[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [lojban] Re: emotions

At 06:38 PM 5/26/03 -0700, Jorge "Llambías" wrote: 
As for usability in lujvo, one that I've often missed is something
correponding to Esperanto -inda, "deserving of".

We considered "deserve" as a gismu, and decided that it was polysemous. It either is the perfective of earned, or it is innately-obliged to

> 5. Words made from one language, as parji was, should be fu'ivla.

The "ji" part seems like it could be from Chinese, but I wouldn't
really know. English-only would have given 'parsi', no?

I don't know.

> Whether
> people think there is a lot of meaning to the 6-language word-making, it
> offers a couple of things: an objective way to decide the "best form",
> dissociation of the word from the keyword in any single source language, so
> that it is less likely to be encoded English (or whatever language).

This goal was sort of defeated by the English keyword list. People
learn the keywords to the point that they sometimes use the wrong
place structure because of a misleading keyword.

I think that depends on whether they memorize the keywords before they start using the language.. Certainly that error happens, but I think people move beyond it quickly. The words I have the most problem with, are the words that most resemble English.

>  This
> is also why fu'ivla should be dispreferred when one can make a lujvo:  a
> lujvo has its own lojbanic meaning, whereas a fu'ivla starts with the
> meaning in some other language and is not really lojbanic.  lujvo-making
> forces you to think about meaning, and jvajvo force you to think about
> place structures (whether you choose to follow jvojva or not, considering
> them is a good idea).
>          Nora looked up other experimental gismu in jbovlaste, and points
> out that even more than parji, "mango" has no business as a gismu, and
> benzo is almost as questionable.

And that's about the whole list, isn't it? It seems like you are making
the issue seem far bigger than it really is. Even if all the experimental
gismu from the wiki were transferred to jbovlaste, I don't think they are
more than 50, and almost all of them are cultural words. A few words like
mango, pitsa or taksi have a special status in that they are international
_and_ are already gismu-form without any need of adaptation.

But we HAVE resisted those up till now, and equally important, it was decided in strong terms (and not just be lojbab) that we wanted to resist these. My own reason for resisting the obvious gismu is because we ridiculed JCB for allowing them willy-nilly, after having lots of people complain. (The number of complaints about "billiards" and "olive" getting gismu were highly prejudicial when we thought of "pitsa". The reaction was more or less "ya gotta be kidding!"

 It is hard
to resist those, since they don't even need a dictionary definition in order
to be understood.

Which is a good reason to resist them. You understand the word without a place structure, and you have a bunch on 1-place predicates - the language isn't much of a predicate language if most relations are unary.

> 6. (hard to explain) the list of existing gismu slants the choice of how
> one makes and interprets lujvo. The semantics of the language is based on
> what has gone before. Adding a new gismu to the coverage of semantic space
> changes the semantic map, and thus could change the color of meaning of
> other words in unexpected ways.

Unexpected = bad ?

Yes.

Changes the semantics after usage has been established with the earlier semantics. Forces relearning.

>          If it is "easy" to add words without thinking about meanings,
> place structures, people will do so.  I contend that, for gismu, this is
> NOT a good thing.

I agree. Not only for gismu, but also for lujvo and fu'ivla. They should
not be added willy-nilly and without due consideration. Especially so in
the case of gismu forms.

We agree for once %^) That is why I pushed for so long with the old noralujv file that people keyword them, then define them with place structures, then decide if they were worth keeping, all before adding them to a dictionary.

All this takes time, which I've been short of while tackling other fires (Nick keyworded most of the file on one of his visits here).

Those other fires are why I resent people saying I should be adding bonafide words to compete with the illegitimate ones. People also want book orders kept up, the accounting transitioned and turned over, and when I have time for technical work, I'm 3 weeks behind in looking at byfy.

> 8. Finally, before there was a byfy, adding gismu to the original baseline
> list was consider fundamental enough that each one was put to a membership
> vote (at LogFest). People were expected to make a case for their word and
> submit it for consideration by the members, and to abide by the
> result. Hence I abided by the elimination of gumri. The current method of
> putting words out there, and having them see usage without the debate,
> without the research, without the discussion, and without abiding by what
> was decided in the past, is disparaging of stability, tradition, and the
> opinions of members who put time and effort into the language in the past.

The members will have to realize at some point that the language will
belong more and more to the users than to the members.

Certainly. What was once done by the members should now be passed to byfy, which for the most part consists of users. But the prescription shouldn't go uncontrolled before it is fully documented (and we wanted the 5 year period on top of that, though lujvo could be added).

lojbab

--
lojbab                                             lojbab@lojban.org
Bob LeChevalier, President, The Logical Language Group, Inc.
2904 Beau Lane, Fairfax VA 22031-1303 USA                    703-385-0273
Artificial language Loglan/Lojban:                 http://www.lojban.org