Re: [lojban] lojban for research on human-machine interfaces

2015-08-06 1:15 GMT+03:00 Martin Felipe Perez-Guevara <mperezguevara@gmail.com>:

First of all, Wei and Kyle thank you for your answers.

@Kyle I see the point about the grammar being unambiguous due to its parsing and the direct mapping between phonemes and graphemes also taking pronunciation ambiguity out. Nonetheless I refer to other aspects of the language or even the grammar itself in the sense of knowing how the different components were decided to stay like that. For example any sounds could have been selected to be the phonemes arbitrarily, before creating a nonambigous mapping to graphemes, or this phonemes could have been selected to maximize their auditory discrimination based on some sound variable or something like that. Also for example the grammar rules themselves must have a history due to loglan, but I just wondered if you could point me to some easy guide on the evolution of the grammatical rules selected.

The first version of Loglan was published in Scientific American: http://mw.lojban.org/papri/Scientific_American_article

So it's supposed that phonemic inventory reflects most common phonemes in the world.

I don't think that the current inventory of Lojban matters much for existing speakers. For new speakers you just use such tools as la cirkotci trying to imitate how it is uttered.

No phonemic inventory can be satisfactory for everyone. Two more links:

a. https://docs.google.com/document/d/1_w-c7yM6beFUc_G-XCSLlRfrCewhQosFdQuPD1DwhuU/edit#heading=h.dqp8c7hl3yvp

b. https://listserv.brown.edu/archives/cgi-bin/wa?A2=ind1505A&L=conlang&P=19780

As for grammar most notable aspects of it compared to e.g. English:

1. A very rich vocabularly. People usually stick to the set of words and constructs used in their native language(s). They don't need or don't want to use anything else (unless they start learning Lojban).

One of the recent and still standing Whorfian (or pseudo-Whorfian depending on your choice) case is that {co'i} particle marks what is called in linguistics "aorist" or "perfective aspect". English mostly lacks a separate word or affix for it mixing it with past tense, perfect tense but never separating it from similar aspects. Russian and chinese do have more clear tools to separate them but speakers of those languages simply use them without realizing what they really mean. So the only way for an English speaker to understand this concept is from usage by other Lojbanists. If you ask them to use {co'i} more then passively you will understand this particle. Lengthy academic explanations might not help here (just as children don't learn their first language(s) from books but from usage).

2. Dealing with raising and metonymy. In English we can say "Chomsky and Plato are here in this room, on the same shelf" whereas in Lojban you would probably want to use additional markers to separate Plato as a human and Plato as a book with his works.

3. Superficially visible in grammar "Principle of least effort". Not that some usable language violates this principle but simply isolating grammar of Lojban additionally allows dropping particles that are unremovable in English. Like

{lo mlatu} can means "a cat, the cat, cats, the cats" because Lojban doesn't force you to add definite/indefinite articles or grammatical number.

However, this principle of least effort in general doesn't go further into having semantically ambiguous words.

So Lojban lacks a word that both means "The accordion box is too small" and "It will rain next Wednesday".

This is not a strict rule, but simply a trend. A generic verb {co'e} can denote any relation depending on context.

In general I mean some material or maybe a person to contact that could introduce me to the path of creation of the language and not just to its final state.

It would be really nice if we could discuss this by voice/chat, thank you again for your guidance.

There is a text chat http://mw.lojban.org/extensions/ilmentufa/irci/ It is active in certain hours (we all live in different timezones across the globe so it is inactive every day in certain hours). As for voice chats we usually do them either via Mumble or Skype. Announcements are usually published in this thread: https://groups.google.com/d/topic/lojban/y0YvUItExQs/discussion

Or just watch this mailing list for new messages.

On Wednesday, August 5, 2015 at 2:45:02 AM UTC+2, Kyle Roucis wrote:
Hello Martin,

I think I can speak for many of us in saying “Welcome!” (or { coi }). Has anyone other than Wei and myself responded to your inquiry? If not, I have a small amount of information that may help.

1) For the grammar and other aspects of the language like the phonetics, the ortography and vocabulary. Are there any informal records, simulations or experiments leading to settling down on the particular aspects of the language? For example selecting the set of phonemes could be based on some kind of optimization.

Most, if not all, of these things are already codified in The Complete Lojban Language (also known as the Reference Grammar or CLL). This set of documents includes orthography, pronunciation, bridi structure, word classes, and so much more. Additionally, jbovlaste is a dictionary system used by the community to find, categorize, and occasionally define words.

I’m not certain what kind of records, simulations, or experiments you would be looking for. The CLL is largely unambiguous about its pronunciation, orthography, and grammar (though some changes are in the works). On the other hand, there is a growing corpus of lojban creations, conversations, and proposals that help to demonstrate the practical applications of the CLL. Virtually everything you could want related to lojban can be found at the Lojban Mediawiki.

2) What is the estimate of people in the community and their fluency. Do you think people in the community would be interested in participating in cognitive experiments designed for Lojban?
Some experiments might be paid and all results would be public and I hope leading to peer reviewed publication.

Sadly, I am not well-integrated into the lojban community and have been relatively inactive in the last few months. I would wager a guess that the total number of people who have or actively do engage in lojban study is between 200-300, though admittedly that seems a little high. Of that group, I only know of a small handful (3-4) that the community at large might be comfortable calling ‘fluent’. Again, this is just my perspective; and I frankly know nothing, John Snow.

I would be very excited to help in any way I can. I have been fumbling with building parsers, speech-to-text and text-to-speech synthesizers, and simple fiction of my own for a while now. If there is any way I can help further, please feel free to ask!

mu’o mi’e la keidji

Kyle Roucis
www.kyleroucis.com
ky...@kyleroucis.com
719-651-8007

On Aug 4, 2015, at 07:33, Martin Felipe Perez-Guevara <mperez...@gmail.com> wrote:

Hi all.

I am currently a PHD student in cognitive and computational neuroscience. I was looking for some conlang as a starting base to explore some ideas to improve human information processing assisted by computers (lets simplify and say improve reading). I particularly needed it as a start point to be easily to parse so I got very interested by Lojban unambigous grammar and decided to use it for my experiments.

So I have two main questions for the community:

1) For the grammar and other aspects of the language like the phonetics, the ortography and vocabulary. Are there any informal records, simulations or experiments leading to settling down on the particular aspects of the language? For example selecting the set of phonemes could be based on some kind of optimization.

2) What is the estimate of people in the community and their fluency. Do you think people in the community would be interested in participating in cognitive experiments designed for Lojban?
Some experiments might be paid and all results would be public and I hope leading to peer reviewed publication.

I think research on Lojban might also be great advertisement for the language and the community. I particularly started learning it myself, I find it fascinating.

Thank you for taking the time to read this and for your future answers.

--
You received this message because you are subscribed to the Google Groups "lojban" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lojban+un...@googlegroups.com.
To post to this group, send email to loj...@googlegroups.com.
Visit this group at http://groups.google.com/group/lojban.
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "lojban" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lojban+unsubscribe@googlegroups.com.
To post to this group, send email to lojban@googlegroups.com.
Visit this group at http://groups.google.com/group/lojban.
For more options, visit https://groups.google.com/d/optout.