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

Re: [lojban] Re: Computing in lojbanistan



At 10:57 AM 06/27/2000 -0400, Invent Yourself wrote:
On Mon, 26 Jun 2000, Arnt Richard Johansen wrote:
> --- In lojban@egroups.com, "LYlun.martins." <lalo@h...> wrote:
> > I'd like to invite experient lojbanists who use Unix, GNU/Linux
> > or similar systems, and use software internationalized trough
> > potfiles (such as most GNU/Linux software), to translate some
> > potfiles to lojban.
>
> We need lots of specialized computer jargon to do that. If we were to
> suddenly start inventing lots of new computer terms, they would, as
> Lojbab has pointed out, run the risk of being flawed and/or malrarbau.

Chicken and egg. People will find it hard to write a document if they
themselves have to invent a great number of the important words as they
are writing.

The problem is that we have a lot of "important words" left to invent that are not jargon, and indeed are the meat of the language, the 2nd tier of abstractions that build a language lexicon up from 1000 words to the 20000-50000 that constitute the lexicon of a college educated speaker, excluding jargon. Inventing these is a different problem from inventing jargon, and is worth spending good time on. Jargon is not.

A major fear I have with ad hoc word-building with one genre (computerese) dominating the writing is that words "important" to that topic and relatively useless outside that field will have speakers inventing 2 and 3 term lujvo for them that rely on knowing that the word is being applied to the computer field in order to understand the metaphor (i.e. jargon), occupying the short-lujvo space with an "assigned meaning" backed with actual usage and thus preventing a more important (as in generally useful) word from making use of that tanru. This has been the norm in the TLI Loglan community, where they have had articles for example coining 50 words for cheeses, and 50 words for stereo equipment, etc. that are not only mediocre tanru but are chewing up lujvo space without asking both questions one must ask of a lujvo: "is this a good tanru to represent the concept" and "can I think of any other concepts that this tanru might suggest that would be more general or more useful". That second question is very hard to ask when we are working on a specialized topic because we have to drop the mind set of the topic to even consider it properly.

Yet, a list of lujvo carries little weight until it has been
tested and proven by history. So let's start at an arbitrary point, with a
weak and untested word-list, and allow the forces of peer-review and usage
to work!

I'm of two minds on this. One mind says that any Lojban usage promotes the language, the other is based on the early history of Loglan, when people spent a lot of time doing just this, trying to find ways to coin the necessary computer jargon because that was what they wanted to talk about, and a) wasting too much time on it and b) coming up with quite miserable results because of the lack of broader experience with the language.

Based on the latter, in translating technical text in the computer and other fields, I would say that new jargon should be added using borrowings from English or whatever, the klutzier the better because that klutziness means that you did not waste a lot of time trying to be elegant but focussed on the structure and logic of the Lojban, which is the hard part of going from idiomatic English to Lojban. If elegance strikes you, go for it, and feel free to go back and revise a borrowing when the perfect tanru comes to you, but don't be afraid to have a text filled with skamrbroda and skamrspageti and whatever else you need, even several skamr- words per sentence. It may look like a hodgepodge of Lojban and Pig Latin at first, but the lexicon will catch up once the texts are being written. With such a writing style, even relative beginners in the language can probably learn to write in the language quite quickly.

How do we stick "x-lojban" where a TWO letter code is allowed?

Data compression? zo'o
--
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