Print

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

[GoblinCookie]

The semantics of what defines a clan, state, or society has nothing to do with the inheritance of names. It's a tangent. The parts actually relating to the handling of names and titles are fine.

My idea was to figure out how to implement clans in the game, but Six of Spades did not really cooperate.  Clans are pretty much a standard dwarfy thing for most fantasy settings afterall.   

If we're bothering to simulate a change in naming systems, there's no reason to have that change propagate faster than light. It can spread using the rumors system.

You at least have to be reasonable about your worst case scenarios. If the player does something extreme, like a 16x16 embark, then they must bear the performance issues.

The player is not intentionally doing anything, he just chose to have few civilizations, a lot of sites and then played adventure mode long enough to have spoken to most everyone before setting up a fortress and getting maximum population.  I do not claim to know the inner working of Dwarf Fortress, none but Toady One does, however the rumours system is not likely to make things worse, actually it is possible that it would be harder for it to do things than to just calculate everything in an instant. 

The reason is that rumours are exponential.  One person tells ten people, who tell 100 people, who tell 1000 people.  At the end of the chain we end up with a lot of calculations for rumours, on top of the number of calcuations themselves.  In any case, it is not the case that we *need* to create the problem in the first place, simply because we can make it less bad. 

As long as you grudgingly admitted (again) that clans are not governments, I'm good with that. Clans are, first and foremost, large families that share a name, and I shall continue to refer to them as such. And that's all I'll say about the issue in this thread.

It is fine to think that, as long as you don't go around saying clan when you just mean a large family.  Saying clan implies other things are the case, so when you mean family just say family.

I think it's an interesting way of adding flavor, but I definitely think it shouldn't be common, and when it does occur it should show equal-opportunity bias.

An equal-opportunity bias implies that it is all somehow random.  Prejudices come about for a reason, they are not just created out of nothing by the RNG to fill some prejudice quota.  Prejudices if they exist should be based upon the ultimately semi-random events in the game and in mythology, rather than simply arbitrarily conjured up. 

The shorthand isn't supposed to be reader-friendly, it's for computers. You get the text description. I've updated my post to match your specs (I think), and included a preview of what happens when you adjust "the maximum allowable number in the entity file".

You are writing your post to be read by computers! 

You do end up with enormously long 'true' names, which is fine because the majority of should not be seen in most contexts and it's really just raw numbers which modern computers can store a staggering number of.  It does seem I do need a hard limit on the number of supportable generations, 5 is a fine number since it is too ridiculously large a number of names that nobody would ever want to mod in more. 

Expect the name algorithm to come back with a lot of "null"s where there should be males, or along both lines if the kid is a foundling or whatever. The longer the names, the longer the child and its descendants are going to be bearing the mark of bastardy.

Nobody can tell at a glance using my system that anyone is illegitimate, they would have to know the relevant mother or grandmother's personal name first and then they can see if it is the same as that of their child; you could say this system both hides and acknowledges illegitimacy.  I think there are no nulls involved, since the personal name of the mother is taking up the whole slot that would normally be taken up by the marriage-name of the mother AND father. 

Foundlings however are not a problem since they can just be given the full marriage name of their adopted family, or the personal name of their adopted parent, which could be a male I supposed.  The interesting question here is whether illegitimate children should adopt an ordinary marriage name if their parents THEN get married?

Wait--do you mean parthenogenesis? As in, dwarf females spontaneously impregnating themselves with their own clones? This is certainly the first that I've heard of such a suggestion, especially considering that it doesn't happen among mammals. Although it would solve the "dwarves die out because they don't get married" problem . . .

I was obviously not talking about dwarves.  There are afterall other creatures in the world with names and families. 

[Bumber]

The player is not intentionally doing anything, he just chose to have few civilizations, a lot of sites and then played adventure mode long enough to have spoken to most everyone before setting up a fortress and getting maximum population.  I do not claim to know the inner working of Dwarf Fortress, none but Toady One does, however the rumours system is not likely to make things worse, actually it is possible that it would be harder for it to do things than to just calculate everything in an instant. 

The reason is that rumours are exponential.  One person tells ten people, who tell 100 people, who tell 1000 people.  At the end of the chain we end up with a lot of calculations for rumours, on top of the number of calcuations themselves.  In any case, it is not the case that we *need* to create the problem in the first place, simply because we can make it less bad.

Speaking to everyone in a civ is an undertaking (pun not intended) more difficult than killing everyone in a civ. (The primary roadblock is that not everyone can be loaded simultaneously, and hist-figs take priority.) The issue correlates to an absurd time investment, and exists inevitably by the creation of too many historical figures regardless.

What I meant about the rumors system is that it can be used as a trigger to divide the workload to a handful of sites at a time, as they learn about the new policy. We know the system performs adequately right now. However, you may be correct that this won't help much if it is the case that rumors simply enter a shared global pool once they leave the site of origin.

There doesn't exist a flawless solution. The only way to avoid creating problems is to avoid adding features. Not that I particularly care if name systems can change, but simulating such things is well within DF's scope.

Granted, I'm not entirely sure I understand what you mean by needing new names. Am I correct in assuming you're talking about adding an additional segment onto each person's name? If you want them to be unique, you merely have to hand them out in order. If you want them to have meaning, you don't expect everyone to figure it out simultaneously.

[GoblinCookie]

Speaking to everyone in a civ is an undertaking (pun not intended) more difficult than killing everyone in a civ. (The primary roadblock is that not everyone can be loaded simultaneously, and hist-figs take priority.) The issue correlates to an absurd time investment, and exists inevitably by the creation of too many historical figures regardless.

I regularly go around the place talking to various random folks and asking them questions about the surrounding area.  It is really a matter of time before I would end up talking to everyone, point in any case is hyperbole; I was really talking about how the total number of historical characters tends to go up over time.   You cannot take for granted that there are only a small number of historical characters since the ultimate state of the game is for everyone to be historical if played long enough in both modes.

What I meant about the rumors system is that it can be used as a trigger to divide the workload to a handful of sites at a time, as they learn about the new policy. We know the system performs adequately right now. However, you may be correct that this won't help much if it is the case that rumors simply enter a shared global pool once they leave the site of origin.

Doing this by rumors system is quite possibly more difficult than simply having them instantaneously recalculate thousands of names.

There doesn't exist a flawless solution. The only way to avoid creating problems is to avoid adding features. Not that I particularly care if name systems can change, but simulating such things is well within DF's scope.

Granted, I'm not entirely sure I understand what you mean by needing new names. Am I correct in assuming you're talking about adding an additional segment onto each person's name? If you want them to be unique, you merely have to hand them out in order. If you want them to have meaning, you don't expect everyone to figure it out simultaneously.

No Bumber, a lot of times the problem does not have to exist in the first place, because there is a better way of doing it than the way that you are thinking of.  This happens a lot, people come up with a flawed idea, there are better ideas about and the person comes up with elaborate solutions that rectify the problems with their original idea, rather than simply coming up with a better idea that does not cause the problem to begin with. 

The better idea is to have all the different systems of naming be reducible to a single naming system which the player does not see, the differences between civilizations are simply a list of instructions to convert that universal single naming system into what people actually see.  That way if the system suddenly changes, there is no load on the system since because nobody is seeing the *actual* names at all, there is no need to actually change them. 

[SixOfSpades]

-Killed "creature".
-Priest of "deity".
-Winner of "world gen event".
-First to climb "mountain peak"
-Created "artifact".

If we're going to be adding specific names, especially creature names, onto other names, then the added names must be designated as titles, not actual names. Otherwise, you could get a near-infinite chain of "Bob who killed Tom who killed Sally who killed Albert who killed . . . " etc. Also, each of these achievements are going to need different "prestige" ratings associated with them, so a dwarf who has accomplished 2 or more noteworthy things can judge which one to add to their name.
Honestly, I'd rather be happier if the combat titles were tweaked so they were actually relevant to the manner in which they were earned. Something about who was killed (goblins, bandits, etc.), what kind of weapon their slayer used, the big-picture result of the victory, etc. Specific names seem less important, unless it was a truly legendary kill . . . which, again, involves prestige.

I think we will need to link at least one title to the life goal,

Eh, depends on what the goal is. I don't think a dwarf who "dreams of raising a family" should be honored for accomplishing that feat with an additional name; to me, that sounds more like a cultural distinction--in a given civilization, everyone who has children has a 'parent' string added to their name, whether that was their dream or not. But that may just be my personal preference: I believe that dwarves having their own kids should be much more common, so honoring them for this "distinction" seems a bit silly to me.

I personally think that if you detach the process of gaining a title from the actual event it is related to, . . . .  imagine Urist being known as dragonslayer since the day Kulet made that awesome poem about it, that got really famous.

I agree, that's a nice touch, but adding that extra "publicity" layer of realism is non-essential. Sure, it can be done: 'Dragonslayer' gets added to Urist's true name, and the rumors system propagates the news of his deed and Kulet's poem, and Urist's Self-Importance and Modesty traits influence how aggressively he self-promotes his title. The game can then use all of that to calculate whether 'dragonslayer' catches on a part of Urist's external name. But all told, I think the CPU cycles would be better spent elsewhere.

. . . we should be able to bypass the title changing problem by setting thresholds prohibitively high, so that titles are kind of rare in order to save precious computation power and grant them more importance and meaningfulness.

Agreed. If I see a dwarf with a combat title, I don't know if that means "I've tanked dragons, hydras and bronze colossi by myself", or "Nine of my heavily armed friends and I beat the shit out of a bunch of naked, starving goblins, one by one".

There is some last thing which I would like to mention, which is the english translations of that gibberish and how they're used in the UI. See I allways found it very confusing when and how the game applies which language. Building in translations so that we don't need to know all the gibberish grammar when following conversations should be fairly easy. But maybe it's time to reconsider the "what language where" from a comprehensive design perspective, or just make it completly customizable?

I don't play Adventurer mode, so I'm not sure if you're referencing that. Is there part of the game that actually tries to translate dwarven (or any other of the DF languages) into English? I doubt that's possible, as the in-game vocabularies are very incomplete, missing tons of words necessary to create a complete sentence. They're fine for their current purposes (just the names of people, cities, landforms, etc), but you can't actually say anything in them.

Information pertaining to the current identity needs to remain distinct from any previous ones until cover is blown, however. Not sure how the new secret identities system maintains this. Does legends mode spoil things, or does info get filtered out by name?

I believe vampires don't simply make up a fake name, they actually assume the identity of one of their victims. So their fake name should also include a pointer to the real (dead) dwarf's background, so they can lie convincingly without requiring any additional memory space. (Of course, how well the vampire succeeds at this is for another thread.)
As for aliases other than vampires, I lack sufficient knowledge of examples to say how they might be handled.

My idea was to figure out how to implement clans in the game, but Six of Spades did not really cooperate.  Clans are pretty much a standard dwarfy thing for most fantasy settings afterall.

A clan is a family of related people who value the surname that unites them. This entire thread is about how best to give related dwarves family-based surnames. And literally the very first reply on this thread is you stating that family names are a waste, and random names are the way to go. As for my 'lack' of cooperation with implementing clans, who's the one compiling example name structures?

I think [discrimination] is an interesting way of adding flavor, but I definitely think it shouldn't be common, and when it does occur it should show equal-opportunity bias.

An equal-opportunity bias implies that it is all somehow random.  Prejudices come about for a reason, they are not just created out of nothing by the RNG to fill some prejudice quota.  Prejudices if they exist should be based upon the ultimately semi-random events in the game and in mythology, rather than simply arbitrarily conjured up.

Whether there's an actual worldgen event where the founders of a culture have reasons to sit down & literally decide "Okay, we're going to oppress X caste of our citizens", or whether Toady just simulates that with a much simpler dice roll, is largely invisible (and likely quite insignificant) to the player.

The shorthand isn't supposed to be reader-friendly, it's for computers. You get the text description. I've updated my post to match your specs (I think), and included a preview of what happens when you adjust "the maximum allowable number in the entity file".

You are writing your post to be read by computers! 

Is that . . . scorn? You're aware that we're discussing improvements to a computer game, yet when I reduce descriptions of text strings to a representation of abstract data that more closely approximates code, you express . . . surprise? that I should do such a thing? 

Nobody can tell at a glance using my system that anyone is illegitimate, they would have to know the relevant mother or grandmother's personal name first and then they can see if it is the same as that of their child; you could say this system both hides and acknowledges illegitimacy.  I think there are no nulls involved, since the personal name of the mother is taking up the whole slot that would normally be taken up by the marriage-name of the mother AND father.

Ehh, maybe. The way I've got your system described right now, dwarves have three names all their own: 1[] and 2[] are their given names (random, the way you like it), and 5[] is their half of the "marriage name", also random. Which of those, precisely, are you calling your dwarf's "personal name"? Regardless, in the current description, 5[] isn't (necessarily) used until the dwarf marries--each one then takes their spouse's 5[] name as their 6[]. This means that, if there's no dad, then mom isn't going to have a 6[] to append to her 5[], and the bastard child is going to be missing the names of ALL of his paternal ancestors (unless the mother just writes her own name in there, over & over).

Foundlings however are not a problem since they can just be given the full marriage name of their adopted family, or the personal name of their adopted parent, which could be a male I supposed.  The interesting question here is whether illegitimate children should adopt an ordinary marriage name if their parents THEN get married?

All answers are possible, and indeed desirable, for increased cultural variation.

Wait--do you mean parthenogenesis? As in, dwarf females spontaneously impregnating themselves with their own clones? This is certainly the first that I've heard of such a suggestion, especially considering that it doesn't happen among mammals. Although it would solve the "dwarves die out because they don't get married" problem . . .

I was obviously not talking about dwarves.  There are afterall other creatures in the world with names and families.

If some user comes on the forums and starts talking about bizarre stuff like "I can't wait until Toady finally implements flight, and shapeshifting," then yes, I would automatically give that user the benefit of the doubt and assume that they must be talking about some creatures other than dwarves, because otherwise the suggestion would be absurd.
But when that user is you, GoblinCookie? I'm sorry, but at this point I just can't take anything for granted.

[dragdeler]

If we're going to be adding specific names, especially creature names, onto other names, then the added names must be designated as titles, not actual names.

I'm not suggesting that their string is the same than the event it relates to. For example Urist Familyname has gained the title "Nokzambibam-" (battleball) "-zora" (killer of) for defeating the megabeast Ithi Questbreaches. It's just an arbitrary random string. As I said earlier, Imagine them obscure references that don't translate well into english. Whether you have one or two strings, constituting one or two titles, could be balanced by civ preferences, or by the importance of a title. I'm just using two in the example above for the ring of it.

Dragdeler: I think we will need to link at least one title to the life goal,

Six Of Spades: Eh, depends on what the goal is. I don't think a dwarf who "dreams of raising a family" should be honored for accomplishing that feat with an additional name; to me, that sounds more like a cultural distinction--in a given civilization, everyone who has children has a 'parent' string added to their name, whether that was their dream or not. But that may just be my personal preference: I believe that dwarves having their own kids should be much more common, so honoring them for this "distinction" seems a bit silly to me.

Everybody having historical figures as parents would carry a familyname, but creatures would reference to eachother according to civilisation preferences. The player could choose to show or cull any part of a name as he pleases. Those who dream of founding a family are kind of the lame duck cases to me because of precisely that.

The solution that pops up to my mind: a creature dreaming of founding a family, will only gain a title for something else, under the condition that he has enough children, of which a sufficient amount have a grown old enough.

Dragdeler: I personally think that if you detach the process of gaining a title from the actual event it is related to, . . . .  imagine Urist being known as dragonslayer since the day Kulet made that awesome poem about it, that got really famous.


Six Of Spades: I agree, that's a nice touch, but adding that extra "publicity" layer of realism is non-essential. Sure, it can be done: 'Dragonslayer' gets added to Urist's true name, and the rumors system propagates the news of his deed and Kulet's poem, and Urist's Self-Importance and Modesty traits influence how aggressively he self-promotes his title. The game can then use all of that to calculate whether 'dragonslayer' catches on a part of Urist's external name. But all told, I think the CPU cycles would be better spent elsewhere.

My theory is that the name giving event would allow us to reduce to number of checks, to the number of naming events actually happening within a tick, instead of all the events happening in that tick. The rumor system isn't my first concern, I guess it should be able to pass on reputations causing changes in what strings creatures are most referred by other creatures... but I would not "freak out" if names propagated in lightspeed; that was somebody else. To be honest the fake identity and rumor system confuse me more than anything else, so I failed at getting my point across: if the game can really simulate fakeness it should, but else it's no drama if the player is in the same position as a movie viewer (you know the killer is behind the curtains but no matter how loud you scream the actor won't hear you).

I don't play Adventurer mode, so I'm not sure if you're referencing that. Is there part of the game that actually tries to translate dwarven (or any other of the DF languages) into English? I doubt that's possible, as the in-game vocabularies are very incomplete, missing tons of words necessary to create a complete sentence. They're fine for their current purposes (just the names of people, cities, landforms, etc), but you can't actually say anything in them.

Legendsviewer and Legendsbrowser only reference Urist Battleball, while the game oftentimes only shows me Nokzambibar and I don't necessarely want to go look at his name trough another menu.

[GoblinCookie]

A clan is a family of related people who value the surname that unites them. This entire thread is about how best to give related dwarves family-based surnames. And literally the very first reply on this thread is you stating that family names are a waste, and random names are the way to go. As for my 'lack' of cooperation with implementing clans, who's the one compiling example name structures?

Yet clans have *nothing* to do with surnames.  As we discussed earlier, surnames were first used by Feudal houses, the purpose being to distinguish those who are the true members of the household from those who are simply living in the household.  Surnames were then extended to common folk, in Europe at least this happened after the period this game is based off. 

So if we are going to go all historical about it, surnames 'worked' initially because each household of importance had a surname and because whenever new households were set up the surnames changed accordingly we did not get the proliferation of a few surnames throughout society.  Once surnames became truly hereditary outside of the context of the household, so the children of Smith are called Smith even if they go off to form their own independent household, then whole system stops working because it no longer reveals information to anyone that you did not already know. 

Whether there's an actual worldgen event where the founders of a culture have reasons to sit down & literally decide "Okay, we're going to oppress X caste of our citizens", or whether Toady just simulates that with a much simpler dice roll, is largely invisible (and likely quite insignificant) to the player.

None of it is ever insignificant, unless the player is quite the heartless psychopathic bastard, which he may well be.  If we want things to be fixed and unchanging, then what you say is basically true, we can just have things set randomly to begin with and then be forced to live with it.  It is when we have new oppressions emerging and/or old one's being overcome then the "folks getting round a table deciding to oppress Group X" starts to verge on the ridiculous. 

Is that . . . scorn? You're aware that we're discussing improvements to a computer game, yet when I reduce descriptions of text strings to a representation of abstract data that more closely approximates code, you express . . . surprise? that I should do such a thing? 

It is odd to talk in French in England, even if the folks could theoretically go learn to speak French. 

Ehh, maybe. The way I've got your system described right now, dwarves have three names all their own: 1[] and 2[] are their given names (random, the way you like it), and 5[] is their half of the "marriage name", also random. Which of those, precisely, are you calling your dwarf's "personal name"? Regardless, in the current description, 5[] isn't (necessarily) used until the dwarf marries--each one then takes their spouse's 5[] name as their 6[]. This means that, if there's no dad, then mom isn't going to have a 6[] to append to her 5[], and the bastard child is going to be missing the names of ALL of his paternal ancestors (unless the mother just writes her own name in there, over & over).

I don't blame you for not understanding this bit, it is a bit confusing even to me sometimes.  The thing is that you don't inherit the family name of your parents, you inherit the marriage name of your parent and all the other generations are marriage names also.  The 'family name' is simply the marriage name of your parents, appearing after your own marriage name in order so we can still tell whose children we are dealing with should those children get married. 

To put it another way, it never happens that anyone inherit the family names of either of their parents, this is how I avoid the issue of gender.  You inherit the marriage name created when your parents married and also inherit the marriage names created when your each pair of grandparents married, you do not inherit the family names of your parents (which are the marriage name of their lines grandparents). 

The personal name is the present last name system, which has now replaced the first names we presently use, the cultural functionality of the first names having been replaced with a Civ+Site name where applicable.  In cases of illegitimacy, we take the mother's personal name (or for foundlings potentially the father's) and use that *as* the absent marriage name.

Hence there will never be the gaps you refer to in the lineage, since all that happens is that we fill the slot that would normally be filled by the marriage name with that of the mother. 

All answers are possible, and indeed desirable, for increased cultural variation.

The key concern however is to avoid the situation ever arising where potentially thousands of names will have to be redone in an instant because the culture changed it's naming system.  To a certain extent we also want to keep down the size of files as well, so we don't want absolutely enormous names for every eventuality even if nobody ever sees them. 

If some user comes on the forums and starts talking about bizarre stuff like "I can't wait until Toady finally implements flight, and shapeshifting," then yes, I would automatically give that user the benefit of the doubt and assume that they must be talking about some creatures other than dwarves, because otherwise the suggestion would be absurd.
But when that user is you, GoblinCookie? I'm sorry, but at this point I just can't take anything for granted.

I am not sure I should take the final point as a compliment or as an insult. 

I am sure the general plan is to have creatures other than dwarves become citizens of our fortress and also to develop systems of reproduction beyond everyone being either a placental mammal or a chicken.  There are rather few creatures that reproduce parthagenetically in DF, ants are one of them since male ants oddly do not have fathers (but female ones do), but ant people living in your fortress is I would hope ultimately on the cards.  On their own it would not be a problem since we would not use family names for ant-people anyway, but if ant people become 'dwarves' culturally then it gets interesting since the male ant people do not have fathers. 

My system deals with this fairly well, since our ant person queen is married to a ant person male and their children regardless of whether the male one's have fathers will be called by the marriage name of their mother OR the personal name of the mother, as with illegitimate children either would work.  The really hard part is when we have creatures that reproduce impersonally, as certain fish do in which the females will deposit the eggs and the males will then fertilize the eggs without ever meeting the mother than laid them.  The same also applies to plump helmet men presumably, in this case the reproduction is presumably that they shed spores which turn into baby plump helmet men. 

But we can use the system for foundlings there, the adopted parent of the impersonally reproduced creature imparts their own personal name as the family name of their adopted child.

[Bumber]

I believe vampires don't simply make up a fake name, they actually assume the identity of one of their victims. So their fake name should also include a pointer to the real (dead) dwarf's background, so they can lie convincingly without requiring any additional memory space. (Of course, how well the vampire succeeds at this is for another thread.)
As for aliases other than vampires, I lack sufficient knowledge of examples to say how they might be handled.

"Rumors of my death at the hand of a vampire were greatly exaggerated!"

I'll have to look into the new aliases when I get the chance.

[...] if the game can really simulate fakeness it should, but else it's no drama if the player is in the same position as a movie viewer (you know the killer is behind the curtains but no matter how loud you scream the actor won't hear you).

Unfortunately for the killer, the overseer can, in fact, orchestrate their demise.

[SixOfSpades]

I'm not suggesting that their string is the same than the event it relates to. For example Urist Familyname has gained the title "Nokzambibam-" (battleball) "-zora" (killer of) for defeating the megabeast Ithi Questbreaches. It's just an arbitrary random string. As I said earlier, Imagine them obscure references that don't translate well into english.

I really can't agree with it being some random string. If I kill a dragon named Albert Trapfurnace Funnywalk, I can see taking the name Albertkiller, or Furnacekiller, or most likely Dragonkiller . . . but not Bucketkiller, or Cloisterkiller, or Inkkiller, whether the words are translated or not. If the title is going to be limited to just 1 or 2 words (with or without a suffix), then it would best say the correct type of creature, rather than attempt to name the precise one.

"Rumors of my death at the hand of a vampire were greatly exaggerated!"

Just about the dumbest thing a vampire could do is to assume the identity of one of their victims, and then emigrate to a fort that houses a friend or family member of the dwarf they're impersonating. If they're smart enough to avoid that particular pitfall (and they are . . . aren't they?), then they should also be smart enough to avoid taking the name of someone they failed to kill. (Then again, should it really matter? The safest way to avoid family members would be to travel a long distance, ideally into a different civilization, each time you changed your identity. If someone in a completely different country just "happened" to have your name, it's not likely you'd even hear about it, let alone do anything.)

Yet clans have *nothing* to do with surnames.

Clans have pretty much everything to do with surnames. Stop it.

Whether there's an actual worldgen event where the founders of a culture have reasons to sit down & literally decide "Okay, we're going to oppress X caste of our citizens", or whether Toady just simulates that with a much simpler dice roll, is largely invisible (and likely quite insignificant) to the player.

None of it is ever insignificant, unless the player is quite the heartless psychopathic bastard, which he may well be.  If we want things to be fixed and unchanging, then what you say is basically true, we can just have things set randomly to begin with and then be forced to live with it.  It is when we have new oppressions emerging and/or old one's being overcome then the "folks getting round a table deciding to oppress Group X" starts to verge on the ridiculous.

I agree with the first part of that, whether Toady creates a "Social Bias" type of civ-level event or not will make little difference to the player, if the only time it ever gets used is just after worldgen. But, if it can be used more often (especially if it's witnessed or even caused by the player), what's so "ridiculous" about simulating events such as Europe's Catholic/Protestant spasms, or America's Civil Rights Era? But still, as this issue isn't really relevant to names, I won't dwell on it.

I am not sure I should take the final point as a compliment or as an insult.

To put it bluntly, it was an insult. It meant that I have seen you argue too many points that I found ludicrous for me to have any remaining faith in your powers of judgement. But despite this, I am still trying to represent your naming system as accurately as I can, because I feel that ideas deserve to be approved or rejected on their own merits, not those of their creator(s). So I'll keep asking questions about your convention until I'm sure it's the way you want it. (Or, until you tell me not to, you'd rather handle it yourself--you certainly have that right.)

I don't blame you for not understanding this bit, it is a bit confusing even to me sometimes.  The thing is that you don't inherit the family name of your parents, you inherit the marriage name of your parent and all the other generations are marriage names also.  The 'family name' is simply the marriage name of your parents, appearing after your own marriage name in order so we can still tell whose children we are dealing with should those children get married.   

To put it another way, it never happens that anyone inherit the family names of either of their parents, this is how I avoid the issue of gender.  You inherit the marriage name created when your parents married and also inherit the marriage names created when your each pair of grandparents married, you do not inherit the family names of your parents (which are the marriage name of their lines grandparents).

I can't find how my transcription of your system (with generation tracking level = 2) deviates from that in any way. Copied from Page 5:
1 = Given name, random style.
2 = Given name, random style.
3 = First word of the individual's home civilization's name
4 = Second word of the individual's home settlement's government's name
5 = Given name, random style. When combined with [6], creates the individual's "marriage two-string". Can be left null until marriage.
6 = Event type, 4th variation (marriage), pointing at the spouse's 5th name element. If the dwarf has no spouse, this name is null.
7 & 8 = The marriage names of the dwarf's parents, older one first.
9 & 10 = The marriage names of the dwarf's maternal grandparents.
11 & 12 = The marriage names of the dwarf's paternal grandparents.
All inherited names are marriage names. The marriage name of your parents appears after your own (and your spouse's) marriage name. You also inherit the marriage names created when each pair of your grandparents married, but not (with generations set to 2) your parents' grandparents. As far as I can tell, this system matches your specifications. Which is why it confuses me when you say

In cases of illegitimacy, we take the mother's personal name (or for foundlings potentially the father's) and use that *as* the absent marriage name. . . . Hence there will never be the gaps you refer to in the lineage, since all that happens is that we fill the slot that would normally be filled by the marriage name with that of the mother.

Using one of the mother's given names (1 or 2) would be all right to fill the empty slot 6 left by an unknown father. But what about slots 11 and 12, the marriage names of the absent father's parents? Since you say there will be no gaps, where precisely should the computer look for names to fill those slots? Sure, you can use the mother's 1 and 2 as a stopgap . . . but what if the generational level is raised to 3, and society expects you to be able to mention your father's grandparents as well? That's 4 more name elements for mommy to fill in . . . what does she write, there?

The key concern however is to avoid the situation ever arising where potentially thousands of names will have to be redone in an instant because the culture changed it's naming system.  To a certain extent we also want to keep down the size of files as well, so we don't want absolutely enormous names for every eventuality even if nobody ever sees them.

Realistically, in most cases there's no need for most people's names to change at all. Suppose the change came from the top down--either the ruling Chinese dynasty was replaced with a Mongol one, or the Pharaoh decided there was actually only one god and changed his name accordingly, or whatever. In such cases, the only people to actually change their names would be those with (relatively) direct contact with the ruler(s)--mostly in order to curry favor and show deference to his whims. Everybody else, meanwhile, would simply name their babies in the new style, while the old one gradually died out. The game can easily keep pace with a civ-wide name change taking place over the span of an entire generation.

[dragdeler]

I really can't agree with it being some random string. If I kill a dragon named Albert Trapfurnace Funnywalk, I can see taking the name Albertkiller, or Furnacekiller, or most likely Dragonkiller . . . but not Bucketkiller, or Cloisterkiller, or Inkkiller, whether the words are translated or not. If the title is going to be limited to just 1 or 2 words (with or without a suffix), then it would best say the correct type of creature, rather than attempt to name the precise one.

Yes we could fiddle with spheres, or try to match at least one string, it could be very nice, but I think in the long run names are going to be very repetetive that way. On the other hand: if we had a dictionnary worth of vocabulary, that would be very different.

[Bumber]

"Rumors of my death at the hand of a vampire were greatly exaggerated!"

Just about the dumbest thing a vampire could do is to assume the identity of one of their victims, and then emigrate to a fort that houses a friend or family member of the dwarf they're impersonating. If they're smart enough to avoid that particular pitfall (and they are . . . aren't they?), then they should also be smart enough to avoid taking the name of someone they failed to kill. (Then again, should it really matter? The safest way to avoid family members would be to travel a long distance, ideally into a different civilization, each time you changed your identity. If someone in a completely different country just "happened" to have your name, it's not likely you'd even hear about it, let alone do anything.)

The implication was more like:
Urist McCitizen: Tell me about yourself.
Urist McMayor: One year ago, Urist McVampire killed me.
Urist McCitizen: It was inevitab-- Wait a second...
Urist McMayor: Uhhh... I got better?
Urist McCitizen: Works for me!

It needn't be the vampire that blows the cover. Basically anyone talking about them could bring up the rumor. I guess vampires would need to chose aliases of victims nobody saw die.

[GoblinCookie]

Clans have pretty much everything to do with surnames. Stop it.

I have already explained that clans did not widely use surnames at all for most of their history and when they did it was imposed from outside by another forms of society.  You have so far made no attempt to establish any relationship between the two things, but just stamp your feet and declare it so. 

I agree with the first part of that, whether Toady creates a "Social Bias" type of civ-level event or not will make little difference to the player, if the only time it ever gets used is just after worldgen. But, if it can be used more often (especially if it's witnessed or even caused by the player), what's so "ridiculous" about simulating events such as Europe's Catholic/Protestant spasms, or America's Civil Rights Era? But still, as this issue isn't really relevant to names, I won't dwell on it.

Because those two events are not the same.  There is a difference between an internal conflict between Protestants and Catholics, which are two rival groups competing for power and a situation like the racist laws and policies of the Civil Rights era.  Both are quite difficult to replicate in DF, but they are quite different.  The problem with religious conflict along the lines of the Reformation is basically the same as that of simulating rebellions in general, why doesn't the rebellion just get swiftly crushed or effortlessly sweep away the old order?

Racism is quite different to that because the Catholics do not want there to be Protestants and vice versa.  Racists on the other hand are quite insistent on maintaining in existence of separate races, hence why they are so threatened by racial intermarriage (and made it illegal).  The problem with racism is why as a powerful person invent a permanently discontented underclass that will perpetually threaten your power when you can give everyone equal status as *your* lowly minions?

But no, it is very much relevant to the question of surnames, because the origin and original purpose of surnames is very much related to class. 

To put it bluntly, it was an insult. It meant that I have seen you argue too many points that I found ludicrous for me to have any remaining faith in your powers of judgement. But despite this, I am still trying to represent your naming system as accurately as I can, because I feel that ideas deserve to be approved or rejected on their own merits, not those of their creator(s). So I'll keep asking questions about your convention until I'm sure it's the way you want it. (Or, until you tell me not to, you'd rather handle it yourself--you certainly have that right.)

Well, that's charming  and also against the forum rules.  You also seem to think that the merits of the creation can somehow be separated from the merits of the creator, that people can legitimately steal all my ideas and still disparage my 'intellectual judgement' all the same.

I can't find how my transcription of your system (with generation tracking level = 2) deviates from that in any way. Copied from Page 5:
1 = Given name, random style.
2 = Given name, random style.
3 = First word of the individual's home civilization's name
4 = Second word of the individual's home settlement's government's name
5 = Given name, random style. When combined with [6], creates the individual's "marriage two-string". Can be left null until marriage.
6 = Event type, 4th variation (marriage), pointing at the spouse's 5th name element. If the dwarf has no spouse, this name is null.
7 & 8 = The marriage names of the dwarf's parents, older one first.
9 & 10 = The marriage names of the dwarf's maternal grandparents.
11 & 12 = The marriage names of the dwarf's paternal grandparents.
All inherited names are marriage names. The marriage name of your parents appears after your own (and your spouse's) marriage name. You also inherit the marriage names created when each pair of your grandparents married, but not (with generations set to 2) your parents' grandparents. As far as I can tell, this system matches your specifications. Which is why it confuses me when you say

In cases of illegitimacy, we take the mother's personal name (or for foundlings potentially the father's) and use that *as* the absent marriage name. . . . Hence there will never be the gaps you refer to in the lineage, since all that happens is that we fill the slot that would normally be filled by the marriage name with that of the mother.

Using one of the mother's given names (1 or 2) would be all right to fill the empty slot 6 left by an unknown father. But what about slots 11 and 12, the marriage names of the absent father's parents? Since you say there will be no gaps, where precisely should the computer look for names to fill those slots? Sure, you can use the mother's 1 and 2 as a stopgap . . . but what if the generational level is raised to 3, and society expects you to be able to mention your father's grandparents as well? That's 4 more name elements for mommy to fill in . . . what does she write, there?

I see the confusion now. 

The mother has a full set of grandparents already.  We are only replacing a singular name slot (one that which would normally be filled by the name created when the parents marriage) with that of the mother in this case.  That would mean that the illegitimate child's paternal grandparents will be those of their mother, not their own.  The point is however that no nulls end up in the system, but we instead end up with a paternal lineage by which the absent real father is treated as though he were the half-brother of the mother, so that the babies parental line follows that of his grandfather but not his grandmother. 

Realistically, in most cases there's no need for most people's names to change at all. Suppose the change came from the top down--either the ruling Chinese dynasty was replaced with a Mongol one, or the Pharaoh decided there was actually only one god and changed his name accordingly, or whatever. In such cases, the only people to actually change their names would be those with (relatively) direct contact with the ruler(s)--mostly in order to curry favor and show deference to his whims. Everybody else, meanwhile, would simply name their babies in the new style, while the old one gradually died out. The game can easily keep pace with a civ-wide name change taking place over the span of an entire generation.

That is an idea that would generally work, except for immortal elves and other creatures that are very long-lived; the realism of such a model however is lacking.  Any change carried out from the top is typically uniform in nature, the present surnames we have now were imposed from the top and were uniformly adopted instantaneously, with some crossover with whatever names people were informally using in various contexts before. 

I guess what you are saying only works if it begins at the top but is not deliberately imposed on everyone else as real-life surnames were.  In that situation what you are describing might be realistic in real-life, but in DF it is not.  That is because there are not many layers between the ruler and the people, everyone in the ruler's site is going to wish to 'show deference to his whims', (even though it isn't his explicit will at all), since everyone who lives in the rulers site has 'known the person since they were born' or 'known the person all their life' (I am quoting adventure mode here). 

Your system is sitting on a knife-edge between compulsory and voluntary, it works if it is somehow neither.  But most things will realistically either be compulsory (everyone adopts it at the same time) or will be voluntary (in which case things are adopted exponentially) and both potentially lead to us having to make lots of calculations in a short time frame. 

[SixOfSpades]

. . . If the title is going to be limited to just 1 or 2 words (with or without a suffix), then it would best say the correct type of creature, rather than attempt to name the precise one.

Yes we could fiddle with spheres, or try to match at least one string, it could be very nice, but I think in the long run names are going to be very repetetive that way. On the other hand: if we had a dictionnary worth of vocabulary, that would be very different.

Well, if we make titles far more difficult to achieve, the repetition potential goes down considerably. As for the vocabulary: Things are happening. That's all I'll say. 

It needn't be the vampire that blows the cover. Basically anyone talking about them could bring up the rumor. I guess vampires would need to chose aliases of victims nobody saw die.

More importantly, victims with very few people who could recognize them. If you kill & impersonate some reclusive hermit who lives out in the woods, nobody's going to say, "Hey, wait . . . YOU'RE not that guy I've never seen before who never comes into town!" In any event, the vampire would still be very prudent to relocate a considerable distance away, to outrun rumor.

Well, that's charming  and also against the forum rules.

You have so far made no attempt to establish any relationship between the two things, but just stamp your feet and declare it so.

It can hardly be a violation of the rules, or indeed even etiquette, to state that I listen carefully to what you say, and try to give every point just as much recognition as it merits. It is also not uncivil of me to say that I consider your credibility on this forum to be greatly diminished. As far as my disregarding your claims is concerned, it's not that I don't need to prove my points--it's that I don't need to prove my points, to you, personally. I have seen you cheerfully derail several threads with your off-topic tangents, and I decline to waste more of the forum's attention on such pursuits. This is not to say that I find you unintelligent--far from it. I freely admit that you usually raise valid concerns that deserve at least consideration. You use the language well, often with fair arguments . . . it's what you argue for that I sometimes find exasperating, and even inexplicable.

You also seem to think that the merits of the creation can somehow be separated from the merits of the creator, that people can legitimately steal all my ideas and still disparage my 'intellectual judgement' all the same.

Intellectual property theft is pretty much the opposite of what I do. I assure you, I have absolutely NO intention of associating my name with your naming convention.

That would mean that the illegitimate child's paternal grandparents will be those of their mother, not their own.

Okay, I interpret that to mean that the child lists the mother's paternal grandparents as if they were the child's own paternal grandparents. Is that exactly what you meant? Because I can make that work (provided that Generations =2, at least), but it conflicts with what you said next:

The point is however that no nulls end up in the system, but we instead end up with a paternal lineage by which the absent real father is treated as though he were the half-brother of the mother, so that the babies parental line follows that of his grandfather but not his grandmother.

In the previous quote, the mother reached up (to her own grandparents' names) and pulled them down to be her child's grandparents. But in this one, you say the father is treated as the same generation as the mother--meaning, the grandparents' names stay where they are. And why would the baby's "parental" line ignore his grandmother, when it's his father who is presumably anonymous?

But regardless, please, just fill this out. Given an unknown father, and an unwed mother whose name is:
Mebzul Dodok TreatyLobster Ezum ItonStakud DatanBer CogLolor
What, precisely, is the child's name? I'm only 100% sure about this much:
random random TreatyLobster random6[null until marriage] 7[Ezum]8[  ?  ] 9[Iton]10[Stakud] 11[  ?  ]12[  ?  ]
(I'm assuming a single parent takes the "older parent" position of Slot 7 by default.) I will convert your answer into an additional "Illegitimacy" variation in the list of naming systems.

The game can easily keep pace with a civ-wide name change taking place over the span of an entire generation.

That is an idea that would generally work, except for immortal elves and other creatures that are very long-lived;

That's true, I overlooked the immortals. Well, it's still a reasonable option for the short-lived races, at least . . . the elves & goblins will have to choose other means. Or, potentially, not change at all: Tolkien's elves, at least, are portrayed as being very resistant to change, and the very nature of (near-)immortality lends itself well to the idea of being a staunch traditionalist, not apt to change one's name at the drop of a hat.

[Starver]

In any event, the vampire would still be very prudent to relocate a considerable distance away, to outrun rumor.

Reminds me of:

[...] 'Are you really from Genua?'

'Are you really from Pseudopolis?' Madam smiled at him. 'I find, personally that it pays never to be from somewhere close at hand. It makes life so much easier. [...]'

The first voice being of someone actually assuming a dead person's identity (though they hadn't actually done the killing), and the second is of someone who is probably not a vampire (canon doesn't definitively say, but certainly no more a vampire than John Not-A-Vampire-At-All Smith and that's good enough for me!), and we certainly can't trust surnames when it comes to aliased people, so this is a definite foray off-topic.


Speaking of which,  once we're arguing about how we're arguing, it probably puts the cherry on the top of the circular arguments that aren't going to help anybody, least of all the main characters in the Toady And ThreeToe Show who might well be discouraged from implementing any version of this whole idea, on the basis that there's plenty of vocal opinion that would consider any direction they went as being utterly wrong.

Can I suggest that there's not many more arguments/counter-arguments/counter-counter-arguments left that haven't yet been said, at least in passing. Even the meta-arguments are getting stale, IMO, but whether you take the same view or not I leave up to you.

[GoblinCookie]

It can hardly be a violation of the rules, or indeed even etiquette, to state that I listen carefully to what you say, and try to give every point just as much recognition as it merits. It is also not uncivil of me to say that I consider your credibility on this forum to be greatly diminished. As far as my disregarding your claims is concerned, it's not that I don't need to prove my points--it's that I don't need to prove my points, to you, personally. I have seen you cheerfully derail several threads with your off-topic tangents, and I decline to waste more of the forum's attention on such pursuits. This is not to say that I find you unintelligent--far from it. I freely admit that you usually raise valid concerns that deserve at least consideration. You use the language well, often with fair arguments . . . it's what you argue for that I sometimes find exasperating, and even inexplicable.

What you said earlier.

To put it bluntly, it was an insult. It meant that I have seen you argue too many points that I found ludicrous for me to have any remaining faith in your powers of judgement. But despite this, I am still trying to represent your naming system as accurately as I can, because I feel that ideas deserve to be approved or rejected on their own merits, not those of their creator(s). So I'll keep asking questions about your convention until I'm sure it's the way you want it. (Or, until you tell me not to, you'd rather handle it yourself--you certainly have that right.)

Insulting people in a wordy officious fashion is no less insulting people and against the forum rules than doing so crudely.  The effect is quite the same, even if the means is less obvious.  Saying "I don't have to prove myself to the likes of *you*" when that is the person you are talking too is a pretty vicious insult and one that pretty much sums up half of what you are saying in the above quotes in fewer words.  Naturally it is better to leave it there on this point, since further discussion is now both seriously unwise and off-topic. 

As for the other half, I don't go around deliberately derailing threads.  It just happens to be the case that people tend to like to make specific suggestions and pretty much all suggestions are really about a few general issues about the DF economy and society.  Folks however don't like to think about that kind of stuff, they would rather just stamp their feet and demand that Feature X be implemented because They Say So and are only looking for as many more foot stampers to back them up as possible, so they prefer to complain about So and So ruining their party by actually revealing what the specific subject is really about than actually coming up with new ideas and hence looping things back from the general to the specific. 

They just want to shout I WANT THAT! as loudly as possible.

Intellectual property theft is pretty much the opposite of what I do. I assure you, I have absolutely NO intention of associating my name with your naming convention.

I never said that you were intending to do it, only that it legitimately follows from what you were saying.   

Okay, I interpret that to mean that the child lists the mother's paternal grandparents as if they were the child's own paternal grandparents. Is that exactly what you meant? Because I can make that work (provided that Generations =2, at least), but it conflicts with what you said next:

The point is however that no nulls end up in the system, but we instead end up with a paternal lineage by which the absent real father is treated as though he were the half-brother of the mother, so that the babies parental line follows that of his grandfather but not his grandmother.

In the previous quote, the mother reached up (to her own grandparents' names) and pulled them down to be her child's grandparents. But in this one, you say the father is treated as the same generation as the mother--meaning, the grandparents' names stay where they are. And why would the baby's "parental" line ignore his grandmother, when it's his father who is presumably anonymous?

The system follows the female grandparental lineage which is known and skips back to the first known male ancestor.  This is why I said that it is in effect the same as what would happen if somebody incestuously reproduced with their half-brother, specifically somebody that shared the same father but different mothers. 

There are no nulls in the system just as I said, since we don't have to worry about the male and female line until we end up with the grandparents.  The illegitimacy part however happens in the present generation, it involves swapping the marriage name which is normally created upon the parents marriage, with the personal name of the mother.  The main issue here is the point that if the father's name is not known how can we know who the paternal grandparents are, that is solved by simply treating the paternal grandparents of the mother as those of the baby. 

But regardless, please, just fill this out. Given an unknown father, and an unwed mother whose name is:
Mebzul Dodok TreatyLobster Ezum ItonStakud DatanBer CogLolor
What, precisely, is the child's name? I'm only 100% sure about this much:
random random TreatyLobster random6[null until marriage] 7[Ezum]8[  ?  ] 9[Iton]10[Stakud] 11[  ?  ]12[  ?  ]
(I'm assuming a single parent takes the "older parent" position of Slot 7 by default.) I will convert your answer into an additional "Illegitimacy" variation in the list of naming systems.

You have not indicated which of the above words fill which slot, also you have translated the third word into English but none of the other words, but that does not really matter.  I will impose the following order onto what you wrote.

1. Personal Name: Mebzul.
[Entity surname value 1]
2. Own Marriage Name: Dodok
3. Parents Marriage Name: TreatyLobster
[Entity surname value 2]
4. Paternal Grandparents Marriage Name: Ezum
5. Maternal Grandparents Marraige Name: ItonStakud
[Not relevant]
6. Name of Site Government+Civilization: DatenBer
7. Name of something else: CogLolor

The last two don't matter here, so our mother is Mebzul Dodok TreatyLobster Ezum ItonStakud.  She decides to name her illegitimate baby Kinzul, that is it's personal name.  The baby then ends up being called Kinzul Mebzul Ezum ItonStakud, as Mebzul is the mother's name.  Now to explain how things fit back together again, Kinzul marries a woman called Yoplin Yellowpage Ipolkin Loinclan.  The happy couple adopt GoblinSpade as their marraige name. 

1. Personal Name: Yoplin.
[Entity surname value 1]
2. Own Marriage Name: GoblinSpade
3. Parents Marriage Name: Yellowpage
[Entity surname value 2]
4. Paternal Grandparents Marriage Name: Ipolkin
5. Maternal Grandparents Marriage Name: Loinclan

So we end up with Kinzul GoblinSpade Mebzul Ezum ItonStakud and Yoplin GoblinSpade Yellowpage Ipolkin Loinclan, we fill the empty second slot with the newly created marraige name.  Now let's just have DatenBer increase their Entity surname value to 3, so we now have 7 family related names and have them produce a (legitimate) daughter called Ulop.  We end up with the following name. 

1. Personal Name: Ulop.
[Entity surname value 1]
2. Own Marriage Name:
3. Parents Marriage Name: GoblinSpade
[Entity surname value 2]
4. Paternal Grandparents Marriage Name: Mebzul
5. Maternal Grandparents Marriage Name: Yellowpage
[Entity surname value 3]
4. 1st Paternal Great-grandparents Marriage Name: Ezum (actually great-great-grandparents)
5. 1st Maternal Great-grandparents Marriage Name: ItonStakud
6. 2nd Paternal Great-grandparents Marriage Name: Ipolkin 
7. 2nd Maternal Great-grandparents Marriage Name: Loinclan

The baby is therefore called Ulop GoblinSpade Mebzul Yellowpage Ezum ItonStakud Ipolkin Loinclan.  The whole system works with only a small bit of deception, Ezum is actually the marriage name of the great-great grandparents rather than the great-grandparents, because we don't know who the grandfather of Ulop actually is.  Also, because nobody can tell at a glance that Mebzul is the name of her grandmother rather than a marraige name, the whole thing hides the illegitimacy while still allowing us to trace lineage backwards.

That's true, I overlooked the immortals. Well, it's still a reasonable option for the short-lived races, at least . . . the elves & goblins will have to choose other means. Or, potentially, not change at all: Tolkien's elves, at least, are portrayed as being very resistant to change, and the very nature of (near-)immortality lends itself well to the idea of being a staunch traditionalist, not apt to change one's name at the drop of a hat.

Well Tolkien is a great one for implementing cool stuff while seemingly not realising the practical implications of what he just did.  I want dwarves to live in cool underground fortress called Erebor, but why don't they just suffocate?  I want my elves to naturally live forever, but doesn't that mean that there are elves about the place with thousands of children+grandchildren?  Why don't we read about the ventilation shafts of Erebor and Galadriel's thousands of grandchildren?

Does living forever actually have those consequences?  To the oldest elves things like tradition are effectively meaningless because they were around when the traditions were newly invented and hence were not traditions yet.  Being resistance to change is also basically suicidal if you live forever, the world of baby-Galadriel is afterall going to so be completely different to the world she presently lives in that to wax on about how the "world is changing" as she does in the movie is actually funny.

[dragdeler]

Now I kind of get it... Except, why? I mean like, whyyyyyyyyy (sorry)

Atm there are no such things as illegitimate children, and DF's definition of incest is very narrow. And I think if we amend the fact that widows don't remarry, and also make some races/civ's polygamous, and/or detach procreation from marriage, that's fine. Don't we exclude cultural differences if we settle on that kind of system? Please don't tell me you have at least one of those for each race, or that nuances are expressed by something that would require to check the name and the sex of every ancestor.
You can just as easily signify illegitmacy by not giving a child the name it's supposed to have in it's culture. As in: it's a stigma or an honour or whatever, to not carry the familyname of your mom/dad as everybody else does around you. In fact, if you just pick the name of any ancestor by the same naming convention (patriarchal civ's, matriarchal civ's), it doesn't even matter if they're married; but you'd still have the possibility to simply pick a new familyname to signify more nuances (such as bastard).

If you're just bothered by the unoriginality of this father, mothername thing we could also add closest relative (like a caretaker), designated master (slaves, students), particular constellations combining 2 out of the 4 strings of the parents familynames etc. But I just don't see why it would be more practical to not put that kind of information into titles, and leave familynames be. In fact. I can't imagine how drowning the player with that kind of information allows to bypass the help of third party tools, or a glance at legends mode, in a practical and intuitive way... at all.

Also if the supposed merit of the system is to be less arbitrary and or more reliable, interesting, relevant (whatever; you name it) than our real world inspirations, why did we discuss historical accuracy? Because I can't see how that would solve the whole fake ID thing, which is the only reason I can come up with.
Atm we create completly fake personas, I think vampires and other NPC do the same tough I'm not sure.  It would be cool to impersonate someone specific, but just... it wouldn't matter how both names, real and fake, "sound". Only which one it "shows" would matter. And as a sidenote flavor could be added by differenciating which parts of a name creatures use to refer to eachother.

Is that 12 strings I'm counting? Dude. why?