What you want would be a context-based generator I think. That means, the next syllable is based on the previous one (for a very simple example using Markov Chains). This is quite a bit more difficult to have as data than this simple construction.
The best way I can come up with would be some sort of data aligned like this:
"syllable" => { possible next syllables }
A simple example would be "To" : [ "m", "ni", "da" ], "m" : [ "on" ], which could produce Tom, Toni, Toda and Tomon. This is a simple version; if you want to have real usages (using the chains as mentioned above, i.e. that some syllables appear more often after a certain one) the file becomes increasingly complex:
"second to last syllable" : { "last syllable" : { "new rare syllable" : 1, "new common syllable" : 99 } }.
This is just a chain (not really Markov, I can’t remember its name) over the last two syllables (it’s possible to do it over just one, of course), but you get the idea. It will really bloat your file up and it’s almost impossible to generate something like that from hand (with a text editor).
That’s what I planned initially with only single links though. I made mine with morphemes though but syllables would probably work better for consistency, it’s a pity there isn’t just a single (complex) method for defining how a language sounds. The problem being that every language has it’s own set of rules on what makes suitable names, something like the ascendency seems like it might use the patronymic/matronymic surnames or names based off where they’re from. From the respective time periods it’s similar to surnames weren’t even that common, it’s frustrating really as something like nihonjin or rayya’s children would probably have a completely different system, I don’t know as much about naming patterns from those regions.
If we could have some way of defining multiple systems and making it choose in the files perhaps? It’s a fairly tricky issue.
I personally don’t think you should use Indian names for Raya’s Children as I don’t feel that Raya’s Children are drawing influence from India (perhaps a bit with the trading part). For me they seem to be more a mix of Buddhism and some of the Middle Eastern countries in a bygone age.
The names at least feel more middle eastern to me, you’d need a dev to tell though. The appearance definitely seems to be asian based though, really the only people who know are the devs. The northmen seem more nordic, and the ascendancy more english? It’s all a bit difficult to tell.
Two hours later, done! I’ve separated the name generator and the faction proxy, it just means that you have a bit more freedom at the moment. The faction proxy is required to “fake” to the game that we are that faction (which we are not). To be bluntly honest, the faction design in the current game is a mess (to get a faction you also need to provide a path to its manifest - why? I would much more prefer if we specified the faction name à la stonehearth:factions:ascendancy and have the game figure out if it has been loaded or not this way instead of “civ”).
Alright, enough ranted about the current state of the game. To get started, download rp_alternate_naming.smod and rp_proxy_factions.smod. You can find example config files that do the whole magic as usually on GitHub. To get nihonjin instead of ascendies, simply replace the kingdom in both config files (proxy_factions and alternate_naming).
@phoriist: I’m going to keep nihonjin.json in the current repos, but it’s going to be deleted when 1402 hits. You can now easily make your own mod folder that takes care of this stuff - without modifying configs or anything. I’ve created a template for you that you can use to get started. This template will replace the normal faction with your nihonjin (which would also allow you to use different workers and all that stuff - whatever the faction json file does).
That means, I’m not going to change (or merge) anything that affects nihonjin.json in my rp_alternate_naming. If you want to tweak ascendies.json however, be my guest :D. I think it’s better (and fair towards others) if my mods are kept as minimal as possible (and as neutral as possible).
I definitely imagine that this is something that will clear up and become a lot more streamlined as the team go forward and have the different factions implemented? Like most things I suppose! Such is the dilemma of modding this early on
In other news it’s great to see what people are doing this early on, definitely gets me excited for the future!
Every single person may have different feelings. My feelings about the descriptions gave me those ‘real world locations’ of Kingdoms:
Yes, maybe not all of India, but at least some of them (Buddhism has been created in Northern India) coloured in Paint. Green - Ascendancy, Blue - Alliance, Red - Children
It’s like Scandinavia and Western Europe, I think.
It’s @Avairian’s Nihonjin. I am only trying to help him with names.
And don’t worry - I have this one already on PC.
…ascendies.json, northmen.json and rayas.json - remember, we will have Three Kingdoms in Stonehearth (only one is playable at the moment), so it’s better to be prepared
True enough, but what I meant was this: ascendies is basically ascendancy using the alternate name generator (i.e. my generator). It’s the only file that I am going to bundle with my mod; if you want to create alternate name generator data for the other factions then go ahead! rp_alternate_naming will only contain ascendies, however, because I think as an example JSON it does quite well.
Well I downloaded the template and took a peek at the files. Not exactly sure what’s going on with the code but it looks all fancy. I appreciate the work everyone here is putting into the name generator!
Just took a look at some of those lua files and the lua files from RP. I so have a long way to go for programming, I don’t understand any of it. xD Still looks fancy.
I appreciate the work everyone here is putting into the name generator! 