Your thoughts on nouns as distinct from objects?

(Björn Paulsen) #1

So should a noun actually be the same as the object itself?

This is more of a question about IF authoring systems in general, but I think it’s worthwhile to ask. I tend to use Inform 7, but every IF system thus far seems to share the same assumption, that being that the gender of the noun is equal to that of the object it represents. This holds true for English, for the most part, but in other languages it’s not.

In traditional maritime parlance, ships are considered female. Yet they are clearly nonliving objects. At any rate, suppose we X SHIP, then type BOARD HER. The result would differ depending on whether the ship object was defined as feminine or neuter. But it shouldn’t. Conceptually, when it comes things that are not adult humans, it is the noun that is gendered, not the actual object.

In the English, that’s academic. But consider a language like German or Swedish, where you constantly run into situations like the child / das Kind / barnet contra the boy / der Junge / pojken. Wouldn’t it be more logical for the parser to be aware of that distinction and handle it sensibly than to go, “no, there’s clearly no ‘it’ in sight” because the object in question is defined as a ‘girl’?

What do you guys think?

0 Likes

(Hanon Ondricek) #2

I’m not sure I understand the problem that differentiating a noun from a thing would solve in English. In Inform 7 it’s possible to set gender for purposes of correct pronouns. I’d imagine in another language translation the same convention could hold - “A tree is usually female” or whatever.

0 Likes

(Andrew Plotkin) #3

Are you imagining a situation where (in some language) the word “ship” is feminine but “boat” is masculine, but the author wants to use them as synonyms for the same game object?

It’s true this would require more modelling. The object’s gender should be the gender of its printable name, so that the game can display text properly. (Printing “You board [the boat]” should display the appropriate article.) But then you’d want to (also) attach gender flags to words, so that input pronouns could be parsed properly, as in your example above.

However, this is potentially messy. Imagine this sequence:

Now does HER refer to “boat”, the last object mentioned, or “ship”, the last term used by the player? If the genders are different, this has the potential to wrong-foot the player.

(I never liked the Inform 7 convention that pronouns are set whenever an object name is printed. But it is the current Inform 7 convention.)

0 Likes

#4

In the German translation of I6/I7 we have the so-called changing gender:
Understand “mantel” or “jacke[f]” as the coat.

So “Mantel” is male, but synonym “Jacke” is female. This occurs quite often in the german language.
It’s a rather complicated thing to implement but it works fine since years :wink:

0 Likes

(Nathanaël Marion) #5

That use of an “[f]” token is quite clever! However, what if the player types a complicated expression rather than a single word? In your example, a player could type >X MANTEL JACKE to refer to the coat. What pronouns should be set then?

In French, if we have a “manteau” object (masculine) with the synonym “veste” (feminine), but the coat belongs to me, a player could type >X VESTE DE NATRIUM729. “Veste” is feminine but I am masculine. How the parser will distinguish wich word is the main word and set the pronouns accordingly? We would need a token for words that do not affect the gender, and I think it can quickly become tiresome for an author.

In French, we didn’t bother and the only gender that matters is the one of the object. It has never caused any issues as far as I can tell. Once one knows that the convention is to use the gender of the “canonical” term used by the game for the object, everything is fine.

(But then, we are not a lot of people playing French parser games, and I believe most of us do not use pronouns, because it’s not a widely known feature, a bit like OOPS. In fact, there was a French game that was intensely tested, but had a scenery that had the wrong gender, and nobody found it; I found it years later because I tried to refer to it with a pronoun and it didn’t work.)

0 Likes

(Piergiorgio d'errico) #6

well, as Naval historian, I must note that sooner or later the acronym HMS will change genre, but as Naval tradition goes, in Italian we use male for our warship and ships, and on top of it, many ship types are at the feminine (e.g. Il Grecale (male referred to ship)/ la fregata Grecale (feminine referred to ship type) )

[long and winded exposition on ship naming, declination and sailor’s mindset elided] and let’s not debate on ships being inanimate… actual sailors will strongly disagree, trust me.

0 Likes