, but it is not filled with values. No point in exposing an attribute that is not filled. First it needs a strategy how to fill the attribute.

Marc

[WWW]
leifuss



Joined: 25/08/2009 15:18:46
Messages: 1
Offline

Hi Marc (and others)

Following a brief email exchange with Barry Hunter (the UK GeoNames Ambassador) I'd like to throw in my two cents on this topic too

I'm a Computer Science PhD Student at the University of Southampton (UK) researching the use of Linked Data in archaeology and other past-oriented humanities disciplines (history, classics, etc) and whilst GeoNames has provided extremely valuable for providing a global URI gazetteer, the lack of a standardised way to refer to historical places is the biggest challenge I've encountered with the service.

In particular, I'm currently Lead on an experimental project (partly inspired by GeoNames) to provide URIs and Linked Data for temporal periods (http://commoneras.ecs.soton.ac.uk - please note it's very slow and not very useful yet!) and we need to be able to tie periods down to their relevant geographical scope, for which GeoNames would be perfect. To do this properly however, we'd need to be able to refer to historical locations.

I'm aware of Marc's concern about mission creep (especially as regards to timelines etc.), so basically just here to ask for suggestions for the best approach, but I'd like to ask a couple of concrete questions to clarify what would be useful. If any of these aren't appropriate uses of GeoNames, let me know!

Q1 is it appropriate to add places that no longer exist and have no modern equivalent? Eg. A Roman province?

Q2 In cases where there is a modern equivalent, is it preferable to add an additional name, or just add the ancient place?

I appreciate this is something of a grey area. For example, the historic towns of Clausentum and Hamwic both lie under modern Southampton, but they are not the same, (it was founded on a separate site and then expanded over them). It's therefore not really appropriate to call 'Clausentum' the 'latin' name for Southampton. (It also seems to me there are some pretty compelling reasons for not calling Zimbabwe and Zambia 'Rhodesia' ). 'Londinium' could legitimately be seen as the 'latin' name for London though.

Ultimately my concern is not to pollute GeoNames with locations that don't fit its overall policy so any guidance at all is appreciated.

Best wishes

Leif

Leif Isaksen
Electronics & Computer Science/Archaeology
University of Southampton
codepoet



Joined: 23/05/2008 21:37:34
Messages: 26
Offline

Hello again,

I've noticed that the Getty Thesaurus of Geographic Names (TGN) contains some historical names of places. Would that be a suitable way to populate isFormerName?

Best
Mike
marc



Joined: 08/12/2005 07:39:47
Messages: 3759
Offline

I just see that I have not yet answered to the posting above by Leif. Sorry about this.

Leif:
There are some ideas how geonames could be improved with a timeline. Some table already have date attributes, like the subdivisions: http://www.geonames.org/LA/administrative-division-laos.html

q1: if the name of the roman provinces is still somehow in use today then they could be added as area or region, but not as ADM1,2 etc.

q2: I wouldn't add new records for Clausentum and Hamwic. Adding new places would make them exist today. I would rather add them as alternate names. That they have historically not had the same extent as todays Southampton is not a big problem. These names would get the language code 'latin' and it would not lead to confusion that Southampton all of a sudden would be called Clausentum or Hamwic.

Mike:
TGN is not a free source, so you have to be careful how to use it and not violate their license conditions. If you manually check some names then it will be ok, but it cannot be done in an automated process.

Best

Marc

[WWW]
codepoet



Joined: 23/05/2008 21:37:34
Messages: 26
Offline

Ok Marc, thanks for the clarification.
Best
Mike
ross_drew



Joined: 15/10/2010 10:49:34
Messages: 1
Location: Glasgow
Offline

I realise this is an old thread but I'm working on a project called CHALICE which is a historic place name Gazetteer for the UK academic sector. It's open source and can be found at:-

https://sourceforge.net/projects/chalice/

it's in the early stages but it's functional. Thoughts and comments on how this should be done and use cases would be helpful if you so wish.
[WWW]
yahalimu



Joined: 08/05/2007 23:30:26
Messages: 3
Offline

Hmmm. Interesting.

Will definitely check that out as I'm looking at a postcode history tour app for blackberry..

May be suitable..

Flexible chimney liners
speaklanguages



Joined: 11/08/2010 01:04:22
Messages: 11
Location: Estonia
Offline

It appears there are over 6000 place names in the database which include the word "(historical)" in the name. For example,

Edo (historical)

is given as one of the alternate names for Tokyo.

However, surely the "name" field isn't the right place to mark somewhere as being historical. "historical" isn't part of the name, it's rather an attribute which applies to the place. So is there any way the word "(historical)" could be removed from the name field for those places, and mark those places as historical in some other way?

Also, is there any possibility the "isFormerName" field could be opened up for editing? The current situation is a little unsatisfactory as there's no way of telling which place names are current or not. At least if the "isFormerName" field was open for editing, we would have a way to start marking place names as former names as we come across them. Then it shouldn't take long before the most well-known historical places have been correctly tagged.

Best wishes,
Chris
[WWW]
marc



Joined: 08/12/2005 07:39:47
Messages: 3759
Offline

Hi Chris

There are two new flags in the alternate name edit tool:
isColloquial for slang and colloquial names
isHistoric for names of the past that are no longer used.

Last week I was at the Annual Meeting of the Association of American Geographers. The topic was how GeoNames could/should be extended to improve support for the timeline. Some more attributes that were discussed:
- fromPeriod (date, year or period of usage)
- toPeriod
- source (book or map where the name is used), publication date of the source
- notes

The discussions for the other fields are still ongoing and any feedback is welcome.
The two new flags isColloquial and isHistoric are not yet included in the daily extract as we will probably add some more attributes and don't want to change the extract files too often. We will extract the records with the two new flags in separate files as a temporary workaround till everything is stable.

Best Regards

Marc

[WWW]
speaklanguages



Joined: 11/08/2010 01:04:22
Messages: 11
Location: Estonia
Offline

Hi Marc,

Thanks for the update. That's a very positive development.

Kind regards,
Chris
[WWW]
saphir



Joined: 05/06/2010 22:44:39
Messages: 123
Offline

Hi Marc

My opinion about the flags > Toponyms, alternate names and flags must be searchable. The old name for Tokyo “Edo (historical)” is searchable… for “Edo” and for “(historical)”…
This is not a specialized historian’s database. The general user who did not even hear about flags should see the facts in the toponym name. Everybody knows that Tokyo is the present name for the city - but what about an obscure town behind the yellow mountains…


isColloquial > that’s OK

isHistoric > OK for A and P feature classes, but even there, without a timeline, it is not more explicit than the (historical) in the toponym.
For all other feature classes I see no advantage for this. The toponym must tell what it is, e.g. “Abandoned Airfield WWI”, “Assault Beach WW2” or “Former Coconut Grove Washed Away”.

from/toPeriod > History is a science; precision is often relative to the state of the art at a given time. Who will arbitrate a war between historians?

source > Meaningful in a dissertation… but everybody can ask Google…and the war above will be here too > this guy is not a reference...

notes > Interesting, sometimes I don’t like to let the reader guess about the meaning of a forcefully concentrated toponym text.


Best regards

Urs

kholerik



Joined: 19/09/2011 16:43:07
Messages: 1
Offline

Hi,

As this thread has been stale for over a year, I'd like to bump it and ask for the current status.

I am in the situation that I have to like to no-longer existing countries like the Soviet Union, Yugoslavia or West and East Germany.

Did someone come up with a solution for this in the past 17 months?


Cheers,
K
 
Forum Index -> Discussion of GeoNames Toponyms
Go to:   
Powered by JForum 2.1.5 © JForum Team