zasni gerna mekso

[guskant]

I have two questions.

1. 

The current peg

http://subvert-the-dominant-paradigm.net/~jbominji/code/lojban_grammar.peg

has "quantifier" in "fragment", and parses numbers and vei-ve'o clause alone.

However, zasni gerna

http://www.lojban.org/tiki/zasni+gerna

does not parse them. Is it intended, or maybe do we need "quantifier" in "sentence" as follows?

sentence <- (term+ CU#?)? bridi-tail / TUhE# paragraphs TUhU#? / gek sentence GI# sentence / term+ ZOhU# sentence / term* VAU#? / quantifier 

2.

la gleki's message on the main list

https://groups.google.com/d/topic/lojban/3mWfITBZweQ/discussion

made me realize that zasni gerna cannot deal with connectives of MEX operators, because MEX operators themselves are joik. Is it intended, or do we need to solve the issue?

[selpa'i]

la .guskant. cu cusku di'e

> la gleki's message on the main list
> https://groups.google.com/d/topic/lojban/3mWfITBZweQ/discussion
> made me realize that zasni gerna cannot deal with connectives of MEX
> operators, because MEX operators themselves are joik. Is it intended, or
> do we need to solve the issue?

This is precisely why I included "connective + connective" in my
connective proposal as things that could be connected by a connective.
It's needed for mekso.

http://selpahi.weebly.com/lojban1/how-to-substantially-simplify-the-lojban-connective-system-my-connective-system

Of course it means that not only {su'i je pi'i} but also {je je ja}
becomes grammatical, but that doesn't seem like a problem to me.

mi'e la selpa'i mu'o

[Jorge Llambías]

On Wed, May 21, 2014 at 10:28 AM, guskant <gusni...@gmail.com> wrote:

I have two questions.

1. 

The current peg

http://subvert-the-dominant-paradigm.net/~jbominji/code/lojban_grammar.peg

has "quantifier" in "fragment", and parses numbers and vei-ve'o clause alone.

However, zasni gerna

http://www.lojban.org/tiki/zasni+gerna

does not parse them. Is it intended, or maybe do we need "quantifier" in "sentence" as follows?

sentence <- (term+ CU#?)? bridi-tail / TUhE# paragraphs TUhU#? / gek sentence GI# sentence / term+ ZOhU# sentence / term* VAU#? / quantifier 

It was intended, along with isolated connectives, relative clauses, linkargs, etc. I probably should have excluded "term* VAU#?" too since it's not really a sentence. "da poi do" or "cumki fa lo nu do" probably shouldn't be grammatical. If I was to add them back I'd do it at the text level rather than at the sentence level.

 

2.

la gleki's message on the main list

https://groups.google.com/d/topic/lojban/3mWfITBZweQ/discussion

made me realize that zasni gerna cannot deal with connectives of MEX operators, because MEX operators themselves are joik. Is it intended, or do we need to solve the issue?

 I think we need to consider more carefully the difference between logical connectives on one hand and non-logical connectives  (including MEX operators) on the other. The latter are essentially functions that take two sumti and return a new sumti, "ko'a (connective) ko'e" should always be paraphraseable as "lo broda be ko'a bei ko'e" for a suitable broda, whereas the former are functions that take two bridi and return a new bridi. It is reasonable for logical connectives to turn up all over the place as a form of shorthand when connecting two bridi with repeating sections, but it's not so clear to me that non-logical connectives and operators should connect anything other than sumti. The meaning of the result is usually rather obscure. I agree with selpa'i that "je je ja", or basically anything else connected with logical connectives, is generally fine, because it just gets expanded into two bridi and we know exactly what it means, but things like "je vu'u ja" would have a much more mysterious meaning, if any. 

mu'o mi'e xorxes

[guskant]

Le jeudi 22 mai 2014 06:20:42 UTC+9, Jorge Llambías a écrit :

On Wed, May 21, 2014 at 10:28 AM, guskant <gusni...@gmail.com> wrote:

I have two questions.

1. 

The current peg

http://subvert-the-dominant-paradigm.net/~jbominji/code/lojban_grammar.peg

has "quantifier" in "fragment", and parses numbers and vei-ve'o clause alone.

However, zasni gerna

http://www.lojban.org/tiki/zasni+gerna

does not parse them. Is it intended, or maybe do we need "quantifier" in "sentence" as follows?

sentence <- (term+ CU#?)? bridi-tail / TUhE# paragraphs TUhU#? / gek sentence GI# sentence / term+ ZOhU# sentence / term* VAU#? / quantifier 

It was intended, along with isolated connectives, relative clauses, linkargs, etc. I probably should have excluded "term* VAU#?" too since it's not really a sentence. "da poi do" or "cumki fa lo nu do" probably shouldn't be grammatical. If I was to add them back I'd do it at the text level rather than at the sentence level.

iesai e'o galfi fi zoi peg

statement-1 <- sentence (I# joik? tag? BO# sentence)* / term* VAU#? / quantifier 

sentence <- (term+ CU#?)? bridi-tail / TUhE# paragraphs TUhU#? / gek sentence GI# sentence / term+ ZOhU# sentence

peg pe'u 

 

2.

la gleki's message on the main list

https://groups.google.com/d/topic/lojban/3mWfITBZweQ/discussion

made me realize that zasni gerna cannot deal with connectives of MEX operators, because MEX operators themselves are joik. Is it intended, or do we need to solve the issue?

 I think we need to consider more carefully the difference between logical connectives on one hand and non-logical connectives  (including MEX operators) on the other. The latter are essentially functions that take two sumti and return a new sumti, "ko'a (connective) ko'e" should always be paraphraseable as "lo broda be ko'a bei ko'e" for a suitable broda, whereas the former are functions that take two bridi and return a new bridi. It is reasonable for logical connectives to turn up all over the place as a form of shorthand when connecting two bridi with repeating sections, but it's not so clear to me that non-logical connectives and operators should connect anything other than sumti. The meaning of the result is usually rather obscure. I agree with selpa'i that "je je ja", or basically anything else connected with logical connectives, is generally fine, because it just gets expanded into two bridi and we know exactly what it means, but things like "je vu'u ja" would have a much more mysterious meaning, if any. 

mu'o mi'e xorxes

ie 

Le mercredi 21 mai 2014 23:07:49 UTC+9, selpa'i a écrit :

la .guskant. cu cusku di'e 

> la gleki's message on the main list 
> https://groups.google.com/d/topic/lojban/3mWfITBZweQ/discussion 
> made me realize that zasni gerna cannot deal with connectives of MEX 
> operators, because MEX operators themselves are joik. Is it intended, or 
> do we need to solve the issue? 

This is precisely why I included "connective + connective" in my 
connective proposal as things that could be connected by a connective. 
It's needed for mekso. 

http://selpahi.weebly.com/lojban1/how-to-substantially-simplify-the-lojban-connective-system-my-connective-system 

Of course it means that not only {su'i je pi'i} but also {je je ja} 
becomes grammatical, but that doesn't seem like a problem to me. 

mi'e la selpa'i mu'o 

 

mi tugni la selpa'i fo lo ka gerna

i ku'i lo ka smuni zo'u ma'a nitcu za'u lo ka tcila

i mu'a tu'e

1 mai

lu mi jajejijoju do klama li'u ma se smuni

ibo e'u la'e lu li'o [ [ [ [ ja je ] ji ] jo ] ju ] li'o li'u iepei

i

2 mai

xu loi re terjonma'o cu gau gendra

i va'o lo nu da'i go'i kei

lu li by su'i va'a cy pi'i dy li'u pe lo zasni gerna 

cu mintu 

lu li (vei by su'i (pe'o va'a cy ku'e) ve'o) pi'i dy li'u

ji

lu li (vei by (su'i va'a) cy ve'o) pi'i dy li'u (to va'o lo nu da'i lu su'i va'a li'u fancu fo da toi)

vu'o pe lo cabna gerna lo si'o smuni

tu'u 

i sei srana se'u 

2-1 mai

va'o lo nu zo pe'o je zo ku'e jeji'a zo ma'o jeji'a zo na'u na cmima lo zasni gerna kei

lo pe'o zei fancu be lo na'ebore mei cu gau mo

i pei e'u gau gendra fa lo nu loi za'u no zo gi cu nenri lo mekso

i mu'a tu'e

lo'u li vei geigi by gi cy gi dy ve'o su'i ny le'u 

basti lu li pe'o gei by boi cy boi dy ku'e su'i ny li'u

i

lo'u li va'agi cy le'u 

basti lu li pe'o va'a cy ku'e li'u no'u lu li tu'o va'a cy li'u

i

lo'u li fu'u xify gi xy gi zy le'u 

basti lu li pe'o ma'o fy boi xy boi zy ku'e li'u

tu'u