Is English the ultimate language for IF?

10 views
Skip to first unread message

Karl Filenius

unread,
Aug 14, 2001, 10:16:22 AM8/14/01
to
A previous thread about translating IF to Swedish got me thinking about the
pros and cons of English as a language for IF.

What I'm looking for are purely linguistical merits.


pros:

1) English is a poorly inflected language. You don't have to worry about if
the nouns "dwarf" and "axe" in "hit dwarf with axe" are dative, acusative or
instrumentalis. You simply use the nominative. Inflection might complicate
things in a language like German and it would most certainly complicate
things in languages like Polish, Russian or Finnish.

2) English uses the latin alphabet. Anyone who thinks this is a trivial
detail should try to write a game in Russian. (By the way, has this been
done?) Futhermore, English has no diacritical signs.

3) The imperative is identical with the infinitive. The imperative is not
conjugated.

4) English rarely uses suffixes. The definite and indefinite forms are
distinguished by articles and not, as in Swedish and other Scandinavian
languages, by suffixes.

5) There is no grammatical gender in English. You don't have to worry about
making sure that the adjective has the same gender as the noun it describes.


cons:

I can't think of any.

Branko Collin

unread,
Aug 14, 2001, 2:07:43 PM8/14/01
to
On Tue, 14 Aug 2001 14:16:22 GMT, "Karl Filenius"
<karl_f...@hotmail.com> wrote:

>A previous thread about translating IF to Swedish got me thinking about the
>pros and cons of English as a language for IF.
>
>What I'm looking for are purely linguistical merits.

In order to that, I should be able to compare some appropriate
languages, which I cannot. (I forgot too much).

However:

>pros:

>2) English uses the latin alphabet. Anyone who thinks this is a trivial
>detail should try to write a game in Russian. (By the way, has this been
>done?) Futhermore, English has no diacritical signs.

English does use diacritical signs, in loan words. For instance:
re'sume'.

>3) The imperative is identical with the infinitive. The imperative is not
>conjugated.

This is not necessarily an advantage, for instance in games where you
want to be able to use both, but with different effects. (I cannot
think of an example, though, so this may be purely academic.)


>4) English rarely uses suffixes. The definite and indefinite forms are
>distinguished by articles and not, as in Swedish and other Scandinavian
>languages, by suffixes.

What do you mean with 'suffixes'? English uses lots of suffixes in
verbs to indicate number in verbs or to indicate genitive in nouns.

>5) There is no grammatical gender in English. You don't have to worry about
>making sure that the adjective has the same gender as the noun it describes.

There are some gender issues, for instance when you refer to objects
or ships.

--
branko collin
"mooi woord, verorberen, help me eraen herinneren
da'k het vaeker gebruik"
izak van langevelde in n.e.s.

Stephen Bond

unread,
Aug 14, 2001, 3:49:18 PM8/14/01
to
Karl Filenius wrote:
> 2) English uses the latin alphabet. Anyone who thinks this is a trivial
> detail should try to write a game in Russian. (By the way, has this been
> done?)

Yes--see 'Suddenly the Trains Departed' by Anna Rostovtseva. It's very
good, apparently.

Stephen.
www.maths.tcd.ie/~bonds/


Magnus Olsson

unread,
Aug 14, 2001, 3:58:32 PM8/14/01
to
In article <Woae7.10711$z21.1...@newsc.telia.net>,

Karl Filenius <karl_f...@hotmail.com> wrote:
>pros:
>
>1) English is a poorly inflected language. You don't have to worry about if
>the nouns "dwarf" and "axe" in "hit dwarf with axe" are dative, acusative or
>instrumentalis. You simply use the nominative. Inflection might complicate
>things in a language like German and it would most certainly complicate
>things in languages like Polish, Russian or Finnish.

Agreed.

>2) English uses the latin alphabet. Anyone who thinks this is a trivial
>detail should try to write a game in Russian.

This is true, but only because almost all existing IF is written
in English - the existing tools and interpreters handle the English
character set best because that's what most of them were constructed
to do in the first place.

>Futhermore, English has no diacritical signs.

Not quite true: there are loan words such as naïve and outré.

>3) The imperative is identical with the infinitive. The imperative is not
>conjugated.

In other words, there is only one possible verb form to use for
commands, and this is the same form that is used in questions like
"what do you want to eat?"

>4) English rarely uses suffixes. The definite and indefinite forms are
>distinguished by articles and not, as in Swedish and other Scandinavian
>languages, by suffixes.

There is of course the third person singular 's' on verbs. And the
indefinite article is either "a" or "an".

>5) There is no grammatical gender in English. You don't have to worry about
>making sure that the adjective has the same gender as the noun it describes.

But pronouns are still gendered. Finnish is simpler, since there's
no difference between "him", "her" or "it".

>cons:
>
>I can't think of any.

Let's see:

There are some irregular plurals, so you can't just add an
's' to get the plural form of a noun.

There are lots of uncountable nouns that complicate the handling
of indefinite forms (you don't want "there's a flour here").

There are rather a lot of cases where plural nouns describe singular
objects and vice versa. There can even be cases where you have one
singular and one pluarl nound for the same object.

There are two possible ways of handling indirect objects, with
different word order: "give the dog the bone" or "give the bone
to the dog".

Compound verbs like "look up" complicate parsing: in "look up 'Zork'
in the dictionary" the verb is "look up", but in just "look up"
the verb is "look" and "up" is an adverb.

--
Magnus Olsson (m...@df.lth.se, m...@pobox.com)
------ http://www.pobox.com/~mol ------

Marek Teraszkiewicz

unread,
Aug 14, 2001, 4:44:38 PM8/14/01
to
"Branko Collin" <col...@xs4all.nl> skrev i meddelandet
news:3b7966f0...@news.xs4all.nl...

> On Tue, 14 Aug 2001 14:16:22 GMT, "Karl Filenius"
> <karl_f...@hotmail.com> wrote:
>
> >A previous thread about translating IF to Swedish got me thinking about
the
> >pros and cons of English as a language for IF.
> >
> >What I'm looking for are purely linguistical merits.
>
> In order to that, I should be able to compare some appropriate
> languages, which I cannot. (I forgot too much).
>
> However:
>
> >pros:
>
> >2) English uses the latin alphabet. Anyone who thinks this is a trivial
> >detail should try to write a game in Russian. (By the way, has this been
> >done?) Futhermore, English has no diacritical signs.
>
> English does use diacritical signs, in loan words. For instance:
> re'sume'.

In that case I stand corrected. However, I never felt the need to use
diacritical signs in a game.

> >3) The imperative is identical with the infinitive. The imperative is not
> >conjugated.
>
> This is not necessarily an advantage, for instance in games where you
> want to be able to use both, but with different effects. (I cannot
> think of an example, though, so this may be purely academic.)

Neither can I.

> >4) English rarely uses suffixes. The definite and indefinite forms are
> >distinguished by articles and not, as in Swedish and other Scandinavian
> >languages, by suffixes.
>
> What do you mean with 'suffixes'? English uses lots of suffixes in
> verbs to indicate number in verbs or to indicate genitive in nouns.

I meant that English uses suffixes much less than other languages.

> >5) There is no grammatical gender in English. You don't have to worry
about
> >making sure that the adjective has the same gender as the noun it
describes.
>
> There are some gender issues, for instance when you refer to objects
> or ships.

I can see what you're driving at but the fact that sailors and weapon
fetischists refer to their ships and guns using personal feminine pronouns
doesn't constitute grammatical gender.


English is the most suitable language for IF not only because it is the
lingua franca of the computer world but also because its grammar fits the
needs of the genre.

TheCycoONE

unread,
Aug 14, 2001, 5:45:56 PM8/14/01
to

"Magnus Olsson" <m...@df.lth.se> wrote in message
news:9lbvt8$aln$1...@news.lth.se...
I don't know... I think latin might be a nice language to do it in... if
enough people spoke latin. Absolutly no accents etc. and a shorter alphabet
(if it matters), they do use suffix to determine who does the action and
what case it's in, but everything would be in first person present anyway.
NO ARTICALS, programs never do them right anyway so this would be an
improvement (hence no more %thedesc%) No appostraphies! Sentences begin
with lowercase letters (nice for people who'd like to use an item
description mid way through a sentince) No question mark (is this an
improvement?)

That's enough of my latin rant for today.
quid est pater. (Who's your daddy?)

TheCycoONE


Carl Muckenhoupt

unread,
Aug 15, 2001, 2:09:27 AM8/15/01
to
If we're going to have a "latin rant", we should probably have an
Esperanto rant as well.

For a language that was specifically designed to be simple, Esperanto has
a fair bit going against it - notably, the use of diacriticals found only
in the Latin-3 character set. Also, both nouns and adjectives are
inflected by case and number. But the inflections are simple and
regular. Output would be no more difficult than in English. What about
input?

Esperanto's unusually regular morphology and grammar should make parsing
easy. Like Latin, however, Esperanto relies more on endings than word
order to determine a word's role in a sentence. I've contemplated
writing an Esperanto language definition file for Inform from time to
time, but I strongly suspect that it would be easier to write an
Esperanto parser from scratch.

Unfortunately, interpreting the command once you've parsed it can be
tricky. Esperanto is agglutinative; speakers routinely invent new words
by piecing together basic roots, and are easily understood by other
Esperantists. But it's probably reasonable to ask the player to curb
this urge, just as it's reasonable to ask that they only enter simple
sentences.

But there are other subtleties of interpretation that the programmer
would have to take into account, such as the affect of an accusative
ending on the object of a preposition. "Saltu en la boaton" means "jump
into the boat", whereas "saltu en la boato" means "jump while in the
boat". How many English-language games understand commands like "jump
while in the boat"? Unfortunately, in Esperanto, that's a very simple
command, separated from "jump into the boat" by only one letter. It
seems like a game that accepts the one should accept the other.

L. Ross Raszewski

unread,
Aug 15, 2001, 2:42:09 AM8/15/01
to
On Tue, 14 Aug 2001 18:07:43 GMT, Branko Collin <col...@xs4all.nl> wrote:
>>4) English rarely uses suffixes. The definite and indefinite forms are
>>distinguished by articles and not, as in Swedish and other Scandinavian
>>languages, by suffixes.
>
>What do you mean with 'suffixes'? English uses lots of suffixes in
>verbs to indicate number in verbs or to indicate genitive in nouns.

Perhaps he means that English is rather more limited in the number of
distinct suffix endings used by its grammar. -ing, -ed, and -(e)s
cover almost the entirety of grammatic suffixes.

>
>>5) There is no grammatical gender in English. You don't have to worry about
>>making sure that the adjective has the same gender as the noun it describes.
>
>There are some gender issues, for instance when you refer to objects
>or ships.
>

It is, however, never gramatically incorrect to refer to a ship
as "it"

I've had lots of grade school English teachers insist that English is
the hardest of all languages to learn because "It has so many figures
of speech" and "There's an exception to every rule". Somehow
overlooking that this is true of languages with much more complex
grammatical rules.


Magnus Olsson

unread,
Aug 15, 2001, 3:44:44 AM8/15/01
to
In article <tnj7lq9...@corp.supernews.com>,

TheCycoONE <cyc...@hotmail.com> wrote:
>I don't know... I think latin might be a nice language to do it in... if
>enough people spoke latin. Absolutly no accents etc. and a shorter alphabet
>(if it matters), they do use suffix to determine who does the action and
>what case it's in, but everything would be in first person present anyway.

You can't avoid cases by ahving everything in the first person present.
Cases have to do with which part of speech a noun is, so you still
have to handle all six cases, such as (with the words translated
to English)

"Floyd (vocative), give the broom (accusative) to the ensign (dative)."
"The ensign (nominative) is not interested"

>Sentences begin with lowercase letters

Depends on which convention you're using. There were no lowercase
letters in antiquity, and all modern texts I've seen start sentences
with upper-case letters.

Magnus Olsson

unread,
Aug 15, 2001, 4:37:55 AM8/15/01
to
In article <MPG.15e3dc22c...@News.CIS.DFN.DE>,

Carl Muckenhoupt <ca...@wurb.com> wrote:
>But there are other subtleties of interpretation that the programmer
>would have to take into account, such as the affect of an accusative
>ending on the object of a preposition. "Saltu en la boaton" means "jump
>into the boat", whereas "saltu en la boato" means "jump while in the
>boat".

Or just "jump in the boat" (which would be a bit unusual syntax,
perhaps, but have the same meaning).

The fact that cases change the maning of prepositions is common in
Indo-European langauges - in German, for example, "in" means "into"
when followed by an accusative, but "in", "inside" when followed by a
dative ("Ich springe ins (== in das) Boot" - I jump into the boat /
"Ich springe im (== in dem) Boot" - I'm jumping (while) in the boat).

And in Finnish (which is not Indo-European), you don't use
prepositions for things like this - there are so many cases that the
case alone of the boat is enough to tell whether you're in it, going
into it, moving away from it, doing something to it, using it for
something, and so on.

>How many English-language games understand commands like "jump
>while in the boat"? Unfortunately, in Esperanto, that's a very simple
>command, separated from "jump into the boat" by only one letter. It
>seems like a game that accepts the one should accept the other.

I'm not so sure of that. In English, the difference between singular
and plural is just one letter, and we do accept that a parser
which accepts "jump into boat" will respond to "jump into boats"
with "You can't see that here".

On the other hand, we've talked (in the "Swedish translation" thread)
about letting the parser ignore case.

How does German IF handle this? I suspect that the "ins"/"im"
distinction can be handled by treating them as two different
prepositions, but if the noun is feminine you have "in die"/"in der"
which is more tricky.

TheCycoONE

unread,
Aug 15, 2001, 7:15:07 AM8/15/01
to

"Magnus Olsson" <m...@df.lth.se> wrote in message
news:9ld99c$jhk$2...@news.lth.se...

> In article <tnj7lq9...@corp.supernews.com>,
> TheCycoONE <cyc...@hotmail.com> wrote:
> >I don't know... I think latin might be a nice language to do it in... if
> >enough people spoke latin. Absolutly no accents etc. and a shorter
alphabet
> >(if it matters), they do use suffix to determine who does the action and
> >what case it's in, but everything would be in first person present
anyway.
>
> You can't avoid cases by ahving everything in the first person present.
> Cases have to do with which part of speech a noun is, so you still
> have to handle all six cases, such as (with the words translated
> to English)
>
> "Floyd (vocative), give the broom (accusative) to the ensign (dative)."
> "The ensign (nominative) is not interested"

Good point! furcifer. Not only that the case changes between the sentence
and the responce, but also that there is no vocative case in latin.... but
they did give orders to people like that... now I'm confused. Anywho they
use the um object ending so it's like the accusative. I get to go back to
class in September. Hopefully they'll straighten it all out. Floydum...
Bobum... Bobis is not interested.

Thank you, this destroys that idea. Though it's perfectly acceptable for
simpler phasers, it's not better, and in fact worse on complex phasers.


>
> >Sentences begin with lowercase letters
>
> Depends on which convention you're using. There were no lowercase
> letters in antiquity, and all modern texts I've seen start sentences
> with upper-case letters.
>

My latin experiance so far has been of the Cambrige Latin Course which is
used by all the schools in my area. They all focus on latin of the 1'st
century AD at which time either the entire text was CAPS, or only the first
letter of a name, title, etc.


Magnus Olsson

unread,
Aug 15, 2001, 7:44:14 AM8/15/01
to
In article <W4ge7.14145$e5.20...@newsb.telia.net>,

Marek Teraszkiewicz <eraszk...@usanet.com> wrote:
>> >5) There is no grammatical gender in English. You don't have to worry
>about
>> >making sure that the adjective has the same gender as the noun it
>describes.
>>
>> There are some gender issues, for instance when you refer to objects
>> or ships.
>
>I can see what you're driving at but the fact that sailors and weapon
>fetischists refer to their ships and guns using personal feminine pronouns
>doesn't constitute grammatical gender.

There is still grammatical gender for personal pronouns.

>English is the most suitable language for IF not only because it is the
>lingua franca of the computer world but also because its grammar fits the
>needs of the genre.

You already seem very convinced. Why are you starting a debate if you
already think you know the answers?

Besides, I think you're sort of preaching to the choir - almost
all IF is written in English today, regardless of whether that's
the "ideal" language or not.

Richard Bos

unread,
Aug 15, 2001, 5:46:25 AM8/15/01
to
"TheCycoONE" <cyc...@hotmail.com> wrote:

> I don't know... I think latin might be a nice language to do it in... if
> enough people spoke latin. Absolutly no accents etc. and a shorter alphabet
> (if it matters), they do use suffix to determine who does the action and
> what case it's in, but everything would be in first person present anyway.

Would it? Why?

> NO ARTICALS, programs never do them right anyway so this would be an
> improvement (hence no more %thedesc%)

YM articles, I suppose? Not literally speaking, but that also means that
you have to use "ille" (not officially an article) to distinguish
between "any old dog" and "that dog you saw before".

> No appostraphies!

So? Are you implying that apostrophe's[0] are a problem?

> Sentences begin with lowercase letters

Erm, no, they don't. Either everything is CAPS, BECAUSE THE ANCIENT
ROMANS DIDN'T HAVE ANY LOWER CASE, or you decide that you're writing
Latin _now_, and use caps like everyone else.

> No question mark (is this an improvement?)

No. Besides, only classical Latin doesn't have question marks. More
recent Latin has used punctuation basically since it was invented.

> quid est pater. (Who's your daddy?)

_What_ is father, rather. "Who is your father" would be "_Qui_ est pater
tuus?".

Richard

Magnus Olsson

unread,
Aug 15, 2001, 10:19:04 AM8/15/01
to
In article <tnkmb8k...@corp.supernews.com>,

TheCycoONE <cyc...@hotmail.com> wrote:
>
>"Magnus Olsson" <m...@df.lth.se> wrote in message
>news:9ld99c$jhk$2...@news.lth.se...

>Good point! furcifer. Not only that the case changes between the sentence


>and the responce, but also that there is no vocative case in latin....

There is indeed a vocative case in Latin, as in "Et tu, Brute"
(nominative "Brutus").

>> Depends on which convention you're using. There were no lowercase
>> letters in antiquity, and all modern texts I've seen start sentences
>> with upper-case letters.
>>
>
>My latin experiance so far has been of the Cambrige Latin Course which is
>used by all the schools in my area. They all focus on latin of the 1'st
>century AD at which time either the entire text was CAPS, or only the first
>letter of a name, title, etc.

I don't think lower-case letters had been invented that early,
but I may be wrong.

I know that Greek lower-case letters weren't invented until early
medieval times, but modern editions of the classics usually use the
system you describe, with lower case everywhere except in proper
names.

TheCycoONE

unread,
Aug 15, 2001, 12:27:16 PM8/15/01
to

"Magnus Olsson" <m...@df.lth.se> wrote in message
news:9le0co$oue$1...@news.lth.se...

> In article <tnkmb8k...@corp.supernews.com>,
> TheCycoONE <cyc...@hotmail.com> wrote:
> >
> >"Magnus Olsson" <m...@df.lth.se> wrote in message
> >news:9ld99c$jhk$2...@news.lth.se...
>
> >Good point! furcifer. Not only that the case changes between the
sentence
> >and the responce, but also that there is no vocative case in latin....
>
> There is indeed a vocative case in Latin, as in "Et tu, Brute"
> (nominative "Brutus").
>
"Even you Brutus" I know we've done lots of sentences like this (including
the one mentioned) which I why I was so confused. We only learned
Nominative, Accusitive and Dative in class and we were told those were the
only ones, we learned word endings for all these cases, so I would assume
vocative has the same endings as one of the others and they were never
diffrentiated.

> >> Depends on which convention you're using. There were no lowercase
> >> letters in antiquity, and all modern texts I've seen start sentences
> >> with upper-case letters.
> >>

see below


> >
> >My latin experiance so far has been of the Cambrige Latin Course which is
> >used by all the schools in my area. They all focus on latin of the 1'st
> >century AD at which time either the entire text was CAPS, or only the
first
> >letter of a name, title, etc.
>
> I don't think lower-case letters had been invented that early,
> but I may be wrong.

Yes they were, Belimicus's will is written in lowercase on papyrus. approx.
84 AD

TheCycoONE

unread,
Aug 15, 2001, 12:29:48 PM8/15/01
to

"Richard Bos" <in...@hoekstra-uitgeverij.nl> wrote in message
news:3b7a33a1....@news.worldonline.nl...

> "TheCycoONE" <cyc...@hotmail.com> wrote:
>
> > I don't know... I think latin might be a nice language to do it in... if
> > enough people spoke latin. Absolutly no accents etc. and a shorter
alphabet
> > (if it matters), they do use suffix to determine who does the action and
> > what case it's in, but everything would be in first person present
anyway.
>
> Would it? Why?
>
> > NO ARTICALS, programs never do them right anyway so this would be an
> > improvement (hence no more %thedesc%)
>
> YM articles, I suppose? Not literally speaking, but that also means that
> you have to use "ille" (not officially an article) to distinguish
> between "any old dog" and "that dog you saw before".
>
No such word, and they didn't difrentiate... your confused with french or
something.

> > No appostraphies!
>
> So? Are you implying that apostrophe's[0] are a problem?
>
> > Sentences begin with lowercase letters
>
> Erm, no, they don't. Either everything is CAPS, BECAUSE THE ANCIENT
> ROMANS DIDN'T HAVE ANY LOWER CASE, or you decide that you're writing
> Latin _now_, and use caps like everyone else.

WRONG!!!


>
> > No question mark (is this an improvement?)
>
> No. Besides, only classical Latin doesn't have question marks. More
> recent Latin has used punctuation basically since it was invented.
>
> > quid est pater. (Who's your daddy?)
>
> _What_ is father, rather. "Who is your father" would be "_Qui_ est pater
> tuus?".
>

Still wrong I made a lot of mistakes... what you just wrote is "Who is his
daddy" it's actually "Qui es pater"

> Richard


Richard Bos

unread,
Aug 15, 2001, 12:26:13 PM8/15/01
to
"TheCycoONE" <cyc...@hotmail.com> wrote:

> "Magnus Olsson" <m...@df.lth.se> wrote in message

> news:9le0co$oue$1...@news.lth.se...


> > There is indeed a vocative case in Latin, as in "Et tu, Brute"
> > (nominative "Brutus").
> >
> "Even you Brutus" I know we've done lots of sentences like this (including
> the one mentioned) which I why I was so confused. We only learned
> Nominative, Accusitive and Dative in class and we were told those were the
> only ones,

You're not only missing vocative, but genitive and (rare) ablative, as
well.

> we learned word endings for all these cases, so I would assume
> vocative has the same endings as one of the others and they were never
> diffrentiated.

They do differ. Genitive differes markedly. Vocative differs only for
some declensions. Ablative is, IIRC, identical in form as dative, except
in some rare cases, BICBW.

> > I don't think lower-case letters had been invented that early,
> > but I may be wrong.
>
> Yes they were, Belimicus's will is written in lowercase on papyrus. approx.
> 84 AD

No, they weren't. Anything written that early is probably in uncials,
which look, in part, like lower-case letters, but most look like
capitals and they all function like capitals, anyway. Either that, or in
capitalis (the origin of the word capital!) or in rustica, which looks
like (and basically are) speedily written capitals.
Of course, they also had cursive hands, but that's quite a different
thing from the capital/lower case distinction. I wouldn't be surprised
if a will were written in a cursive, and if someone would use the term
"lower case" for such, but they really have nothing in common with what
we call "lower case", and only the name and purpose with the renaissance
cursives.

Richard

Richard Bos

unread,
Aug 15, 2001, 12:35:21 PM8/15/01
to
"TheCycoONE" <cyc...@hotmail.com> wrote:

> "Richard Bos" <in...@hoekstra-uitgeverij.nl> wrote


> > YM articles, I suppose? Not literally speaking, but that also means that
> > you have to use "ille" (not officially an article) to distinguish
> > between "any old dog" and "that dog you saw before".
>
> No such word, and they didn't difrentiate... your confused with french or
> something.

No such word as what? Ille? There most definitely is such a word. Look
it up in a dictionary if you're not too cock-sure to stand corrected.

And no, I'm not confusing Latin with French. I may be batty enough to
confuse Latin with Italian, but I'm not quite stupid enough to confuse
it with French.

> > Erm, no, they don't. Either everything is CAPS, BECAUSE THE ANCIENT
> > ROMANS DIDN'T HAVE ANY LOWER CASE, or you decide that you're writing
> > Latin _now_, and use caps like everyone else.
>
> WRONG!!!

Says fscking _who_? The same teacher that taught you that "ille" is not
a Latin word?
FYI, most books I've got in Latin (not that many, but still) use normal
capitals the way modern languages do. So much for "WRONG!!!".

> > > quid est pater. (Who's your daddy?)
> >
> > _What_ is father, rather. "Who is your father" would be "_Qui_ est pater
> > tuus?".
>
> Still wrong I made a lot of mistakes... what you just wrote is "Who is his
> daddy"

Eh? Since when does "tu" mean "he", then?

> it's actually "Qui es pater"

No, that's ungrammatical. "Es" is second person. Even "Qui est pater"
would only mean "Who is father" or "Who is the father", not "...your
father".

Richard

Adam Biltcliffe

unread,
Aug 15, 2001, 3:13:06 PM8/15/01
to
Some are born TheCycoONE. Some achieve TheCycoONE. But rec.arts.int-
fiction had TheCycoONE thrust upon it:

> My latin experiance so far has been of the Cambrige Latin Course which is
> used by all the schools in my area.

Ooh, mine too! In fact, that's the very same course I began learning
Latin from. However, from your comments elsewhere in this thread, it's
painfully obvious that you're still on book I, or at best the first
few chapters of book IIa, so I'd suggest that you bear in mind that
they go all the way up to IV before you start screaming 'WRONG!' at
people when they offer corrections to your Latin.


jw

ally.mon

unread,
Aug 15, 2001, 3:46:55 PM8/15/01
to
m...@df.lth.se (Magnus Olsson) wrote in message news:<9ldcd3$kle$1...@news.lth.se>...

As "ins"/"im" are contractions of "in das"/"in dem", those wouldn't be
any less tricky I guess. (Note I said "guess".) I suppose you could simply
define each pair ("ins"/"in das", "im"/"in dem", "in die", "in der") as an
individual preposition, the way TADS handles a player-input "out of" as
"outof", associating it with the internal "out" preposition object.

German IF--haven't played all that much. Starrider, which uses Toni Arnold's
translation of Inform's Lib 6/10, is extremely tolerant with regards to
gender, case and prepositions. Maybe most or all are basically synonyms
of the general-purpose English "the" (big "haven't-analyzed-it" caveat here.)

Thus,

">lege dem handtuchs in das tasche"
(>put the [masc||neut/dat] towel's [neut] in the [neut/nom||acc] bag [fem])

may contain three grammatical errors, but yields the same result as

">lege das handtuch in die tasche"
(which would be correct.)

I don't mind that; I suppose that if I wanted to "jump while in the boat"
(or "walk while in the kitchen") I'd go about that the same way I would in
an English-language game (i.e., ">enter boat. jump").

Is that (tolerance) the way it's done in Swedish IF?

There are a few authoring systems written for German right from the start
though, maybe those handle things differently? Mh.

g'daze,~*~*
ally.mon

ems...@mindspring.com

unread,
Aug 15, 2001, 5:57:33 PM8/15/01
to
Whoo ah. Flameage. Let me see if I can disentangle a few things
gently here.

in...@hoekstra-uitgeverij.nl (Richard Bos) wrote in message news:<3b7aa363....@news.worldonline.nl>...


> "TheCycoONE" <cyc...@hotmail.com> wrote:
>
> > "Richard Bos" <in...@hoekstra-uitgeverij.nl> wrote
> > > YM articles, I suppose? Not literally speaking, but that also means that
> > > you have to use "ille" (not officially an article) to distinguish
> > > between "any old dog" and "that dog you saw before".
> >
> > No such word, and they didn't difrentiate... your confused with french or
> > something.
>
> No such word as what? Ille? There most definitely is such a word. Look
> it up in a dictionary if you're not too cock-sure to stand corrected.
>
> And no, I'm not confusing Latin with French. I may be batty enough to
> confuse Latin with Italian, but I'm not quite stupid enough to confuse
> it with French.

"Ille" certainly exists; it is a demonstrative pronoun meaning "that",
matched with "hic" meaning "this." It is the word that eventually
gave rise to articles in the Romance languages (le/la etc. being
degenerate forms of ille), but was not itself used with anything like
the frequency or freedom of an article.



> > > Erm, no, they don't. Either everything is CAPS, BECAUSE THE ANCIENT
> > > ROMANS DIDN'T HAVE ANY LOWER CASE, or you decide that you're writing
> > > Latin _now_, and use caps like everyone else.
> >
> > WRONG!!!
>
> Says fscking _who_? The same teacher that taught you that "ille" is not
> a Latin word?
> FYI, most books I've got in Latin (not that many, but still) use normal
> capitals the way modern languages do. So much for "WRONG!!!".

As Magnus pointed out, modern editions of classical works tend to rely
(mostly) on modern punctuation and capitalization, for the sake of
clarity. [There are some differences in Greek punctuation, but they
derive from practices developed in the Byzantine period and do not
date back to the Classical Greek to which they are often applied. The
semi-colon, for instance, stands in place of our question mark to
indicate the end of a question in Greek.]

What we now know as lower-case letters are the descendents of
handwriting developed in the middle ages; writing in the ancient world
tended not to be graced with conveniences like a distinction between
upper and lower case, markers at the ends of sentences, or spaces.
Important inscriptions in the Roman world sometimes had a decorative
sign inserted (a small leaf shape was popular) or some spacing added
to separate words when one sentence ended and another began, but that
was not universally applied. They also often relied on abbreviations
and orthographical tricks (blending two or more letters together,
representing combinations of letters by special signs, placing small
letters such as 'I' actually *inside* the next letter over) in order
to use up as little marble as possible.

In short, what you see in your school edition, or even your Loeb,
Teubner, or OCT, is not a visually accurate representation of the form
of Roman lettering on papyrus or on stone, but something reduced for
your convenience to a readable format. They are themselves often
derived from the Medieval manuscript tradition, and are created
through the laborious comparison of the remaining texts. This is why
an academic edition of an ancient text will be accompanied by a thick
rind of footnotes (called the apparatus criticus) which details the
variant forms and words. The majority of ancient texts that we
possess in good shape have come to us by this route. Text taken
directly from ancient media such as stone or papyrus tend to be much
more fragmentary, though occasionally there are pieces large enough to
be useful. (The discovery of Menander's Dyscolus is interesting in
this regard, in that most of the text was found on papyrus in this
century after the play was regarded as lost forever-- but this sort of
thign is very rare.) Even those who have had many years' practice
reading Classical languages need special training in inscriptions and
in papyrology in order to make head or tail of most of them.



> > > > quid est pater. (Who's your daddy?)
> > >
> > > _What_ is father, rather. "Who is your father" would be "_Qui_ est pater
> > > tuus?".
> >
> > Still wrong I made a lot of mistakes... what you just wrote is "Who is his
> > daddy"

> Eh? Since when does "tu" mean "he", then?

> > it's actually "Qui es pater"
>
> No, that's ungrammatical. "Es" is second person. Even "Qui est pater"
> would only mean "Who is father" or "Who is the father", not "...your
> father".

This is true enough. CycoOne seems to have confused the function of
the verb with the function of the possessive pronoun. The verb must
agree with the subject and predicate nominative of this sentence
(Qui... pater) and belongs in the third person.

As for the matter of cases, mentioned earlier in this thread: Latin
has Nominative, Genitive, Dative, Accusative, Ablative, and Vocative.
Ablative is not especially 'rare': it is the most grammatically
ambiguous of the cases, since it is a sort of catch-all case that
inherited from the much larger number of cases in earlier
indo-European languages, and it incorporates many functions of the
locative (place where) and works with prepositions. In these manifold
guises it occurs more commonly than the student would like; it was
certainly the bane of my existence when I was first introduced to it.
The same functionality is, in Greek, distributed among the Genitive,
Dative, and Accusative; no Greek Ablative exists.

For those who are struggling with inadequately informative textbooks,
I recommend the thorough Latin grammar and syntax produced by Allen
and Greenough, which is available for online perusal at the Perseus
website (http://www.perseus.tufts.edu). Wheelock's textbook of Latin
also provides a more thoroughly systematic, though less friendly,
approach than many other standard courses. If you are interested in
prose composition (that is, the production, rather than merely the
comprehension, of Latin), I commend to your attention the exercise and
answer books of North and Hillard; this additional practice is
definitely useful if you have mostly been studying Latin-for-Reading.
Similar resources exist for Greek: Crosby and Schaeffer's textbook is
a guide to Attic Greek with unflinching purism of grammar, directed at
someone already competent in the rigors of Latin, and there is a North
and Hillard Greek composition book as well.

It would be a pity to make so elegant a language the cause of
unwarranted antagonism. It is unquestionably difficult, and
elementary courses in it tend to gloss over its more advanced features
and to present it without historical context; there is no reason to
lambast the partially-informed because they have not yet understood
the esoterica, but it is also unsafe to assume that a first-year
acquaintance with it will answer all these questions.

ES

Graham Nelson

unread,
Aug 15, 2001, 7:19:38 PM8/15/01
to
"ems...@mindspring.com" wrote:
> "Ille" certainly exists; it is a demonstrative pronoun meaning "that",
> matched with "hic" meaning "this." It is the word that eventually
> gave rise to articles in the Romance languages (le/la etc. being
> degenerate forms of ille), but was not itself used with anything like
> the frequency or freedom of an article.

The definitive Latin translation of A. A. Milne's classic was
published under the title "Winnie ille Pooh", as I recall.

One of the chapters began "Pluebat et pluebat et pluebat", but,
eheu fugaces, I can't quote anything else from it.

--
Graham Nelson Oxford, UK

TheCycoONE

unread,
Aug 15, 2001, 8:55:26 PM8/15/01
to

"Adam Biltcliffe" <abilt...@bigfoot.com> wrote in message
news:MPG.15e4da1e4...@news.freeserve.net...
So it seems that there are enough people with knowledge of the language that
I really serve no purpose. No I'm past book I (grade 10), and book II
(grade 11), I've even started book III, which I will complete along with
book IV this year. I suppose that means that my mind just fades out in the
summer becase I remember all about the genitive now and assume the other
cases exist. Blast! Now I feel like a fool... Still I think latin would be
a good language to write in, other people do.


TheCycoONE

unread,
Aug 15, 2001, 8:58:13 PM8/15/01
to

"Richard Bos" <in...@hoekstra-uitgeverij.nl> wrote in message
news:3b7aa363....@news.worldonline.nl...

> "TheCycoONE" <cyc...@hotmail.com> wrote:
>
> > "Richard Bos" <in...@hoekstra-uitgeverij.nl> wrote
> > > YM articles, I suppose? Not literally speaking, but that also means
that
> > > you have to use "ille" (not officially an article) to distinguish
> > > between "any old dog" and "that dog you saw before".
> >
> > No such word, and they didn't difrentiate... your confused with french
or
> > something.
>
> No such word as what? Ille? There most definitely is such a word. Look
> it up in a dictionary if you're not too cock-sure to stand corrected.
>
> And no, I'm not confusing Latin with French. I may be batty enough to
> confuse Latin with Italian, but I'm not quite stupid enough to confuse
> it with French.
>

Blast I've been confused... I'll quit being arrogant (in my defence those
posts were all written in a very short time well I was quite tired)
Anyway... please erase these posts from the server immediatly and be done
with it.

L. Ross Raszewski

unread,
Aug 16, 2001, 12:29:38 AM8/16/01
to
On Thu, 16 Aug 2001 00:19:38 +0100, Graham Nelson
<gra...@gnelson.demon.co.uk> wrote:
>"ems...@mindspring.com" wrote:
>> "Ille" certainly exists; it is a demonstrative pronoun meaning "that",
>> matched with "hic" meaning "this." It is the word that eventually
>> gave rise to articles in the Romance languages (le/la etc. being
>> degenerate forms of ille), but was not itself used with anything like
>> the frequency or freedom of an article.
>
>The definitive Latin translation of A. A. Milne's classic was
>published under the title "Winnie ille Pooh", as I recall.

But... everyone knows his name is really Winnie *ther* Pooh. You know
what "ther" means, don't you :-)

John Colagioia

unread,
Aug 16, 2001, 10:14:00 AM8/16/01
to
Karl Filenius wrote:

> cons:
> I can't think of any.

Here are a handful, just off the top of my head. Most are based on some
idealized form of IF parsing, and may or may not reflect the state of the art in
any situation.

- English, inheriting the vast majority of its vocabulary from different
sources, has little regular word formation, meaning that every word must be
coded into the program in order to be recognized. Contrast with, for example,
French, where you can "deglutenize" a word to an unidentifiable root and
distinguish between "this word isn't implemented" and "I have no idea what
you're talking about."

- English, being extremely widespread, has an extraordinary number of dialects.
These range from trivial spelling differences (British/American) to oddities in
word order (I don't remember where, but I vaguely remember a pocket of Americans
that traditionally start sentences with the verb) to almost entirely different
vocabularies (near any Creole- or Pidgin-speaking area, typically).

- English, having no grammatical gender other than personal pronouns, the
pronoun base is significantly smaller (one, in most cases), meaning that things
have to be referenced explicitly more often, or risk absurd ambiguity. In fact,
there was a debate, here, if I remember my research correctly, on whether or not
"put it on the tray" should reset "it" to "the tray" in the Inform library.

- Voice recognition might be more complex, as vowels are only sometimes
important, but when they are, they're critical to understanding. Most other
languages that I've seen are a bit more consistent in their vowel usage.

I'm sure there are others floating around.


John Colagioia

unread,
Aug 16, 2001, 10:18:09 AM8/16/01
to
Magnus Olsson wrote:

> In article <W4ge7.14145$e5.20...@newsb.telia.net>,
> Marek Teraszkiewicz <eraszk...@usanet.com> wrote:
> >> >5) There is no grammatical gender in English. You don't have to worry
> about
> >> >making sure that the adjective has the same gender as the noun it
> describes.
> >> There are some gender issues, for instance when you refer to objects
> >> or ships.
> >I can see what you're driving at but the fact that sailors and weapon
> >fetischists refer to their ships and guns using personal feminine pronouns
> >doesn't constitute grammatical gender.
> There is still grammatical gender for personal pronouns.

I'm not sure that's gender in the same sense, though. It only pertains to
which people are allowed to be referenced by that pronoun, and affects nothing
else about the sentence. "Gendering," as I understand it, pertains not only to
the noun name itself, but also to the agreement of verb, article, adjective,
and possibly other parts of speech I'm forgetting.

[...]


Magnus Olsson

unread,
Aug 16, 2001, 2:33:14 PM8/16/01
to
In article <3B7BD620...@csi.com>,

Yes; that's what called congruence, and having to bother about that
would add complexity to an IF game.

What I meant was that English isn't quite ideal from the gender point
of view, since you still have to give nouns genders if you want
pronouns to work properly. It would be simpler if there were just one
personal pronoun, like in Finnish.

Magnus Olsson

unread,
Aug 16, 2001, 2:39:13 PM8/16/01
to
In article <3B7B0389...@gnelson.demon.co.uk>,

Graham Nelson <gra...@gnelson.demon.co.uk> wrote:
>"ems...@mindspring.com" wrote:
>> "Ille" certainly exists; it is a demonstrative pronoun meaning "that",
>> matched with "hic" meaning "this." It is the word that eventually
>> gave rise to articles in the Romance languages (le/la etc. being
>> degenerate forms of ille), but was not itself used with anything like
>> the frequency or freedom of an article.
>
>The definitive Latin translation of A. A. Milne's classic was
>published under the title "Winnie ille Pooh", as I recall.

"Winnie ille Pu," accoring to a Web search.

It's interesting that the evolution of demonstrative pronouns into
definite articles seems to have occured three times separately
among Indoeuropean languages - both Greek, Germanic and Romance
languages seem to have gone through a similar evolution, but
at widely separated times.

Adam Thornton

unread,
Aug 16, 2001, 8:02:17 PM8/16/01
to
In article <3b7a9ed6....@news.worldonline.nl>,

Richard Bos <in...@hoekstra-uitgeverij.nl> wrote:
>if a will were written in a cursive, and if someone would use the term
>"lower case" for such, but they really have nothing in common with what
>we call "lower case", and only the name and purpose with the renaissance
>cursives.

To further clarify: the phrase "lower case", of course, only makes sense
after the invention of moveable type, and development of a convention
for storing slugs.

Adam

David Given

unread,
Aug 17, 2001, 6:39:23 AM8/17/01
to
In article <9lh40h$j7f$2...@news.lth.se>,
m...@df.lth.se (Magnus Olsson) writes:
[...]

>>The definitive Latin translation of A. A. Milne's classic was
>>published under the title "Winnie ille Pooh", as I recall.
>
> "Winnie ille Pu," accoring to a Web search.

And don't miss the sequel, _Domus Angularum Puensis_ (approximately).

--
+- David Given --------McQ-+ "[One shot of the Death Star's superlaser has
| Work: d...@tao-group.com | enough power] to send Marty McFly to the Big Bang
| Play: d...@cowlark.com | and back 1.4x10^29 times." --- William Clifford
+- http://www.cowlark.com -+

DModus

unread,
Aug 17, 2001, 2:49:57 PM8/17/01
to


"John Colagioia" <JCola...@csi.com> wrote in message
news:3B7BD528...@csi.com...


> Karl Filenius wrote:
>
> > cons:
> > I can't think of any.
>
> Here are a handful, just off the top of my head. Most are based on some
> idealized form of IF parsing, and may or may not reflect the state of the
art in
> any situation.
>
> - English, inheriting the vast majority of its vocabulary from different
> sources, has little regular word formation, meaning that every word must
be
> coded into the program in order to be recognized. Contrast with, for
example,
> French, where you can "deglutenize" a word to an unidentifiable root and
> distinguish between "this word isn't implemented" and "I have no idea what
> you're talking about."
>
> - English, being extremely widespread, has an

Unfortunately as I am unable to speak or even read english all games are a
problem for me. I mean what the heck is a 'grue'? It just boggles the mind
when trying to understand some of the more 'colourful' versions of the
english language, certainly when the person who wrote the game wanted you to
say 'devour the food' when 'eat' will do just as well.
Simplify, simplify, simplify.

--
Darren Sparrow

Brought to you by Wunderspeek, the most wonderful translation tool on the
planet.
Disclaimer: this device when worn in the neck may or may not induce madness,
vomiting and a penchant for Bakewell tarts.


Jonadab the Unsightly One

unread,
Aug 18, 2001, 12:03:47 AM8/18/01
to
"Karl Filenius" <karl_f...@hotmail.com> wrote:

> pros:

[snip]

> cons:
>
> I can't think of any.

1. The vocabulary is positively expansive. Most of us
use words on a daily basis that aren't found in a
typical sixty-thousand-entry dictionary. Then there
are a great many words that we use only occasionally.
The OED is comprised of twenty volumes, and it is by
no means complete. English has words the way a beach
has grains of sand.

2. Verb phrases. The oddities involved with the way
we jostle combinations of auxilliary words to create
different combinations of time, tense-aspect, voice,
and mood are considerable.

3. Someone else already mentioned plural nouns; we have
some dozen different declensions, more than one of
which vary greatly in application from word to word.

#1 is probably the most significant. This is where
guess-the-verb comes from, among other things.

- jonadab

John Colagioia

unread,
Aug 18, 2001, 7:18:00 AM8/18/01
to
DModus wrote in message ...


Given your apparently random editing, I'm not entirely sure, here, but I
suspect you've misinterpreted me--possibly intentionally. My point was that
English requires a vast number of choices for the author to make when
designing his acceptable vocabulary. Does the game contain "a coloring
book"? Is "a colouring book" an acceptable synonym? Is that going to be
done for every "-or" word in the game? By hand?

Ditto for synonyms. The OED is huge and largely incomplete by their own
admission. Plus, words change in meaning at different rates in different
regions.

To put it in simple terms that every IF programmer, here, will understand,
the only way an IF author can avoid guess-the-verb problems is by playing
guess-the-verb, himself, every step of the way. And, that's probably the
least-entertaining aspect of coding a game, when you get right down to it.

The real reason English is "superior" for IF is the fact that Inform, TADS,
Hugo, and most of the other dozens of available construction systems already
support English parsing, so most of the problematic work has already been
done.

Kevin Forchione

unread,
Aug 18, 2001, 1:36:26 PM8/18/01
to
"John Colagioia" <JCola...@csi.com> wrote in message
news:3b7e4...@excalibur.gbmtech.net...

This is, of course, exactly the same dilemma faced by an author of fiction.
Diction is one of the elements that one takes into consideration, at times
painstakingly so. "By hand"? As opposed to the convenience of symbolic
substitution provided by a text editor? Heh!

<snip>


> To put it in simple terms that every IF programmer, here, will understand,
> the only way an IF author can avoid guess-the-verb problems is by playing
> guess-the-verb, himself, every step of the way. And, that's probably the
> least-entertaining aspect of coding a game, when you get right down to it.
>
> The real reason English is "superior" for IF is the fact that Inform,
TADS,
> Hugo, and most of the other dozens of available construction systems
already
> support English parsing, so most of the problematic work has already been
> done.

Now why do you think that is? Why was English chosen as the standard, when
surely there have been competent computer programmers whose native language
is not English. The real reason English is "superior" as a medium for any
literature is that it provides a vehicle that allows an author to express
complex, subtle, varied, and multi-faceted ideas. One can argue otherwise,
but the preponderance of material evidence speaks for itself.

--Kevin


Jonadab the Unsightly One

unread,
Aug 18, 2001, 5:58:11 PM8/18/01
to
m...@df.lth.se (Magnus Olsson) wrote:

> It's interesting that the evolution of demonstrative pronouns into
> definite articles seems to have occured three times separately
> among Indoeuropean languages - both Greek, Germanic and Romance
> languages seem to have gone through a similar evolution, but
> at widely separated times.

The Greek article derives from the _relative_ pronoun, unless
I have misunderstood completely.

- jonadab

Paul Trembath

unread,
Aug 18, 2001, 7:11:55 PM8/18/01
to

"Kevin Forchione" <Ke...@lysseus.com> wrote in message
news:uIxf7.12556$P15.6...@news1.rdc1.sfba.home.com...

> "John Colagioia" <JCola...@csi.com> wrote in message
> news:3b7e4...@excalibur.gbmtech.net...
> > DModus wrote in message ...
<...or perhaps it was someone else...>

> > English requires a vast number of choices for the author to make when
> > designing his acceptable vocabulary. Does the game contain "a coloring
> > book"? Is "a colouring book" an acceptable synonym? Is that going to
be
> > done for every "-or" word in the game? By hand?
>
> This is, of course, exactly the same dilemma faced by an author of
fiction.
> Diction is one of the elements that one takes into consideration, at times
> painstakingly so. "By hand"? As opposed to the convenience of symbolic
> substitution provided by a text editor? Heh!

There's a significant difference between game output and player input, of
course.

"Kevin Forchione" <Ke...@lysseus.com> wrote in message
news:uIxf7.12556$P15.6...@news1.rdc1.sfba.home.com...


>
> Now why do you think that is? Why was English chosen as the standard, when
> surely there have been competent computer programmers whose native
language
> is not English. The real reason English is "superior" as a medium for any
> literature is that it provides a vehicle that allows an author to express
> complex, subtle, varied, and multi-faceted ideas. One can argue otherwise,
> but the preponderance of material evidence speaks for itself.

English is certainly a fine and large language, though this is not always
thought to be a good quality in Cobol or PL/I. However, I doubt that
English is *uniquely* subtle and expressive; I guess all natural languages
are broadly equivalent in expressive power, though mechanical issues such as
alphabet, syntax and inflection may make IF easier or harder.

Most IF uses English because English is widely used in the modern world, and
most people here (anyone who is not here, please speak up) can use it well
enough for the purpose. IF in Russian, Esperanto, or Klingon has its
attractions, but will remain inaccessible to most of us and therefore less
attractive to an author seeking a wider audience. If Charles Babbage had
been born a thousand years earlier, and less of a perfectionist, we could be
enjoying an ancient tradition of Latin IF, predating the modern novel.
Perhaps.

--
pt

Karl Filenius

unread,
Aug 18, 2001, 6:46:31 PM8/18/01
to
"Kevin Forchione" <Ke...@lysseus.com> skrev i meddelandet
news:uIxf7.12556$P15.6...@news1.rdc1.sfba.home.com...

[...]

> Now why do you think that is? Why was English chosen as the standard, when
> surely there have been competent computer programmers whose native
language
> is not English. The real reason English is "superior" as a medium for any

> literature [sic] is that it provides a vehicle that allows an author to


express
> complex, subtle, varied, and multi-faceted ideas. One can argue otherwise,
> but the preponderance of material evidence speaks for itself.

Are you saying that other languages are unable to "express complex, subtle,
varied, and multi-faceted ideas"?

TheCycoONE

unread,
Aug 18, 2001, 10:18:35 PM8/18/01
to

"Paul Trembath" <ptre...@compuserve.com> wrote in message
news:9lmsup$idl$1...@suaar1ab.prod.compuserve.com...

But as demographic's go English quite distant from the most spoken language
in the world (chinese), and we can't say that orintal people have not
created the technology. It seems that we have all decided that computers
and such technology would be more fit for english. I feel the real reason
for this is the languages amazing ability to change. If we need a word it
gets invented, if a word becomes impractical we change it or quit saying it.
If it takes to long to spell we shorten it etc. Our short alphabet was also
very much key, but many langauges have short alphabets. It would be
impossible to make French the language of rapid technology because it would
be impossible for the Academy Frances(sp) to keep up with the lingo, and the
language can't change without there concent. German is almost a nice
language, but it's less spoken then english do to the glory days of England.

Enough rambling, good night.

TheCycoONE


John Colagioia

unread,
Aug 19, 2001, 9:42:47 AM8/19/01
to
Kevin Forchione wrote in message ...

>"John Colagioia" <JCola...@csi.com> wrote
>> DModus wrote in message ...
>> My point was that
>> English requires a vast number of choices for the author to make when
>> designing his acceptable vocabulary. Does the game contain "a coloring
>> book"? Is "a colouring book" an acceptable synonym? Is that going to be
>> done for every "-or" word in the game? By hand?
>This is, of course, exactly the same dilemma faced by an author of fiction.
>Diction is one of the elements that one takes into consideration, at times
>painstakingly so. "By hand"? As opposed to the convenience of symbolic
>substitution provided by a text editor? Heh!


Uhm...no. As opposed to algorithmically checking anything ending in -our to
see if an otherwise identical word ending in -or is in the dictionary. And
possibly changing the dictionary and output strings if this is the case, so
that the game "dynamically" changes from British to American English, for
example. [Aside: I was once going to use something like this as a gimmick
in a game, so I actually did some work in this area. It was not amusing...]

Authors of static fiction rarely, if ever, need to consider the active
vocabulary of the reader. We all know many, many more words than we use.
The IF author doesn't have that luxury and absolutely must compare and
contrast the active and passive vocabularies to make sure the game and the
player can effectively communicate.

><snip>


>> The real reason English is "superior" for IF is the fact that Inform,
TADS,
>> Hugo, and most of the other dozens of available construction systems
already
>> support English parsing, so most of the problematic work has already been
>> done.
>Now why do you think that is? Why was English chosen as the standard, when
>surely there have been competent computer programmers whose native language
>is not English.

I had honestly assumed that this was because England, the United States, and
Germany were the first nations to have electronic, stored-program computers
to any significant degree, and thus laid the groundwork that everyone else
has built on. German was presumably eliminated from "the running" because
of their international status after WWII, leaving English as the effective
lingua franca of the computing community, since if you want to read Turing's
papers, you need to understand English anyway.

>The real reason English is "superior" as a medium for any
>literature is that it provides a vehicle that allows an author to express
>complex, subtle, varied, and multi-faceted ideas.

It obviously can, but I fail to see how this distinguishes English
from...well, any other natural language.

>One can argue otherwise,
>but the preponderance of material evidence speaks for itself.


What material evidence might that be, though? The amount of English text
found in an American (or British) library or bookstore? I don't know of any
other language as historically significant (the British spent a heck of a
lot of time distributing their language around the planet and edging out
their nearest competition) in a region of active scientific and geographic
exploration. So, I don't really have a point of comparison.

L. Ross Raszewski

unread,
Aug 19, 2001, 7:22:44 PM8/19/01
to
On Sun, 19 Aug 2001 09:42:47 -0400, John Colagioia <JCola...@csi.com> wrote:
>><snip>
>>> The real reason English is "superior" for IF is the fact that Inform,
>TADS,
>>> Hugo, and most of the other dozens of available construction systems
>already
>>> support English parsing, so most of the problematic work has already been
>>> done.
>>Now why do you think that is? Why was English chosen as the standard, when
>>surely there have been competent computer programmers whose native language
>>is not English.
>
>I had honestly assumed that this was because England, the United States, and
>Germany were the first nations to have electronic, stored-program computers
>to any significant degree, and thus laid the groundwork that everyone else
>has built on. German was presumably eliminated from "the running" because
>of their international status after WWII, leaving English as the effective
>lingua franca of the computing community, since if you want to read Turing's
>papers, you need to understand English anyway.

In the particular case of China, which has been mentioned elsewhere in
the thread, as there are more native speakers of Chinese (whichever
dialect is most used, I reckon) than any other language, I seem to
recall that the orientation of their computers away from their own
language had something to do with the mechanical difficulties in
making an input device that supported their language without being
either (a) impractically unweildy or (b) restricted to a subset of the
language.

I also recall that a number of Far-east companies (in those countries
having written languages related to Chinese) are putting in grat
amounts of work toward resolving natural language parsing and speech
recognition, since they've pretty much decided that textual input is
just plain too much work to do usefully for their languages.

Richard Bos

unread,
Aug 20, 2001, 6:02:10 AM8/20/01
to
lrasz...@loyola.edu (L. Ross Raszewski) wrote:

> In the particular case of China, which has been mentioned elsewhere in
> the thread, as there are more native speakers of Chinese (whichever
> dialect is most used, I reckon) than any other language, I seem to
> recall that the orientation of their computers away from their own
> language had something to do with the mechanical difficulties in
> making an input device that supported their language without being
> either (a) impractically unweildy or (b) restricted to a subset of the
> language.

Not the language so much as the writing. Several thousands of
complicated ideographs and half-ideographs is not the nicest thing to
use on either a keyboard or a primitive printer. Kana are a lot easier
already - and had the Chinese used something like Korean letters
instead, they'd have had a lot less problems with computers. Something
like Pinyin with accents would do very nicely, and I wouldn't be
surprised if there are already computers using this, or if China in its
entirety would be going to use it a lot more.

Richard

Will Briggs

unread,
Aug 20, 2001, 11:51:47 AM8/20/01
to
Interesting thread. I think the "con" of "too many synonyms" (or
near-synonyms) is well taken. I'd say overall, though, that though
English may be better for IF than Japanese, Spanish, or Cherokee, it's
still tough enough to make IF difficult.

I can speak to "Is English (writing) the ultimate writing system for
computers?" I would say YES, along with Russian and I'm sure others I
don't know. Why:

1. Diacritics exist, but can be dropped.
2. Only 2 alphabets (upper/lower case).
3. Fewer than 256 chars. This means 1-byte encoding, and a keyboard
of reasonable size.
4. Letters don't change shape based on position (Greek, Arabic) or
break into parts (Cambodian).
5. It's always one direction, left to right. (Hebrew and Arabic are
R->L except numbers are L->R; Japanese and Chinese can be R->L down
the page or top->bottom going left across page, except that borrowed
Western words still must be L->R)
6. No pronunciation notations (Japanese).

But Spanish is going to be a lot better for speech synthesis!

David Brain

unread,
Aug 23, 2001, 10:12:00 AM8/23/01
to
In article <3b7de5e7...@news.bright.net>, jon...@bright.net (Jonadab
the Unsightly One) wrote:

> 1. The vocabulary is positively expansive. Most of us
> use words on a daily basis that aren't found in a
> typical sixty-thousand-entry dictionary. Then there
> are a great many words that we use only occasionally.
> The OED is comprised of twenty volumes, and it is by
> no means complete. English has words the way a beach
> has grains of sand.

And yet somehow we still find god-knows-how-many different ways to use the
word "set" (and similar words.)

--
David Brain
London, UK

Xiphias Gladius

unread,
Aug 23, 2001, 11:32:24 PM8/23/01
to
Kevin Forchione <Ke...@lysseus.com> wrote:

>> The real reason English is "superior" for IF is the fact that Inform,
> TADS,
>> Hugo, and most of the other dozens of available construction systems
> already
>> support English parsing, so most of the problematic work has already been
>> done.

> Now why do you think that is? Why was English chosen as the standard, when
> surely there have been competent computer programmers whose native language
> is not English. The real reason English is "superior" as a medium for any
> literature is that it provides a vehicle that allows an author to express
> complex, subtle, varied, and multi-faceted ideas. One can argue otherwise,
> but the preponderance of material evidence speaks for itself.

The real reason that English was chosen as the standard was that Infocomm
was headquartered in Kendall Square, in Cambridge, Massachusetts.

And that the people who started Infocomm had previously been at the
Massachusetts Institute of Technology, also in Cambridge, Massachusetts.

While many languages are spoken in Cambridge the lingua franca (which is
Latin for "French language", unless it isn't) is English. Few people at
MIT cannot speak at least some English, while many people at MIT are
hopeless in Spanish, Russian, Mandarin, Cantonese, German, *and* Japanese.

- Ian
--
Marriage, n: The state or condition of a community consisting of a master,
a mistress, and two slaves, making, in all, two. -- Ambrose Bierce

John W. Kennedy

unread,
Aug 24, 2001, 11:40:48 AM8/24/01
to
TheCycoONE wrote:
> But as demographic's go English quite distant from the most spoken language
> in the world (chinese),

Actually, English is spoken by more people than Chinese, and is a far
more important language, as nearly all speakers of Chinese live in
either China or Chinese ghettos.

=--
John W. Kennedy
(Working from my laptop)


John W. Kennedy

unread,
Aug 24, 2001, 11:40:30 AM8/24/01
to
Assuming a new game engine, the ultimate language for IF is Loglan
(http://www.loglan.org), which actually has a complete YACC parser
already in existence. (It is even 100% possible to completely parse
sentences with unknown words.) It uses the Latin alphabet with no
diacritics except, optionally, in proper nouns.

For Inform use, Loglan has a few problems. It may not map well to
Informese, because verbs, nouns, adjectives, and adverbs are all
subsumed into a single class called "predicates". ee.g., "grada" maps
to English as: "is great", "great thing", "great", or "greatly",
depending on the syntactic context; similarly, "hasfa" may be: "is a
house", "house", "of a house", or "in a houselike manner"; and "prano":
"run", "runner", "running", or "in a running manner". Another problem
is that, in proper names only, a comma with no space after it is used to
indicate that a vowel pair that would ordinarily be pronounced as a
single syllable is to be pronounced as two: e.g., "Lois" is spelled
"Lo,is" to indicate that it is not to rhyme with "rejoice". (I have
myself already proposed that "Loļs" be declared an alternative where
available.) And Loglan uses, as third-person pronouns, letters of the
alphabet; in other words, if English worked the same way, instead of
saying: "Look at the girl. Kiss her," we'd say, "Look at the girl.
Kiss g." Letters of the Greek alphabet are available for second words,
so that we could say: "Look at the goat. Look at the girl. Kiss
gamma."

--
John W. Kennedy
(Have to get back to work on my Java port of the Loglan-English
dictionary, and "Le Nu reible Gramadji pe la Az")


John W. Kennedy

unread,
Aug 24, 2001, 11:40:50 AM8/24/01
to
Jonadab the Unsightly One wrote:
> 1. The vocabulary is positively expansive. Most of us
> use words on a daily basis that aren't found in a
> typical sixty-thousand-entry dictionary. Then there
> are a great many words that we use only occasionally.
> The OED is comprised of twenty volumes, and it is by
> no means complete.

On the other hand, it's been completed (twice, in fact!), which is more
than you can say for most such.

John W. Kennedy

unread,
Aug 24, 2001, 11:40:39 AM8/24/01
to
"ems...@mindspring.com" wrote:
> As for the matter of cases, mentioned earlier in this thread: Latin
> has Nominative, Genitive, Dative, Accusative, Ablative, and Vocative.

And Locative, though that one really _is_ rare, belonging only to names
of cities.

John W. Kennedy

unread,
Aug 24, 2001, 11:40:46 AM8/24/01
to
Paul Trembath wrote:
> English is certainly a fine and large language, though this is not always
> thought to be a good quality in Cobol or PL/I.

PL/I is a large language. COBOL is a remarkably tiny, but garrulous
language.

> However, I doubt that
> English is *uniquely* subtle and expressive;

On the contrary, _all_ languages are uniquely subtle and expressive.

English probably wins "largest morpheme pool", but spelling is both
actively and passively horrid (French spelling is only actively horrid),
and early-21st-century English grammar suffers from being half synthetic
and half analytic, and always being in danger of tumbling into the abyss
between.

John W. Kennedy

unread,
Aug 24, 2001, 11:40:43 AM8/24/01
to
Kevin Forchione wrote:
> Now why do you think that is? Why was English chosen as the standard, when
> surely there have been competent computer programmers whose native language
> is not English.

Yes, but during the Golden Age of IF, the overwhelming majority of
people who could afford to spend large amounts of computer time for
recreational purposes were native speakers of English, and I suspect
that even today a majority of the professional programmers in the world
are at least semi-competent in English.

> The real reason English is "superior" as a medium for any
> literature is that it provides a vehicle that allows an author to express
> complex, subtle, varied, and multi-faceted ideas. One can argue otherwise,
> but the preponderance of material evidence speaks for itself.

Modern day English grammar is full of structural ambiguities that make
complexity difficult, while often lacking ambiguities convenient for
subtlety.

English probably has the largest morpheme set of all known languages,
and the licence to verb a noun or noun a say is not without usefulness,
but those are not the only tools available in the languages of the
world. Instance: as a rule, English stabs at haiku verse always fail
to scan. Instance: I find English attempts at quantity lacking.

Of course, nothing in the above is intended to deny the self-evident
fact that English is, at present, Top Language.

John W. Kennedy

unread,
Aug 24, 2001, 11:40:35 AM8/24/01
to
Magnus Olsson wrote:
> I don't think lower-case letters had been invented that early,
> but I may be wrong.

The lower-case letter forms are from the time of Charlemagne, but they
were conceptualized simply as a new hand for the existing monocase
alphabet. The two-case convention evolved slowly, not really arriving
until after printing, I think, and the modern rules of two-case usage in
English are only about 200 years old.

Larry Smith

unread,
Aug 24, 2001, 1:16:48 PM8/24/01
to
"John W. Kennedy" wrote:

> English probably wins "largest morpheme pool", but spelling is both
> actively and passively horrid (French spelling is only actively horrid),

I guess one shouldn't point out that a lot of English's
actively horrid spelling comes directly from the French
words in has assimilated...

--
.-. .-. .---. .---. .-..-. | Wild Open Source Inc.
| |__ / | \| |-< | |-< > / | "Making the bazaar just a
`----'`-^-'`-'`-'`-'`-' `-' | little more commonplace."
http://www.smith-house.org/ | Need programming? Ask me.

Terry Turner

unread,
Aug 24, 2001, 2:06:49 PM8/24/01
to
Hi!

>>> The real reason English is "superior" for IF is the fact that Inform,
>> TADS,
>>> Hugo, and most of the other dozens of available construction systems
>> already
>>> support English parsing, so most of the problematic work has already been
>>> done.
>
>> Now why do you think that is? Why was English chosen as the standard, when
>> surely there have been competent computer programmers whose native language
>> is not English. The real reason English is "superior" as a medium for any
>> literature is that it provides a vehicle that allows an author to express
>> complex, subtle, varied, and multi-faceted ideas. One can argue otherwise,
>> but the preponderance of material evidence speaks for itself.
>
>The real reason that English was chosen as the standard was that Infocomm
>was headquartered in Kendall Square, in Cambridge, Massachusetts.
>
>And that the people who started Infocomm had previously been at the
>Massachusetts Institute of Technology, also in Cambridge, Massachusetts.
>
>While many languages are spoken in Cambridge the lingua franca (which is
>Latin for "French language", unless it isn't) is English. Few people at
>MIT cannot speak at least some English, while many people at MIT are
>hopeless in Spanish, Russian, Mandarin, Cantonese, German, *and* Japanese.

As for Russian - it's much more difficult for programming, than English.
The reasons are:
1. There are at least four different codepages of Russian.
2. There are lots of wordforms and synonims - the only way to say "touch the
cat" in English is to say "touch the cat" and in Russian there dozen of
possible correct phrases.

Terry.


Alexander Deubelbeiss

unread,
Aug 24, 2001, 4:59:11 PM8/24/01
to

John W. Kennedy schrieb in Nachricht <3B865B27...@bellatlantic.net>...
Well, there _is_ a difference between "complete" and "completed".

Nikita

unread,
Aug 24, 2001, 9:32:40 PM8/24/01
to
brig...@acavax.lynchburg.edu (Will Briggs) claims to have found this in a
bottle:

> But Spanish is going to be a lot better for speech synthesis!

Along with Russian, and probably tons of other languages...

English is really unsuited for speech :)

--
P.S. please if you get a chanse put some flowrs
on Algernons grave in the bak yard.

Nikita

unread,
Aug 24, 2001, 9:35:52 PM8/24/01
to
Terry Turner <terry_...@mail.ru> claims to have found this in a
bottle:

> [..]


> the only way to say
> "touch the cat" in English is to say "touch the cat" and in Russian
> there dozen of possible correct phrases.

Umm.... Ever played Guess the Verb? :)

Jake Wildstrom

unread,
Aug 25, 2001, 9:24:49 AM8/25/01
to
In article <0108250006...@localhost.localdomain>,

Terry Turner <terry_...@mail.ru> wrote:
>As for Russian - it's much more difficult for programming, than English.
>The reasons are:
>1. There are at least four different codepages of Russian.
>2. There are lots of wordforms and synonims - the only way to say "touch the
>cat" in English is to say "touch the cat" and in Russian there dozen of
>possible correct phrases.

To put it succinctly and technically: unlike most languages, English
possesses very rigid syntax (word order) and practically no
grammar. If you're writing a parser which tokenizes word-by-word,
you're mostly happier with lots of syntax and not much grammar, since
it means you can determine a word's function simply by its location in
the sentence without deep analysis of inflection.

+--First Church of Briantology--Order of the Holy Quaternion--+
| A mathematician is a device for turning coffee into |
| theorems. -Paul Erdos |
+-------------------------------------------------------------+
| Jake Wildstrom |
+-------------------------------------------------------------+

John W. Kennedy

unread,
Aug 25, 2001, 9:52:45 AM8/25/01
to
Xiphias Gladius wrote:
> lingua franca (which is
> Latin for "French language",

Lingua Franca for "Frankish language", actually, "Frank" being a generic
term for any western European in the East. (Lingua Franca is a
Mediterranean trade language with simplified Romance, Arabic, and
Turkish elements, now dead, I believe. The easiest way to see a bit of
it is in the induction scene of Moliere's "Le Bourgeois Gentilhomme".