Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[GoblinCookie]

No, [inherited surnames] creates problems irrespective of what my opinion on the subject is.  The present system works very well to identify individuals out of lists, if we replace it with the surnames system the OP proposed then we lose that functionality.

I take it you're again referring to the fact that the RNG often gives different dwarves the same 1st name, and so the 2nd & 3rd names are required to help tell those dwarves apart. Fair enough--but you seem to be forgetting the very simple tweak of telling the RNG to just not give a newborn any name that's already being used by someone in the fort. Being unwilling to consider an easy fix, and instead painting the whole idea as "greatly inferior", does indeed smack of prior emotional commitment on your part. In my opinion, you're clinging to the "1st names are unreliable identifiers" status quo because it supports the "need" for your desired system of longer names.[/quote]

As I said before, the ability to fix a problem that you needlessly created for no gain is not an argument for adding it in.  Why would I be 'emotionally invested' in a system of naming? The people proposing this concept are emotionally attached to the idea of making surnames work like reality, that is the problem; it does not even work very well in reality. 

By making all first names unique, what we are in effect doing is getting rid of first names and replacing them with the present surnames.  The problem with doing that is the present first names also have a purpose; they allow us to identify culture and if we get rid of the present first names in effect by giving every individual a unique first name then it becomes more difficult to tell what culture any individual comes from than it presently is.  With further development we can make it so that first names change over time in a culture, so first names allow us not only to identify culture but what era the person comes from. 

As I recall, you brought up this straw man in the last thread on this topic as well. Something along the lines of, "This plan has a statistical chance of what I consider a failure, therefore the plan has already failed and you should abandon it entirely." I believe that at the time, my reply was something like "Yes, that happens in real life too. A couple of distantly-related dwarves having the same last name can be a minor difficulty . . . but a difficulty is not the same as a flaw."

It is when you are replacing a system that has near 100% accuracy with a system that will in many cases have an accuracy of less than 50%. 

Oh, it makes a hell of a difference if you stand to inherit. If you were (somehow) in the direct legitimate line of Elizabeth I, for instance, that would literally make you (or your father/grandfather) the rightful King of England. You do have a point about bloodlines mixing, but bloodlines aren't names. If everybody's name is Durin, then yes, the whole question becomes moot--which is why not all of his descendants got to keep his name.

Remember there are literally thousands of other people in the same inheritance position as you are.  Hereditary systems simply break down when you end up picking some random peasant out of thousands of other peasants and make him king.  It is like a reducto-ad-absurdum for that kind of system, the point at which it stops making any sense, as there is neither any special genetic or indeed a household connection between the descendant and his ancestor.

In history nobody has ever cared about extremely distant relations inheritance rights.  Their interest tends to stop at about first cousins and aunts/uncles for exactly the reasons I was referring to; it becomes absurd beyond that point.

Point of terminology here: Only a name specifically shared with other family members can actually be called a surname. So your dwarf would have name elements [1] [23] for a three-word given name, and then elements [45] for a surname.

A surname is often referred to as a last name.  But yes, in this case the last name is not a surname.  You summed my idea up well though, in effect we have a non-unique first name, a unique middle name and an inherited surname that is replaced each generation.

True, many animals produce absolute swarms of offspring--as a counterbalance to offset the fact that the vast majority of them will not survive to be able to reproduce. That works for them, because the forces they're working against--hunger and predation--ensure the evolution of the species through the survival of the fittest. Not so with dwarves, because the forces arrayed against them are their own damn social mores about love and sex. From what I've seen, only roughly 5% of dwarves born in the fort will ever have children; not because of goblins or forgotten beasts, but for pitiful reasons like being 11 years apart in age, or having differing views on purring maggots, or having briefly been in love with someone who died 40 years ago. Every dwarf fort is populated primarily by walking Darwin Awards, and in their case the result is NOT a more rigorous gene pool, but instead some very intense inbreeding, as in only a handful of generations you're left with nothing but the same 2 groups of cousins.

Exactly, the majority of the creatures do not get to reproduce, on account of getting eaten; so my point is quite correct.  Thing you are not considering is the exact same thing (nearly everyone gets eaten) can happen in Dwarf Fortress can't it, especially in particularly nasty worlds, which will be a thing fairly soon.  If a forgotten beast turns up and eats most people, then we end up with the exact same situation due to random factors that presently exists due to the weird sexual repression that dominates the current DF world.

Why should it have to be separate? If the dwarf who earned the name feels that it's more prestigious than her link (which might be tenuous) to her famous ancestor, then she might see fit to [/i]replace[/i] her (or her kids') ancestral name with her own combat title.

Because that would imply introducing a dysfunctional system into the game.  The system is only functional if it is kept separate from other information-bearing words, since it carries different information.  It is also of limited value once the distinguished individual that created the name has died, since we are only interested in their fairly immediate family.

[Starver]

Family names should be inherited by one of the parents

Direction that in go not does time...

[dragdeler]

This is what happens when you try do explain something complicated in a foreign language over lunch break.

I chuckled at it then proceeded to edit "civ" into "races" in order to be accurate, duh... so good morning everybody 

[Dorsidwarf]

I already get dwarves with the exact same name popping up occasionally, when it happens I usually just rename them to something sinilar, so it’s not like the “random gibberish” system is flawless

[GoblinCookie]

I already get dwarves with the exact same name popping up occasionally, when it happens I usually just rename them to something sinilar, so it’s not like the “random gibberish” system is flawless

It's nearly flawless.  You can still break it however if you put enough restrictions on the number of available names in the entity file so that the range of available words is brought down to relatively few.  The larger the number of words the exponentially lower the probability of getting the same name.

[SixOfSpades]

As I said before, the ability to fix a problem that you needlessly created for no gain is not an argument for adding it in.  Why would I be 'emotionally invested' in a system of naming? The people proposing this concept are emotionally attached to the idea of making surnames work like reality, that is the problem; it does not even work very well in reality.

And as I've said before, family names do not provide zero gain, multiple users seem to think they would be pretty handy. Every real-world culture that I can think of (except those small enough to give each person only one name) has also come to the same conclusion. Why do you think that is?

By making all first names unique, what we are in effect doing is getting rid of first names and replacing them with the present surnames.

Um--what? You yourself stated that shared first names was a problem (essentially "requiring" the other two names to make an exact match virtually impossible), and yet now you appear to be opposed to solving that problem? Am I reading you right? I don't really understand what you mean with the second half of your sentence.

The problem with doing that is the present first names also have a purpose; they allow us to identify culture and if we get rid of the present first names in effect by giving every individual a unique first name then it becomes more difficult to tell what culture any individual comes from than it presently is.

By 'culture' do you mean race, as in being able to tell a dwarven name from an elven one by sight? Either way, I never suggested that the RNG should make up all-new word on the spot and assign it as a dwarf's first name, I meant only that when a dwarf is born, it should just keep pulling first names until it hits one that isn't already in use. (Besides, the elven and dwarven languages each use letters that the other one doesn't, so telling them apart is pretty easy.)

With further development we can make it so that first names change over time in a culture, so first names allow us not only to identify culture but what era the person comes from.

This suggestion is totally unrelated, why do you bring it up?

As I recall, you brought up this straw man in the last thread on this topic as well. Something along the lines of, "This plan has a statistical chance of what I consider a failure, therefore the plan has already failed and you should abandon it entirely." I believe that at the time, my reply was something like "Yes, that happens in real life too. A couple of distantly-related dwarves having the same last name can be a minor difficulty . . . but a difficulty is not the same as a flaw."

It is when you are replacing a system that has near 100% accuracy with a system that will in many cases have an accuracy of less than 50%.

"In many cases" it's less than 50%? Is that anything like, "sixty percent of the time it works, every time"? And where exactly are you getting this data, on a DF suggestion that has never been implemented?

Oh, it makes a hell of a difference if you stand to inherit. If you were (somehow) in the direct legitimate line of Elizabeth I, for instance, that would literally make you (or your father/grandfather) the rightful King of England. You do have a point about bloodlines mixing, but bloodlines aren't names. If everybody's name is Durin, then yes, the whole question becomes moot--which is why not all of his descendants got to keep his name.

Remember there are literally thousands of other people in the same inheritance position as you are.  Hereditary systems simply break down when you end up picking some random peasant out of thousands of other peasants and make him king.

"Some random peasant" who just happens to be the foremost scion of a senior branch of the family, and who (miraculously) has the historical documents & experts in inheritance law to prove it. Technically, I don't think any two people are ever in the exact same position to inherit--even with identical twins, they still record which one came out first.

In history nobody has ever cared about extremely distant relations inheritance rights.  Their interest tends to stop at about first cousins and aunts/uncles for exactly the reasons I was referring to; it becomes absurd beyond that point.

The Hundred Years' War was fought over the King of England's birthright claim to the throne of France. By its conclusion, royals were still debating the succession rights from seven generations prior. And even today, one of the fundamental differences between Sunni and Shiite Muslims is over which of Mohammed's relatives was his true successor as leader of the faith.

True, many animals produce absolute swarms of offspring--as a counterbalance to offset the fact that the vast majority of them will not survive to be able to reproduce. . . . Every dwarf fort is populated primarily by walking Darwin Awards, and in their case the result is NOT a more rigorous gene pool, but instead some very intense inbreeding, as in only a handful of generations you're left with nothing but the same 2 groups of cousins.

Exactly, the majority of the creatures do not get to reproduce, on account of getting eaten; so my point is quite correct.  Thing you are not considering is the exact same thing (nearly everyone gets eaten) can happen in Dwarf Fortress can't it, especially in particularly nasty worlds, which will be a thing fairly soon.  If a forgotten beast turns up and eats most people, then we end up with the exact same situation due to random factors that presently exists due to the weird sexual repression that dominates the current DF world.

Yes, if the fort falls, the dwarves inside will most likely fail to reproduce, I considered that all right. My point was that dwarves will still reliably fail to reproduce even if the fort DOESN'T fall. Out of 200 dwarves, maybe 10 of them will get married, and maybe 1 of those marriages is homosexual, so you're probably looking at 4 productive couples--who together churn out enough babies to keep the population right at the cap. No migrants can enter, so each child's dating pool is restricted to kids from the other 3 families (on top of all the rules they faced already). With such a reduced pool, the odds of finding a mate becomes very small--there will still be a few successful pairings, but as each family of children drops out of the running, the number of eligible mates becomes lower and lower--because a larger and larger percentage of the fort's inhabitants are one's own siblings. The number of reproducing families will inevitably shrink to 2, then 1, and finally none. Ironically, the only way a dwarf fort can survive in the long term is by failing, reducing the population below the cap so migrants can bring in new blood.

Why should it have to be separate? If the dwarf who earned the name feels that it's more prestigious than her link (which might be tenuous) to her famous ancestor, then she might see fit to replace her (or her kids') ancestral name with her own combat title.

Because that would imply introducing a dysfunctional system into the game.  The system is only functional if it is kept separate from other information-bearing words, since it carries different information.  It is also of limited value once the distinguished individual that created the name has died, since we are only interested in their fairly immediate family.

I think I should clarify--there's a reason I said replace. Consider the dwarf Momuz Thunderbreeches, granddaughter of Stakuz Thunderbreeches. Due to her dual professions of Butcher and Axedwarf, Momuz earns the combat title "Slayer of Geese". She then bears a son, whom she names Datan Gooseslayer, because she figures any name related to a combat title is better than being associated with someone famous only for his flatulence. So yes, your response is largely correct--Datan Gooseslayer might very well have elder siblings already named Thunderbreeches, creating a small but noticeable dysfunction. As long as Momuz herself remains alive, with her name & title acting as a reminder that Thunderbreeches and Gooseslayer are the same family, that's largely okay, but after her death it's purely up to the player to remember . . . unless other individuals decide to keep the clan intact under the same name, unilaterally changing their own surnames to Gooseslayer (or back to Thunderbreeches).
I wasn't implying that combat titles themselves be inherited--for instance, I never meant that the son should be named "Datan Thunderbreeches, Slayer of Geese".

[dragdeler]

At the moment the game has the following name structure:

{a} {b}{c} the {e} of {f}

All variables are proper nouns choosen randomly to create a name. I suggest we leave it that way, but insist that all creatures, items, minerals and other things DF features, should be included in every language of every race.





My actual suggestion is:


We add suffixes to every language, that are not choosen randomly out of a pool of proper nouns, but always the same in every world (but different for each race). Now those pieces of gibberish are applied to {e} and {f}, in order to signify different meanings, such as: "comes from", "killer of", "creator of", "inheritor of" (...).

The {e} and {f} part of a name are what I call titles. Titles are gained after important life events, and the game consistently applies the same titles to different people if it relates to the same thing. It goes without saying that different races should place more emphasis on different kinds of titles, according to their cultural identities. Which brings me to {b}{c}.

The {b}{c} part of a name are what I call familynames. So evidently they should reflect ancestry. Whether that would be the name of the mother, the father, the most famous ancestor or some other thing should be defined by the races. Some races should have different civilisations feature different solutions to that problem. All that according to the cultural identites of those races, of course.

If the game keeps track of all the meaningful names and prohibits them from being reused, and you also have meaningful titles that relate to some historical or geographical fact, we should be able to spot meaningful links in a glance while using the UI and searchfilters as they are today.

And I think that leaves T1 enough leeway to fiddle some procedural mythgen into it, so that there is even more meaning in those words (that are still randomly picked). Also, if false identities keep track of meaning in order to allow distinct lies, that would be great.

[SixOfSpades]

I suggest we . . . insist that all creatures, items, minerals and other things DF features, should be included in every language of every race.
We add suffixes to every language, that are not choosen randomly out of a pool of proper nouns, but always the same in every world (but different for each race). Now those pieces of gibberish are applied to {e} and {f}, in order to signify different meanings, such as: "comes from", "killer of", "creator of", "inheritor of" (...).

I'll generally support the second half of that, but also caution against getting involved with the language. It's a huge can of worms unto itself, and in short it's nothing but a placeholder--there's little point in suggesting improvements on a placeholder.

Another thing to worry about is the length of your {e} and {f}: Artifacts, creatures, and locations can have names that are quite long, and if they get folded into a dwarf's regular name in all menus, there's your truncation problem again. Titles should probably only appear in screens wide enough to properly show them, like the main Units menu or of course the individual dwarf's Thoughts page.

Dwarves should be limited to one title each; if they earn a second one for whatever reason, they have to choose which one to keep. They can also start a maximum of one new clan: Take Momuz Thunderbreeches, from my previous post. Say she names her kid Datan Gooseslayer after herself, and then makes an artifact bone figurine. Because she already started the Gooseslayer clan, she should not be allowed to replace her "Slayer of Geese" title with "Carver of Figures", or name her future children Figurecarver. I'm all for successful dwarves being able to found their own clans, but let's not take it too far.

[dragdeler]

Another thing to worry about is the length of your {e} and {f}: Artifacts, creatures, and locations can have names that are quite long, and if they get folded into a dwarf's regular name in all menus, there's your truncation problem again. Titles should probably only appear in screens wide enough to properly show them, like the main Units menu or of course the individual dwarf's Thoughts page.

Not if you keep assigning them randomly and at one word lentgh (as they are) and simply make a mention in legends or announcements such as: "Urist Abbeyfloor has gained the title of towerkar in honor of his creation of the artifact x" . If add cull familynames, cull title {e}, cull title {f} to the relevant menus and apply them all, you're left with "firstname nickname".

Yeah I'm just considering two titles because we currently have the option to create such names. But since save compability is negligible with the next big update all doors are open. And as you hinted the real difficulty lies in deciding whether titles are permanent or what changes them. My first attempt would be to place the tresholds prohibitely high, or place random timers on creatures that will initiate their "title-fest" to choose the strongest or rarest one. But surely there are more elegant and situation appropriate solutions.


ps: about the random part; just imagine them obscure references and puns that don't translate well into english

[GoblinCookie]

And as I've said before, family names do not provide zero gain, multiple users seem to think they would be pretty handy. Every real-world culture that I can think of (except those small enough to give each person only one name) has also come to the same conclusion. Why do you think that is?

Recall I am not proposing that we do not have family names, I am just proposing that we do not simply copy the systems that are used in real-life but come up with a better system.  The family name system in real-life works extremely poorly, especially under certain conditions, yes it does work better than having no family name system at all, but that may only be if you consider a false conclusion better than no conclusion. 

A lot of things real-life cultures do is quite imbecilic, in fact I would say the majority of things that real-life cultures do is such.  So if you want things to work properly, blindly copying real-life cultures is definitely not the way to go.

Um--what? You yourself stated that shared first names was a problem (essentially "requiring" the other two names to make an exact match virtually impossible), and yet now you appear to be opposed to solving that problem? Am I reading you right? I don't really understand what you mean with the second half of your sentence.

It is not a problem at the moment because there are no family surnames.  It becomes a problem if we replace the present surnames with family ones, since we will end will a large number of people with the same name.  This is on top of the general problem of the dis-functionality of such surnames at identifying family relationships. 

If we intend to solve this manufactured problem by increasing the number of first names, then that is in effect the same as simply replacing them with the present double-string surnames.  This raises the question of what the point of first names is at all and the answer is that it allows us to easily identify the culture of our individual. 

By 'culture' do you mean race, as in being able to tell a dwarven name from an elven one by sight? Either way, I never suggested that the RNG should make up all-new word on the spot and assign it as a dwarf's first name, I meant only that when a dwarf is born, it should just keep pulling first names until it hits one that isn't already in use. (Besides, the elven and dwarven languages each use letters that the other one doesn't, so telling them apart is pretty easy.)

By culture I mean culture, a goblin born in a dwarven civilization, or a non-historical goblin promoted into a historical characters gets a dwarven name.

We should do the opposite of making first names unique.  Having identified that the only function of first names is to identify culture, we can make everyone in the culture have the same 'first name'.  That makes things work a lot better, since we have a shared name that tells us what culture the individual belongs to, one shared name telling us what family household they belong to and a third unique name identifying the specific individual. 

This suggestion is totally unrelated, why do you bring it up?

Because it is a possible development of first names in the future. 

"In many cases" it's less than 50%? Is that anything like, "sixty percent of the time it works, every time"? And where exactly are you getting this data, on a DF suggestion that has never been implemented?

I'm not getting the data from anywhere, I am just guessing.  I know it is going to work poorly because I can think critically, I also know that sometimes it will actually work and that probability of that is going to modified by circumstances.

"Some random peasant" who just happens to be the foremost scion of a senior branch of the family, and who (miraculously) has the historical documents & experts in inheritance law to prove it. Technically, I don't think any two people are ever in the exact same position to inherit--even with identical twins, they still record which one came out first.

Nobody actually cares if you are one of several thousand descendants of a defunct ruler.  If the rulership is not defunct, then the pool of heirs is going to be a lot smaller. 

The Hundred Years' War was fought over the King of England's birthright claim to the throne of France. By its conclusion, royals were still debating the succession rights from seven generations prior. And even today, one of the fundamental differences between Sunni and Shiite Muslims is over which of Mohammed's relatives was his true successor as leader of the faith.

In neither of these cases did the claims originally result from anything beyond the immediate extended family.

Yes, if the fort falls, the dwarves inside will most likely fail to reproduce, I considered that all right. My point was that dwarves will still reliably fail to reproduce even if the fort DOESN'T fall. Out of 200 dwarves, maybe 10 of them will get married, and maybe 1 of those marriages is homosexual, so you're probably looking at 4 productive couples--who together churn out enough babies to keep the population right at the cap. No migrants can enter, so each child's dating pool is restricted to kids from the other 3 families (on top of all the rules they faced already). With such a reduced pool, the odds of finding a mate becomes very small--there will still be a few successful pairings, but as each family of children drops out of the running, the number of eligible mates becomes lower and lower--because a larger and larger percentage of the fort's inhabitants are one's own siblings. The number of reproducing families will inevitably shrink to 2, then 1, and finally none. Ironically, the only way a dwarf fort can survive in the long term is by failing, reducing the population below the cap so migrants can bring in new blood.

Most people don't have families at all since they are not historical characters.  The situation you are describing will however not happen, because the population cap will eventually drop as the older generation die off, which causes existing marriages of the middle-aged dwarves to become fertile again.  That is because the child cap constrains the reproduction of the married couples once there are a certain number of children.  But once the older dwarves die of old age and none of the younger generation can marry since there is only one family, the middle-aged dwarf marriages simply churn out more children. 

In effect, in DF as in nature reproductive potential is seldom actual reproductive outcome.  If all the younger generation are one family, then the older generations families will simply make new offspring for them to breed with. 

I think I should clarify--there's a reason I said replace. Consider the dwarf Momuz Thunderbreeches, granddaughter of Stakuz Thunderbreeches. Due to her dual professions of Butcher and Axedwarf, Momuz earns the combat title "Slayer of Geese". She then bears a son, whom she names Datan Gooseslayer, because she figures any name related to a combat title is better than being associated with someone famous only for his flatulence. So yes, your response is largely correct--Datan Gooseslayer might very well have elder siblings already named Thunderbreeches, creating a small but noticeable dysfunction. As long as Momuz herself remains alive, with her name & title acting as a reminder that Thunderbreeches and Gooseslayer are the same family, that's largely okay, but after her death it's purely up to the player to remember . . . unless other individuals decide to keep the clan intact under the same name, unilaterally changing their own surnames to Gooseslayer (or back to Thunderbreeches).
I wasn't implying that combat titles themselves be inherited--for instance, I never meant that the son should be named "Datan Thunderbreeches, Slayer of Geese".

A large number of people end up being called Gooseslayer.  Perpetually inherited surnames are no less dysfunctional if they are originally assigned to heroes, since most people are not heroes.

At the moment the game has the following name structure:

{a} {b}{c} the {e} of {f}

All variables are proper nouns choosen randomly to create a name. I suggest we leave it that way, but insist that all creatures, items, minerals and other things DF features, should be included in every language of every race.

I have actually spent the last year doing just that.  It is ton of work to add in all the things in the game as words.

[voliol]

Considering this is Dwarf Fortress we're talking about, I'm surprised this discussion has barely mentioned the question of how the values and personalities of individual Dwarves and entities would affect these naming conventions.
Say, the existance of lineage-based names should probably depend on how much the entities value family to begin with (e.g. no goblin surnames), and individuals with high regard for traditions should be less likely to change their names than those with disdain for them.
Also, the personality of an individual should affect what kind of event they eventually take their name after. That is, dwarves that don't like craftmansship likely wouldn't name themselves after it, and dwarves that like working would more often name themselves "Smith", "Miller" etc.

On the other hand, it should be possible for others to name an indivual, so that new knights can get names from whoever knights them, feared demons getting names from their oppressees, or monarchs taking the names their advisors advises them to instead of their own preferences.

All this does kinda merge in a wierd way with nicknames, so I might be getting a bit off topic.

By the way, I assume we all agree that the naming systems should be decided on the entity-level and not the species level?

[dragdeler]

I have actually spent the last year doing just that.  It is ton of work to add in all the things in the game as words.

You freak  ... you didn't also happen to figure out some procedural phonetic generator, that gives a list after you define it's length, of the shortest and most pronouncable syllabes, in order to shorten those dang names?

it should be possible for others to name an indivual

From what I've heard, when Napoleon conquered germany, he forced some sort of census where the german civil servants had to assign familynames to everybody. And the jews refused so the officers just invented those typical weird jewish names such as "Goldberg" or "Rosenstein" (and that's how I know about this). This is a very nice example of a trigger for titles (such as described in my previous posts).

By the way, I assume we all agree that the naming systems should be decided on the entity-level and not the species level?

I'm a bit fuzzy on the raws. By that you mean the different civs (nations?)? If so yes.

this discussion has barely mentioned the question of how the values and personalities of individual Dwarves and entities would affect these naming conventions

I think the character trait "ambitious" for example

Spoiler (click to show/hide)

    91-100    has a relentless drive, completely consumed by ambition
76-90    is very ambitious, always looking for a way to better [his/her] situation
61-75    is quite ambitious
25-39    isn't particularly ambitious
10-24    is not driven and rarely feels the need to pursue even a modest success
0-9    has no ambition whatsoever

would be a very good indicator how fast an individual settles with a title, in how I imagine my suggestion.

too much edit

[Shazbot]

For me, species-level naming conventions give an extra degree of granularity. I'm using castes to create a caste of goblin "untouchables" who wouldn't merit a last name. But I will take what I can get, so long as some degree of lineage and "clannishness" is possible.

[SixOfSpades]

The family name system in real-life works extremely poorly, especially under certain conditions, yes it does work better than having no family name system at all, but that may only be if you consider a false conclusion better than no conclusion.

Wow. I'm amazed you feel so strongly about this, and in such direct opposition to pretty much the entire world--but hey, you do you.

I am not proposing that we do not have family names, I am just proposing that we do not simply copy the systems that are used in real-life but come up with a better system. . . . A lot of things real-life cultures do is quite imbecilic, in fact I would say the majority of things that real-life cultures do is such.  So if you want things to work properly, blindly copying real-life cultures is definitely not the way to go.

True, there are certain cultures that do things that I personally find rather silly--like some Polynesian peoples refusing to ever speak the name of a person who has died, or English speakers making non sequitur nicknames, like shortening "Charles" to somehow get "Chuck". But these are quite definitely the exception, the vast majority of naming protocols seem quite sensible to me--and obviously, to those who use them. Even the real-world chauvinist standard of "the wife takes the husband's surname" would cease to be anywhere near as sexist in DF, if we simply introduce an equally-weighted feminist counterpart. (It makes more sense to claim parentage from the mother, anyway.)
When you say we should come up with "a better system", I hope you don't mean just one system: I've always championed the idea that different races, and different civilizations within those races, should have different cultural behaviors, and naming conventions should definitely be part of that. Besides, having a set of random options present in the raws would likely allow players who dislike certain possible behaviors (like yourself) to go into the raws and edit out the options they'd rather not see.

It is not a problem at the moment because there are no family surnames.  It becomes a problem if we replace the present surnames with family ones, since we will end will a large number of people with the same name.

I think you're making a mountain out of a molehill with this. Sure, let's say the Bunnyhammer family consists of a married couple who produce 15 kids, one of whom manages to marry and have another 9 kids, and outside the fort there's an extended Bunnyhammer clan with 127 living members (I'm not sure how many of those would likely be considered historical figures). So that's about 150 Bunnyhammers . . . but since there are hundreds of viable first names, the game can still name them all with no repeats. (Getting close to an "upper limit" on clan size could be another prompt for members to break off into a new clan.)

By culture I mean culture,

That wasn't helpful, but I'll try to play along. Every creature has a race (species), a birth civilization (set of social customs used by their parents [both parents are likely to share the same one]), a formative civilization (where they grew up), and a current civilization (where they live now). A person's culture is going to depend on ALL of these, and in my opinion their name should as well. A dwarven child snatched off to live as a goblin would probably be given a goblin name (and, as I said, might not ever remember their dwarven one).

This raises the question of what the point of first names is at all and the answer is that it allows us to easily identify the culture of our individual. . . . a goblin born in a dwarven civilization, or a non-historical goblin promoted into a historical characters gets a dwarven name.

But that's hardly limited to just the first name. Just because we also see the English translations of their names, doesn't mean we can't tell the difference between "Stinthad" and "Ngustpuz". Besides, if it's in a fort, the goblin's going to be shown as a 'g' anyway, not a '☺'. And it's not like we can't just overwrite the special cases with a nickname, so it's a moot point.

We should do the opposite of making first names unique.  Having identified that the only function of first names is to identify culture, we can make everyone in the culture have the same 'first name'.  That makes things work a lot better, since we have a shared name that tells us what culture the individual belongs to, one shared name telling us what family household they belong to and a third unique name identifying the specific individual.

Oh HELL no. You criticize the majority of human cultures' naming conventions as "imbecilic", and then suggest THIS as an improvement!? You never cease to amaze me with your ability to make wild, unfounded suppositions and then immediately treat them as established objective facts. But let's mentally put this plan of yours into practice anyway: Since your home civilization is called "The High Candles", every native dwarf in your fort has "Highcandle" as name elements 1 and 2 . . . which serves no purpose whatsoever except to a) take up valuable space, and b) distinguish them from the fort's various merchants, guests, and possible invaders--who of course are already flagged as Merchant, Guest, or Invader in the Units list. I just checked, you can't use nicknames to completely remove a dwarf's first name--and even if you could, to have them all called "Highcandle" by default is what truly deserves to be called imbecilic.
The point of a name is to distinguish. A name, or a name element, cannot distinguish if there is nothing meaningful to distinguish it FROM. Bob Higgins doesn't go around calling himself "Human Bob Higgins" all the time, and even the far more specific "Philadelphian Bob Higgins" still doesn't mean a damn thing if he's in Philadelphia. The sort of race-specific or civ-specific names that you're suggesting could only make sense if they were applied to just the fringe elements in a given society.

The Hundred Years' War was fought over the King of England's birthright claim to the throne of France. By its conclusion, royals were still debating the succession rights from seven generations prior. And even today, one of the fundamental differences between Sunni and Shiite Muslims is over which of Mohammed's relatives was his true successor as leader of the faith.

In neither of these cases did the claims originally result from anything beyond the immediate extended family.

Hmm, yes and no. For instance, because the Hundred Years' War had interludes of peace, those monarchs who came later had to look further back to find "just cause" to resume hostilities. In Shakespeare's Henry V, a big chunk of Act 1 Scene 2 is literally devoted to a tedious and rather arcane history lesson: https://www.youtube.com/watch?v=Pnk1fbGNWrM (and I'm pretty sure this part of the script is actually edited down from the original version).
This tangent has diverted more than far enough from the topic of the thread, and I'd prefer not to continue it. I primarily added this reply just for the YouTube link, because Branagh made a bitchin' Henry V.

Most people don't have families at all since they are not historical characters.  The situation you are describing will however not happen, because the population cap will eventually drop as the older generation die off, which causes existing marriages of the middle-aged dwarves to become fertile again.  That is because the child cap constrains the reproduction of the married couples once there are a certain number of children.  But once the older dwarves die of old age and none of the younger generation can marry since there is only one family, the middle-aged dwarf marriages simply churn out more children. 
In effect, in DF as in nature reproductive potential is seldom actual reproductive outcome.  If all the younger generation are one family, then the older generations families will simply make new offspring for them to breed with.

Yes, but the problem is not "babies don't get born", the problem is "couples don't get married". Toady largely "fixed" the issue by making (apparently) all migrants already heterosexually married upon arrival, so the initial number of productive couples in the fort is VERY high at first. Since most forts don't last more than a generation or two, this band-aid solution is nearly perfect in practice (although the resulting babysplosion does make the first few years even more difficult).
I too must admit that certain of my conclusions are not empirical--I have not personally run even a single fort long enough to see all the original migrants die off, let alone for the clear majority of their descendants to refuse to marry and thus start the inbreeding train. But I know it is going to work poorly because I, too, can think critically, and I know that when the average dwarf considers perpetuating their species to be LESS important than a shared fondness for a particular metal, that's definitely a bad sign. Yes, when a significant number of old dwarves die, more babies will be born to replace them, and the babies will definitely meet the age requirement with each other. But just because they can marry & reproduce doesn't mean a realistic number of them will.

A large number of people end up being called Gooseslayer.  Perpetually inherited surnames are no less dysfunctional if they are originally assigned to heroes, since most people are not heroes.

A large number of people are named Smith, too, even though most of them are not smiths. Everyone knows this. They also know that the name Smith is so common, there is no reason to assume that any two random people named Smith are related--it's more likely that they are not. Yet this is not a dysfunction, because a) it's so well-known, and b) even all the Smiths together would still constitute only a sliver of a minority, against all the other thousands of names out there. It can be the most popular last name, but as long as it doesn't dominate (the way Highcandle would), it still serves its purpose of distinguishing each Smith family from the rest of society as a whole.

All variables are proper nouns choosen randomly to create a name. I suggest we leave it that way, but insist that all creatures, items, minerals and other things DF features, should be included in every language of every race.

I have actually spent the last year doing just that.  It is ton of work to add in all the things in the game as words.

My reply is unrelated to the thread topic, so I sent you a PM.

Considering this is Dwarf Fortress we're talking about, I'm surprised this discussion has barely mentioned the question of how the values and personalities of individual Dwarves and entities would affect these naming conventions. . . .
On the other hand, it should be possible for others to name an indivual, so that new knights can get names from whoever knights them, feared demons getting names from their oppressees, or monarchs taking the names their advisors advises them to instead of their own preferences.
By the way, I assume we all agree that the naming systems should be decided on the entity-level and not the species level?

Agreed, decisions like this should largely be up to the individual dwarf, although certain others should also have enough influence on the dwarf to have some control as well. As previously mentioned, a newly-arrived migrant (or newly-named infant) might be renamed by local authorities if a name is deemed improper or likely to cause confusion. Depending on the social customs of the civ, and a dwarf's regard for tradition & authority, a family elder might be able to control things like baby names or breaking off a new clan. And combat titles should arise from witnesses to the deed(s) in question, so that most likely means the other members of the dwarf's squad . . . or at least those who saw the dwarf returning from the kill.

Also, achieving personal dreams like "mastering a skill" and "creating an artifact" should receive an additional bonus if/when it comes to deciding a dwarf's new name.

[GoblinCookie]

True, there are certain cultures that do things that I personally find rather silly--like some Polynesian peoples refusing to ever speak the name of a person who has died, or English speakers making non sequitur nicknames, like shortening "Charles" to somehow get "Chuck". But these are quite definitely the exception, the vast majority of naming protocols seem quite sensible to me--and obviously, to those who use them. Even the real-world chauvinist standard of "the wife takes the husband's surname" would cease to be anywhere near as sexist in DF, if we simply introduce an equally-weighted feminist counterpart. (It makes more sense to claim parentage from the mother, anyway.)
When you say we should come up with "a better system", I hope you don't mean just one system: I've always championed the idea that different races, and different civilizations within those races, should have different cultural behaviors, and naming conventions should definitely be part of that. Besides, having a set of random options present in the raws would likely allow players who dislike certain possible behaviors (like yourself) to go into the raws and edit out the options they'd rather not see.

It's not one system, well in a way it is  . We have a large number of names, reflecting all different types of important information and we eliminate those that are not culturally relevant.  In some cases, as with family the length of names is controlled by a number of indefinite length.  So if we don't think a civilization would care about a particular piece of information, we leave it out even though the player might want to know it.  We can do this automatically on the basis of values, so a civilization that does not care about [FAMILY] does not bother with surnames at all and the more they care the more generations of family they record. 

I think you're making a mountain out of a molehill with this. Sure, let's say the Bunnyhammer family consists of a married couple who produce 15 kids, one of whom manages to marry and have another 9 kids, and outside the fort there's an extended Bunnyhammer clan with 127 living members (I'm not sure how many of those would likely be considered historical figures). So that's about 150 Bunnyhammers . . . but since there are hundreds of viable first names, the game can still name them all with no repeats. (Getting close to an "upper limit" on clan size could be another prompt for members to break off into a new clan.)

That is effect an alternative means at carrying out the basic function of my naming, getting rid of old surnames so that they do not pass out context and lose accuracy as regards to family identification for the player.  However your idea depends too much on a situation working out in a particular fashion, for instance it breaks down if a forgotten beast scatters the Bunnyhammers about the place and then a whole series of catastrophes ensures that the population never reaches the magic splitting-point which I presume retires that name and creates two new names. 

In any case, where did the clans come from?  We were never talking about clans, which are more family-site government hybrids to put it in game terms. 

By culture I mean culture,

That wasn't helpful, but I'll try to play along. Every creature has a race (species), a birth civilization (set of social customs used by their parents [both parents are likely to share the same one]), a formative civilization (where they grew up), and a current civilization (where they live now). A person's culture is going to depend on ALL of these, and in my opinion their name should as well. A dwarven child snatched off to live as a goblin would probably be given a goblin name (and, as I said, might not ever remember their dwarven one).[/quote]

That would result in us having two names for the same character, one according to the culture of the first civilization from which the child was stolen and the second according to the second civilization.  No, the child keeps the original name as at present, but when they grow up and become formerly part of goblin they adopt the goblin civilizations culture-name, as do all other full immigrants, in addition to having the rest of their name translated into the goblin language.  As a special case we should probably hide their family names (not actually delete) as long as they remain part of a [BABYSNATCHER] civilization, that way if they migrate back into their own civilization (or into any non-babysnatcher one incidentally) their family names will restore, unless they would have been overridden for another reason, for instance they got married to someone in the goblin civilization. 

But that's hardly limited to just the first name. Just because we also see the English translations of their names, doesn't mean we can't tell the difference between "Stinthad" and "Ngustpuz". Besides, if it's in a fort, the goblin's going to be shown as a 'g' anyway, not a '☺'. And it's not like we can't just overwrite the special cases with a nickname, so it's a moot point.

It is a lot easier to do that reliably with first names because you have 'heard that name before' as it were, the uniqueness of two-string names makes it harder, especially if the languages are similarly sounding (modded languages may well be).  In any case, first names are an imperfect system that I am proposing we replace effectively with the two-string surnames and replace their function with a seperate cultural name.  In effect, what I am doing is pointing out the cultural-identification function of non-unique first names, a function lost by making them unique and proposing we simply replace this with a reliably uniform name. 

Oh HELL no. You criticize the majority of human cultures' naming conventions as "imbecilic", and then suggest THIS as an improvement!? You never cease to amaze me with your ability to make wild, unfounded suppositions and then immediately treat them as established objective facts. But let's mentally put this plan of yours into practice anyway: Since your home civilization is called "The High Candles", every native dwarf in your fort has "Highcandle" as name elements 1 and 2 . . . which serves no purpose whatsoever except to a) take up valuable space, and b) distinguish them from the fort's various merchants, guests, and possible invaders--who of course are already flagged as Merchant, Guest, or Invader in the Units list. I just checked, you can't use nicknames to completely remove a dwarf's first name--and even if you could, to have them all called "Highcandle" by default is what truly deserves to be called imbecilic.
The point of a name is to distinguish. A name, or a name element, cannot distinguish if there is nothing meaningful to distinguish it FROM. Bob Higgins doesn't go around calling himself "Human Bob Higgins" all the time, and even the far more specific "Philadelphian Bob Higgins" still doesn't mean a damn thing if he's in Philadelphia. The sort of race-specific or civ-specific names that you're suggesting could only make sense if they were applied to just the fringe elements in a given society.

The game does not entirely consist of Fortress mode.  In adventure mode and legends mode we often get big lists of people with diverse backgrounds to sift through, this idea was directed towards those modes not Fortress mode.  In fortress mode, the idea does not do much I agree, but it was not for the benefit of fortress mode that I proposed it.

My idea was to string together the first string of the civilization's with the last string of the site government.  So if my civilization was called the Guilds of Steel and my site government is called the Society of Rabbits, then all it's members would be called the Guildrabbits.  That means without having to check, whenever I see someone called Guildrabbit, I can immediately tell that they are part of the Society of Rabbits.  I can run a search and I can immediately come up with a list of historical characters that lived in their territory, without having to trawl through lists of "so and so stole a mug in Yr X" with a notepad. 

Hmm, yes and no. For instance, because the Hundred Years' War had interludes of peace, those monarchs who came later had to look further back to find "just cause" to resume hostilities. In Shakespeare's Henry V, a big chunk of Act 1 Scene 2 is literally devoted to a tedious and rather arcane history lesson: https://www.youtube.com/watch?v=Pnk1fbGNWrM (and I'm pretty sure this part of the script is actually edited down from the original version).
This tangent has diverted more than far enough from the topic of the thread, and I'd prefer not to continue it. I primarily added this reply just for the YouTube link, because Branagh made a bitchin' Henry V.

None of those periods of peace lasted very long, certainly not enough for us to be dragging up lineages going back millenia.  But it mostly about France's sexist inheritance laws and England's relatively less sexist laws.  Interestingly we had a similar situation in England with Stephen and Matilda, in which the former ultimately accepted the latter's claim, disinheriting his own children and establishing the opposing principle in France to that of England. 

Yes, but the problem is not "babies don't get born", the problem is "couples don't get married". Toady largely "fixed" the issue by making (apparently) all migrants already heterosexually married upon arrival, so the initial number of productive couples in the fort is VERY high at first. Since most forts don't last more than a generation or two, this band-aid solution is nearly perfect in practice (although the resulting babysplosion does make the first few years even more difficult).
I too must admit that certain of my conclusions are not empirical--I have not personally run even a single fort long enough to see all the original migrants die off, let alone for the clear majority of their descendants to refuse to marry and thus start the inbreeding train. But I know it is going to work poorly because I, too, can think critically, and I know that when the average dwarf considers perpetuating their species to be LESS important than a shared fondness for a particular metal, that's definitely a bad sign. Yes, when a significant number of old dwarves die, more babies will be born to replace them, and the babies will definitely meet the age requirement with each other. But just because they can marry & reproduce doesn't mean a realistic number of them will.

That really does not matter much.  As long as it is not only one family reproducing, which won't happen for the reasons I mentioned, as there own inability to reproduce simply encourages whatever non-related married couples there are to reproduce, basically similar to Fisher's Principle, which explains why we always have roughly equal numbers of both gender in most species.  So in effect it does not matter much, except for 'realism' if we have only a small number of people reproducing but having a lot of babies or whether we have a large number of people having only a handful of babies; those are the only two options really. 

In any case, the real thing that will cause mischief is the inability to remarry and the inability to marry across age-ranges.  It is not how many of them marry, it is restrictions on marriage in general which potentially could cause problems. 

A large number of people are named Smith, too, even though most of them are not smiths. Everyone knows this. They also know that the name Smith is so common, there is no reason to assume that any two random people named Smith are related--it's more likely that they are not. Yet this is not a dysfunction, because a) it's so well-known, and b) even all the Smiths together would still constitute only a sliver of a minority, against all the other thousands of names out there. It can be the most popular last name, but as long as it doesn't dominate (the way Highcandle would), it still serves its purpose of distinguishing each Smith family from the rest of society as a whole.

That is exactly why surnames don't work.  If I meet two Smiths, then I cannot in any way determine them to be related to each-other, so why do surnames even exist?  That there are loads of other surnames does not change that, the situation still applies since if I meet two people who have the same surname I cannot by this fact alone determine them to be related; this problem is inherent to the surname system when played out of a long time period.  That is why current surnames are dumb, the people who invented the system did not think about what would happen in the long-run and come up with any mechanism to ensure that essentially unrelated people don't end up with the same surname. 

The solution is to create a new surname every time two people get married and have this inherited by their children, but when their children get married they replace one of their ancestral surnames with a new surname.  That way whenever I run across a person with the same surname I know they are of the same family group, rather than being somebody who shares a common ancestor a thousand years ago.