Oska - Wtf

A forum for all topics related to constructed languages
Post Reply
User avatar
Micamo
MVP
MVP
Posts: 5671
Joined: 05 Sep 2010 19:48
Contact:

Oska - Wtf

Post by Micamo »

Earlier this morning I had a freaky idea: How about a conlang with built-in syntax for defining words and concepts in a very precise way?

After playing around with it all day (most of the time spent on making a translation) I've got Oska.
num tem suk num o seno met fko em seno to em ot fko em y suk en to fko se suk em es en ot mut em y suk seno to seno ot mut em y suk en to fko se suk em es en ot pex em y seno to suk seno ot pex em y suk en to mut se pex em en es em ot pam pa po tem pam pa po met pun tem te o fe xist ka tem kon ka xist ka o nim met fot ef it y nim to it ot fot ef it y kon ka ko to fot ef se ef ka it es ko ot kont y te em en to em ot kont y fe em en to en ot ketik ek y pam em en to kont ek pam fok em pex se suk suk seno es suk en pam em suk en ot kom em en ex to em se en ex es ot fip ef em en to ef en em ot pokta to fok se fip kon es nim ot fast y pam em en to em ot pinto to fast kom se fot ketik pam seno seno es pokta.
Rules:

to and ot denote definitions of verbs. To denote verb definitions, first state the name of the verb, followed by any number of nicknames for the verb's objects. (Intransitive verbs take one object, transitive take 2, ditransitive 3, and so on). After stating the names, wrap the definition in to-ot. (For example, swim subject to subject moves through water ot).

Similarly tem and met denote noun definitions. The particle 'o' breaks up the constructor definitions. To use an example from my translation, "num tem suk num o seno met" means "A num is either the suk of another num or it's a seno." Think of the constructors like adjectives.

y is a particle that allows you to refer to a defined noun's internal parts when defining verbs rather than referring to the whole thing. For example, "mon y suk em" means "em" is referring to whatever our verb's object is a "suk" of.

Now, as for the translation itself. You're probably wondering what it says. Actually I just gave you everything you need to know. Figure it out like a real man.
Spoiler:
Give up? It's peano arithmetic! Once the definition above is established between all speakers they can simply say something to the effect of "pinto kon oks kon ikt kon petf kon ponk nim" whenever they need to refer to a number. (In that case the number is 1,924.) Really I think expecting people to actually be able to parse this kinda stuff in daily conversation is completely insane but the option sure would be useful to have at times.
My pronouns are <xe> [ziː] / <xym> [zɪm] / <xys> [zɪz]

My shitty twitter
User avatar
Pirka
roman
roman
Posts: 907
Joined: 09 Oct 2010 10:18
Location: Seattle

Re: Oska - Wtf

Post by Pirka »

*brain broke*

Just give me a couple of seconds to gather the pieces...

I have a feeling that this is an interesting idea, but I just don't understand it yet. Something about your particle explanation don't click. Could you give some better examples with a gloss?
User avatar
Micamo
MVP
MVP
Posts: 5671
Joined: 05 Sep 2010 19:48
Contact:

Re: Oska - Wtf

Post by Micamo »

Pirka wrote:*brain broke*

Just give me a couple of seconds to gather the pieces...

I have a feeling that this is an interesting idea, but I just don't understand it yet. Something about your particle explanation don't click. Could you give some better examples with a gloss?
A gloss wouldn't really work: Other than the "axiomatic" particles I defined above every last word in there has a definition described somewhere before. Here's a much simpler example:
tefe tem te o fe met ite te em en to em ot ite fe em en to en ot
This defines if-then-else. Using is as simple as "ite (Predicate) (Positive Result) (Negative Result)." (A caveat: the predicate needs to be defined such that it evaluates to a yes-or-no question. Here you'd place something like "Have you done your homework?" and use it with ite to say "If you do your homework you'll get good grades, or else you will die.")

The defined notions of true and false are encapsulate in "tefe tem te o fe met." Note the definition of te fe doesn't require te or fe to be true or false, what they mean is really dependent on whatever context they're used in: We could have easily switched te and fe in ite's definition to make "fe" mean yes and "te" mean no.
My pronouns are <xe> [ziː] / <xym> [zɪm] / <xys> [zɪz]

My shitty twitter
User avatar
Czwartek
sinic
sinic
Posts: 344
Joined: 12 Aug 2010 15:50

Re: Oska - Wtf

Post by Czwartek »

I'm really not getting this either. Maybe tomorrow when I'm less tired ...
User avatar
jseamus
greek
greek
Posts: 614
Joined: 15 Aug 2010 23:07
Location: Phoenix, AZ

Re: Oska - Wtf

Post by jseamus »

It looks like it should make sense, and I can sorta get the basics, but i still have a few questions:

1. Are the words, aside from the basics (to, ot, tem, met, etc.) arbitrary?

2. If so, as one is parsing a text, how does one tell what is a verb name, what the transitivity of that verb is, and what the nicknames of its objects are? For example, if you see "word1 word2 word3 word4 to word5 word6 word7 ot", how do you pick out which word is the verb name?

3. Can this encode anything besides numbers? If so, how?

4. What is the phonology of this "conlang"?
This is the world.
User avatar
Micamo
MVP
MVP
Posts: 5671
Joined: 05 Sep 2010 19:48
Contact:

Re: Oska - Wtf

Post by Micamo »

jseamus wrote:1. Are the words, aside from the basics (to, ot, tem, met, etc.) arbitrary?
Only the defined ones. The defined words can be anything the definer wishes them to be. There are other "base" words that you can build definitions from (and more realistically a number system would most definitely be one of these base things, but I really wanted to give a really show-offy example) but I haven't come up with any yet.
2. If so, as one is parsing a text, how does one tell what is a verb name, what the transitivity of that verb is, and what the nicknames of its objects are? For example, if you see "word1 word2 word3 word4 to word5 word6 word7 ot", how do you pick out which word is the verb name?
The first one. How many there are between the word and to determines the transitivity. I probably should have mentioned it before but "normal" mode is marked just like definitions are.
3. Can this encode anything besides numbers? If so, how?
I'm fairly certain that this definition system is turing-complete. I don't have a proof written up but it can encode the combinator calculus so I'm pretty sure that'll be sufficient. Are there any comp-sci majors here?

Anyway this means you can define any mathematical structure you want, and basically anything you want if you use the "base" words in your definition.
4. What is the phonology of this "conlang"?
I don't really intend to develop this as a full conlang, it's more just a playground for experimenting with the definition idea. But the working phonology I have now is /a e i o u y p t k ɸ s x m n/ <a e i o u y p t k f s x m n> with (C)(C)V(C)(C). Basically all clusters are allowed. Before and after every word a glottal stop is inserted.
My pronouns are <xe> [ziː] / <xym> [zɪm] / <xys> [zɪz]

My shitty twitter
User avatar
Czwartek
sinic
sinic
Posts: 344
Joined: 12 Aug 2010 15:50

Re: Oska - Wtf

Post by Czwartek »

Okay, assuming I understand the rest of this (which I don't) how do you actually define meanings of verbs and nouns? You said that this isn't actually a conlang, so does that mean you can't? What information is actually contained in the examples you've written?
User avatar
jseamus
greek
greek
Posts: 614
Joined: 15 Aug 2010 23:07
Location: Phoenix, AZ

Re: Oska - Wtf

Post by jseamus »

It's a very cool idea, but my ignorance of computer science means I can't really get it, beyond the most basic level. :oops:
This is the world.
User avatar
Micamo
MVP
MVP
Posts: 5671
Joined: 05 Sep 2010 19:48
Contact:

Re: Oska - Wtf

Post by Micamo »

Czwartek wrote:Okay, assuming I understand the rest of this (which I don't) how do you actually define meanings of verbs and nouns? You said that this isn't actually a conlang, so does that mean you can't? What information is actually contained in the examples you've written?
Lemme answer all of these at once with a single answer: The expression "pinto kon te kon fe kon te kon te nim" is just a shorthand. It expands (though technically it reduces) according to the rules of the definition. The expansion is just "suk suk suk suk... seno." As it doesn't use any base words to tie itself down to that's as far as it goes in terms of meaning. It technically only represents numbers as we know them if we mutually understand seno to be 0, suk seno to be 1, suk suk seno to be 2, etc. The seno just delimits our big suk chains. The pinto kon expression just gives us a shorter way to write the big suk chains; "suk suk suk suk suk suk suk suk suk suk seno" and "pinto kon oks kon seno nim" are equivalent by definition, and that's what the definition explains (in VERY precise, extremely rigorous detail). And really, that's all it contains.

And I can't believe this actually happened (my fault for not proofreading) but you guys have the wrong one: The paragraph I gave at first is actually a version I made which accepts inputs in binary. This is what I originally planned to put up but at the last minute I changed my mind and made a base 10 version (and indeed all of my pinto expressions have been the base 10 version). HERE is the correct one, with lines breaking up the different definitions to help with reading:
num tem suk num o seno met
fko em seno to em ot
fko em y suk en to fko se suk em es en ot
mut em y suk seno to seno ot
mut em y suk en to fko se suk em es en ot
pex em y seno to suk seno ot
pex em y suk en to mut se pex em en es em ot
oks to suk seno ot
petf to suk oks ot
mink to suk petf ot
ponk to suk mink ot
musk to suk ponk ot
ftan to suk musk ot
kfat to suk ftan ot
pto to suk kfat ot
ikt to suk pto ot

pam pa po tem pam pa po met
pun tem te o fe
xist ka tem kon ka xist ka o nim met
fot ef it y nim to it ot
fot ef it y kon ka ko to fot ef se ef ka it es ko ot
ketik ek y pam em en to pam fok em mut ek pex se suk ikt es suk en ot
kom em en ex to em se en ex es ot
fast y pam em en to em ot
pinto to fast kom se fot ketik pam seno seno es pokta
As for the design itself I mostly took inspiration from lambda calculus but ironically it doesn't contain any methods of defining anonymous functions (simple, small functions we make solely for the purpose of passing it to a higher-order function) but honestly you can replace those with well-designed combinators anyway.
My pronouns are <xe> [ziː] / <xym> [zɪm] / <xys> [zɪz]

My shitty twitter
Post Reply