City Name Generator

[davwms]

I was playing around with the idea for a fictional league, but had trouble coming up with fictional city names that were realistic but yet fictional. I found a few random generators online, one that was even pretty close to what I wanted and I based this one on that.

This program will generate a set number of city names with the option to export into a text fiile. Here's a sample of what it does...


Arborline
Aterswell
Beardale
Belhurst
Bluechester
Calmere
Coalman
Frebourne
Galark
Grand Brownsville
Highler
Kanlywell
Kentinglotte
Mountainlyn
Mountainson
North Gamage
Orchardons Springs
Ridgebourne
Springcott
Sterlingshore
Warlyn
Waterlyn
Wetherley
Windark
Woomere

There are several small data files that can be modified all you want.

[Muzamba]

This is quite a neat little program, davwms. Lots of potential here. Thanks for designing it and letting us download it.

I have a couple questions, though.

First off, is there a limit to the number of lines you can have in each dat file? I tried pasting in an extra few hundred entries and it gave me a runtime error when i tried opening it up again.

Also, will it accept duplicate entries? Having more than one of the same entry seems to be causing it to crash on startup as well.

[davwms]

Currently it allows for 500 entries in each file. I can easily allow for more, it was just a number I used when I started creating it...

You can have duplicate entries.. In my next version, I will have another modifier for each entry to allow for how common that entry will be used.. (I hope you can understand that last sentence LOL)

Here is version 1.1... I increased the number of entries allowed by quite a bit.. :-)

Also, please.. if you make changes/additions to the data files, please, pass them back to me so I can add them in for the next version..

thanks!

[ry1220]

Neat little program. Thanks davwms! I'll be modding to allow for some Spanish-sounding city names and will PM them to you or post them here, whichever you'd prefer.

EDIT: To be honest, the export to file should have a directory listing, at least, if not an option as to where to save to.

EDIT2: Also, it would be nice if populations could be entered or randomized and added to the export in the same format as cities.txt.

[Muzamba]

Originally Posted by davwms

Currently it allows for 500 entries in each file. I can easily allow for more, it was just a number I

used when I started creating it... ... I increased the number of entries allowed by quite a bit..

Fantastic!

However, for what I'm looking to accomplish, the number would have to be pretty big... probably somewhere in the neighborhood of 20,000. (I'll explain in a little more detail in a moment, but basically I want to add all the American names from the names.txt db file.)

Originally Posted by davwms

.. In my next version, I will have another modifier for each entry to allow for how common that entry will be used.. (I hope you can understand that last sentence LOL)

Understood perfectly! And this was actually going to be my second suggestion. That's why I was originally asking about duplicates, but using a frequency interger would definitely be much more efficient.

Originally Posted by davwms

Also, please.. if you make changes/additions to the data files, please, pass them back to me so I can add them in for the next version..

Before sending you my additions, I'd like to propose a few ideas I have.

Unfortunately, many of the names that are being generated still sound rather contrived and unrealistic. I think that's just a fundamental drawback to trying to conjoin parts of words into new ones.

So here are my suggestions that I think will give us even better, more realistic-sounding place names. (Note: If there are any English teachers reading this, I confess that I don't remember the correct terms for the grammatical elements I'm using here, so I'm just gonna wing it. )

1. Reorganize the dat files as follows:

city1 = Title prefix (Fort, Mount, New, Port, etc.)
city2 = Descriptive prefix (Big, Cold, Green, Running, etc.)
city3 = Subject (proper names)
city4 = Subject (nouns)
city5 = Conjunctive suffix (burg, field, ton, ville, etc.)
city6 = Title suffix (City, Lake, Park, Village, etc.)
city7 = Secondary title suffix (see "Title suffix")
(optional) city8 = States/provinces

(Quick sidenote here: Although the spacing format could justifiably be hard-coded, it might be better to allow spaces to be defined within the datasets instead.)

2. Control the number of dataset combinations that are generated.

For example:

3 - ("Smith") - 1000
3+5 ("Smithville") - 1000
3+6 ("Smith Falls") - 1000
3+5+6 ("Smithville Falls") - 50
3+6+7 ("Smith Falls City") - 50
3+5+6+7 ("Smithville Falls City") - 1
1+3 ("New Smith") - 250
1+3+5 ("New Smithville") - 250
1+3+5+6 ("New Smithville Falls") - 10
1+6 ("New Falls") - 5
2+4 ("Big Oak") - 250
4+5 ("Oakville") - 1000
4+6 ("Oak Falls") - 500
4+6+7 ("Oak Falls City") - 50
2+4+6 ("Big Oak Falls") - 50
2+6 ("Big Falls") - 100

Ideally, the user would be able to enter the frequency interger themselves, although hard-coding them roughly according to the examples listed above would probably be okay.

3. Use tags to group certain values together. Tags could be numbers, letters, or words... whatever the user wants to use to identify conceptually-similar values.

For instance, say you want to add "Saint" as a title prefix. You wouldn't want a place name of "Saint Smith" to be generated, as that would be unrealistic. Therefore, you might add the value "saints" next to the line for "Saint" in the city1 file. You would then add the tag "saints" to entries like Andrew, Michael, and Paul in the city3 file.

I'd also like to expand upon ry1220's ideas...

Originally Posted by ry1220

...I'll be modding to allow for some Spanish-sounding city names...

Spanish names could be effectively utilized by implementing the tag function I mentioned in suggestion #3 above. Using the same logic as with "Saint" above, one could include "San" in city1 and match it up with names like Andreas, Miguel, and Pablo in city3. You could also add descriptive prefixes like "La", "Las", "El", and "Los".

On the other hand, because Spanish has a different structure, it might be better to have a completely separate Spanish names file. That would allow the descriptive prefix and subject to be juxtaposed (e.g. "Rio Loco", instead of "Loco Rio")

Originally Posted by ry1220

EDIT: To be honest, the export to file should have a directory listing, at least, if not an option as to where to save to.

Agreed. This would make it easier to have separate export files.

Originally Posted by ry1220

EDIT2: Also, it would be nice if populations could be entered or randomized and added to the export in the same format as cities.txt.

This is a great point. Perhaps an arbitrary population and latitude/longitude coordinates could be generated. (Or the lat/long could be based on the states the cities are assigned to.)

I have several other ideas regarding the use of this program to generate other fictional entities for use in OOTP, especially if some of my ideas for future versions are implemented. These include company names, local and national broadcast media, and local print media (which could already be useful with our new storyline engine. )

But I'll save these for another post, so as to not overwhelm you with too many ideas just yet. LOL

Anyway, let me know what you think of these ideas and whether you think they're feasible. (FWIW, I'll be happy to help with the content if you'd be able to handle the coding.)

And thanks again for developing this. I'm looking forward to seeing what all we can eventually accomplish with it!

[davwms]

Originally Posted by ry1220

EDIT: To be honest, the export to file should have a directory listing, at least, if not an option as to where to save to.

Agreed...:-)

Originally Posted by ry1220

EDIT2: Also, it would be nice if populations could be entered or randomized and added to the export in the same format as cities.txt.

This was something else that I was also thinking about adding in. Maybe even a one button click to create a fictional world all ready for OOTP!

..but some of the city names that are produced are not very realistic. So, I need to work on that as well...

In my own little version, I had it creating state/province names that were being based on the names of the states in the U.S. Some of these were really cool but more of them were pretty ridiculous sounding. I would like the this program to be able to generate country names and states/provinces, as well.. or at least, allow the names to be entered and then generate city names for them...


Thanks for the feed back!

[davwms]

Originally Posted by Muzamba

However, for what I'm looking to accomplish, the number would have to be pretty big... probably somewhere in the neighborhood of 20,000. (I'll explain in a little more detail in a moment, but basically I want to add all the American names from the names.txt db file.)

I think I increased it to 50,000 for v1.2 :-) But being that this database is pretty simplistic, I could probably increase it to 200,000 or more without causing any problems.

Originally Posted by Muzamba

Unfortunately, many of the names that are being generated still sound rather contrived and unrealistic. I think that's just a fundamental drawback to trying to conjoin parts of words into new ones.

yup!

Originally Posted by Muzamba

1. Reorganize the dat files as follows:

city1 = Title prefix (Fort, Mount, New, Port, etc.)
city2 = Descriptive prefix (Big, Cold, Green, Running, etc.)
city3 = Subject (proper names)
city4 = Subject (nouns)
city5 = Conjunctive suffix (burg, field, ton, ville, etc.)
city6 = Title suffix (City, Lake, Park, Village, etc.)
city7 = Secondary title suffix (see "Title suffix")
(optional) city8 = States/provinces

(Quick sidenote here: Although the spacing format could justifiably be hard-coded, it might be better to allow spaces to be defined within the datasets instead.)

I like it. I knew that if this ended up being something that someone would want that I was going to have to change the dat format to be more precise. This will stop some of the absurd names. The very first version I put together from scratch in the matter of an hour. I made some minor changes and additions since then, but it was obvious that a complete over-haul was needed.

Originally Posted by Muzamba

2. Control the number of dataset combinations that are generated.

For example:

3 - ("Smith") - 1000
3+5 ("Smithville") - 1000
3+6 ("Smith Falls") - 1000
3+5+6 ("Smithville Falls") - 50
3+6+7 ("Smith Falls City") - 50
3+5+6+7 ("Smithville Falls City") - 1
1+3 ("New Smith") - 250
1+3+5 ("New Smithville") - 250
1+3+5+6 ("New Smithville Falls") - 10
1+6 ("New Falls") - 5
2+4 ("Big Oak") - 250
4+5 ("Oakville") - 1000
4+6 ("Oak Falls") - 500
4+6+7 ("Oak Falls City") - 50
2+4+6 ("Big Oak Falls") - 50
2+6 ("Big Falls") - 100

Ideally, the user would be able to enter the frequency interger themselves, although hard-coding them roughly according to the examples listed above would probably be okay.

Again.. nice.. Silly how simple it looks when you put it like you did. I was having trouble trying to figure out how to do this but you made it easy for me to see.. thanks!

Originally Posted by Muzamba

3. Use tags to group certain values together. Tags could be numbers, letters, or words... whatever the user wants to use to identify conceptually-similar values.

For instance, say you want to add "Saint" as a title prefix. You wouldn't want a place name of "Saint Smith" to be generated, as that would be unrealistic. Therefore, you might add the value "saints" next to the line for "Saint" in the city1 file. You would then add the tag "saints" to entries like Andrew, Michael, and Paul in the city3 file.

That's why I didn't add SAINT to my first setup.. LOL

Originally Posted by Muzamba

I'd also like to expand upon ry1220's ideas...

Spanish names could be effectively utilized by implementing the tag function I mentioned in suggestion #3 above. Using the same logic as with "Saint" above, one could include "San" in city1 and match it up with names like Andreas, Miguel, and Pablo in city3. You could also add descriptive prefixes like "La", "Las", "El", and "Los".

On the other hand, because Spanish has a different structure, it might be better to have a completely separate Spanish names file. That would allow the descriptive prefix and subject to be juxtaposed (e.g. "Rio Loco", instead of "Loco Rio")

Yes, I believe it would require seperate dat files, but this is conceivable and probably not that difficult.



Thanks for your input! Definitely helps guide me to what I want this program to do and how to get it to do that. :-)

[Scott Vibert]

The more I hear the ideas here the more I like it. Can't wait to see the next version.

[davwms]

I rewrote the majority of the code and here's what is being generated at this point...

I need to add a lot more common nouns. Currently I only have about 200..

I have 7 data tables that the program pulls from.. pretty much just going with Muzamba's suggestion here...(including the number of items currently in the data lists)

1) - 6 - Prefix (Title)
2) - 15 - Prefix (Descriptive)
3) - 15,263 - Subject (Proper Names)
3a)- 51 - Subject (Proper First Names - mainly used with "St." prefix
4) - 150 - Subject (Common Nouns)
5) - 57 - Suffix (Conjunctive)
6) - 9 - Suffix (Title)


Here's a sample of 100 randomly generated city names: (Still too many Points and Ports, and Beaches and Groves.. stc.. :-)

Anderson View
Angertown
Animal Falls
Avenel Grove
Ball View
Beauty View
Benz
Big Village
Blue Springs
Bullis Point
Car Point
Cassels Springs
Central Grove
Chair Springs
Chance Heights
Chance Point
Chief Beach
Christianway
Christopherwell Grove
Clark
Cold Heights
Company Beach
Condition Beach
Corney
Dance Falls
D'Anneville Park Grove
Dictionary Park
Doironwell Springs
Drop View
Ease Falls
East Beach
Edington Falls Heights
Enemy Park
Evansapolis
Featherville
Fort Botelertown Village
Fort Markhammere
Fritsch View
Grand Beach
Green Falls
Green Heights
Green Beauty Point
Green Color
Inskipland
Kerwin Springs
Konigside Falls Village
Krieger Beach
Lake Village
Lambage Heights Point
LaPointe
Leasktown
Leedsfield
Little Child View
Livermoreard
Mabeyster
MacBeolain
MacLymont Falls
McCavanagh Grove
McCreight Beach
O'Furry
Ottinger
Pastondale
Peiper Park
Pink Beach
Point Bindermere
Point Coughlanage
Point Heights
Point MacKnightport Falls
Point Toddapolis Heights
Port Denbighmark
Port Dorisford Village
Port Fifettmark
Port Hallland Park
Port Loeschcaster
Port Storerley
Port Village
Pryor
Roofe
Saintmount
Savile View Falls
Scaifemore Beach View
Schantz View
Sherston
South Grove
St. Robert
Tinsdaleway View
Trevenot
West Beach
West Eight Springs
Wilkinsdale View

[Muzamba]

These are already looking a LOT better.

Good call adding the first names category, btw.

I think for the nouns, we need to look at the types of words that were in common use in the 18th and 19th centuries, when most US and Canadian towns were settled. Many were translations of Native American words, which is why there are a lot of animals and other terms pertaining to nature. Pioneers also seemed to name landforms after objects that they resembled and subsequently named adjacent settlements likewise.

I've got quite a few additional entries for 1, 5 and 6 (7) for you that I can post here. They don't have frequency intergers yet (not sure exactly how you need them formatted) but I included duplicate copies of those which should appear most often. (This should help cut down on the occurences of title suffixes like "Point" and "View") And of course I've yet to apply tags.

Note: I had to change the extensions to .txt in order to post them here, so they'll obviously need to be changed back to .dat

[davwms]

UPDATE

I added in a lot more to all of the different data files.. including Muzamba's lists..

Now the totals are: (I tried duplicating the first two files to see how that works but I will be splitting them up with just some entries being duplicated in both files)

1) -33- Prefix (Title)
2) - 33 - Prefix (Descriptive)
3) - 15,263 - Subject (Proper Names)
3a)- 51 - Subject (Proper First Names - mainly used with "St." prefix
4) - 1157 - Subject (Common Nouns)
5) - 57 - Suffix (Conjunctive)
6) - 78 - Suffix (Title)


and here is a list of 100 generated results:

Arrol
Atkinson Heights Bridge
Big Settlement - Must have been named after a divorce hearing...
Blue Limit North
Brother Corner
Brown
Browning
Cape City - Where every super hero gets their capes!
Central Center - Seriously.. it's right in the middle!
Central Howardmill Pines
Changetree
Charles Center Mountain
Chick Butte - Ha! A little change in the spelling and this would have been a popular place!
Child Woods - When he was younger, James was simply known as 'Child'
Claxton Woods Township
Cleveland - Rocks!
Cold Heighwayton
Cold Stidhamhill Centre
Cost Flats
Cough Heights
Digestion North - OK.. I need to remove the word DIGESTION.. LOL
Fairies Square
Feeling Farms - Sounds like a hippie commune to me!
Flashman - Before Batman and Spiderman, there was Flashman!
Fort Smithbank Glen
France - Hmmm... never heard of it!
Gallagherchester - Watch out for the exploding watermelons!
Glen Brinkside
Gordy Center
Gormley
Gourleyron Ridge West
Grand Canyon - Yes, it is!
Grand Lake
Grand St. Georgelyn Butte
Green Hamel
Gregg
Hargroveburg Rock
Hermannage Hook View
Hilderford
Home Center
Hot Bart
Hot Hill
Irwinshire
Jenkinslotte South Point
Jenniffhurst
Jones
Kaye Prairie Valley
Kindlehurst
Kitehurst
LeMay
Lemonver
Little Bend
Little Park
Little Tunnellmond Lake
MacDowalton Bluff
Mapstoneapolis East
Marble East
McClellanebourne
McVicker Prairie
Mint Prairie - Sounds like it came from the game "Candyland"
Moneyapolis - Next to Boardwalk in Monopoly.
Mount Gardens
New Raffertyster Spring
North Moon Mines
North Rapids
North River
Nyebourne Village Fork
O'Duvanyhurst View Island
Olsenville Point
O'Radaghan
Peck Butte Wells
Pen Square
Port Schmitzline Rocks
Red Rayburnver
Revelslotte
Revill
Robbinsline Rock
Roberts
Sayersage
Scallymill
Schaferburg
Shoffner
Shriver Ridge - We'll head them off at Shriver Ridge!
Shumaker Settlement
Side Lakes
South Wallacever
Sparkesver River
Starkeyster
Stricklandster Meadows Center
Taylorbridge
Upper Hook
Upper Winfieldcrest Corner
Waltonmond - After the explosion on the mountain... (Walton's Mountain)
Weightmanmyra
Wier Bluff
Work Grove
Yellow Bluffs
Yellow Friend - Your friend is a coward.. He's yellow!
Yellow Pines

[Scott Vibert]

Nice. Some are pretty amusing, but they are certainly plausible.

[Cryomaniac]

Having seen some US place names, Digestion is a perfectly reasonable thing to include.

[Tony M]

Originally Posted by Cryomaniac

Having seen some US place names, Digestion is a perfectly reasonable thing to include.

Really, you'll be telling me next that a town in New Mexico would change its name to Truth Or Consequences!!!

[smithtm]

This is a pretty sweet mod you've come up with davwms! I'd like to try out your latest version for a league I want to put together, but don't see it. Could you post it please? Much appreciated! I've been wanting to start a fictional league and I like the idea of generating fictional cities for my teams.

[Cryomaniac]

Originally Posted by Tony M

Really, you'll be telling me next that a town in New Mexico would change its name to Truth Or Consequences!!!

[davwms]

The 2nd version is not 100% complete but it works as is. I still haven't set up the "Save as..." dialog. It still exports as a file named CityList.txt in the same directory that the .exe is located.

I (hopefully) made it look better, though. :-)


the current version, we'll call it 2.0a, can be downloaded HERE!

Thanks!

[smithtm]

Woo Hoo! Thanks davwms, I'm really looking forward to playing around with this tonight to set up my fictional league!

[Muzamba]

It's definitely getting there.

Here are my observations on the most recent build.

• While the results are more structurally-realistic than they were in the first build, I think the table content is causing a lot of the output to sound rather silly and implausible. The primary culprit is Table4. The nouns really need to be toned down considerably. Relatively few of the words in this list would have been used in naming locations during the era that most American cities were founded.
• Some sort of tagging system still needs to be applied in order to better match concepts. This is especially true of Tables1 and 2 combining with Table3. For instance, the more descriptive prefixes like "Blue" and "Cold" would normally be followed by a noun from Table4, but not a name or proper noun from Table3.
• Even though they're not very common, there should probably still be a second title suffix table (Table7?) for cities like "Salt Lake City", "Thief River Falls" and "Hot Springs Village".
• Adding a second set of title prefixes, however, was a smart move. I forgot about places like "North Little Rock", "New Port Richey" and "Fort St. John". (However, they still need to be grouped appropriately with tags.)
• This isn't a big deal if it's too inconvenient or impractical, but would it be possible to allow the user to control the frequencies of table combinations? (I've always been of the mindset that the more the user can customize, the better.)
• Finally, on a personal note, is there any way you code the new GUI to accept Windows colors? I'm sure this won't be a problem for most, but because of my Windows settings, the list output shows up as white text on a white background on my computer. I can see them when they're highlighted, but not otherwise. (I have my colors inverted due to a vision problem.)

One thing that I think would really help, once all the programming has been completed, would be to compile statistics on place names. Some of the numbers still seem a little off to me, but I'd rather use accurate frequencies than to speculate on the numbers myself.

Anyway, like I said, this program is coming along nicely. A few more tweaks and content revisions and I think it'll be perfect!

Thanks again and keep up the great work!

[davwms]

Originally Posted by Muzamba

[*]While the results are more structurally-realistic than they

were in the first build, I think the table content is causing a

lot of the output to sound rather silly and implausible. The

primary culprit is Table4. The nouns really need to be toned

down considerably. Relatively few of the words in this list

would have been used in naming locations during the era that

most American cities were founded.

I pretty much created the content for Table4 by copy-pasting

from a list of the most common nouns in the english language.

I intend to edit the list at some point, but wanted to get this

version working before I did that. So, this issue was already

known and expected.

Originally Posted by Muzamba

[*]Some sort of tagging system still needs to be applied in

order to better match concepts. This is especially true of

Tables1 and 2 combining with Table3. For instance, the more

descriptive prefixes like "Blue" and "Cold" would normally be

followed by a noun from Table4, but not a name or proper noun

from Table3.

I duplicated the contents in Tables1 and 2 just to get more

options. What I see is that I need to keep them separate, but

have the option of using Table1 twice for the same city name..

113 as opposed to 123... there is nothing wrong with using the

same table twice, but I wouuld make sure to not duplicate the

same word.. (There won't be a Fort Fort Smith)

Originally Posted by Muzamba

[*]Even though they're not very common, there should probably

still be a second title suffix table (Table7?) for cities like

"Salt Lake City", "Thief River Falls" and "Hot Springs

Village".

I think this is already able to happen... but I will make

sure..

Originally Posted by Muzamba

[*]Adding a second set of title prefixes, however, was a smart

move. I forgot about places like "North Little Rock", "New Port

Richey" and "Fort St. John". (However, they still need to be

grouped appropriately with tags.)

What I think I am going to do is keep it with 2 seperate lists.

Table1 would be "Title" prefixes (Port, Fort, etc...) and

Table2 would be "Descriptive" prefixes (Little, Blue, etc...)

and there should be the ability for them

Originally Posted by Muzamba

[*]This isn't a big deal if it's too inconvenient or

impractical, but would it be possible to allow the user to

control the frequencies of table combinations? (I've always

been of the mindset that the more the user can customize, the

better.)

I plan on adding this in later. Right now, you still can make changes by editing the files, but I would to add this ability right into the program.

Originally Posted by Muzamba

[*]Finally, on a personal note, is there any way you code the

new GUI to accept Windows colors? I'm sure this won't be a

problem for most, but because of my Windows settings, the list

output shows up as white text on a white background on my

computer. I can see them when they're highlighted, but not

otherwise. (I have my colors inverted due to a vision problem.)

I should be able to come up with something, even if it is only the ability to change the colors.

Originally Posted by Muzamba

One thing that I think would really help, once all the

programming has been completed, would be to compile statistics

on place names. Some of the numbers still seem a little off to

me, but I'd rather use accurate frequencies than to speculate

on the numbers myself.

This could be a chore. The only thing I can think of would be to get a list of every city/village in the United States and then search for how often each part shows up ? But then you also have cities that are named after something from another language, like an indian name for instance.


Thanks for your input, I probably wouldn't have even gotten this far if not for your suggestions and enthusiastic responses..