Talk:Tiki forum archive/Forum 5/9327
Source: 2013 Tiki forum 5, thread #9327. Original authors, UTC times, post IDs, and reply-parent IDs are preserved. This is an archive, not an active article discussion.
Specifying natural language ambiguity in Lojban translation?
Specifying natural language ambiguity in Lojban translation?
baisong · 2009-12-18 14:06 UTC · Tiki post #9327
bad puns:
pa What did the fish say when it hit a wall? Dam re The blind man picked up his hammer and saw. ci Why did the vampire snowman die? Frostbite
Let's say we also had a comprehensive English -> Lojban dictionary that could recognize (1) homophones, (2) multiple word senses and (3) possible novel compounds. If we then also had some sort of meta-markup that could tag one or more lojban tokens as correlating to an English word or sound, these types of plays on words could be described.
If there isn't part of the language that does it already, here's what I imagine it looking like functionally:
{ .i zo'e zo'e <?lang:en lemma="dam" sampa="d{m"?> zo'e zo'e }
And then, if you tag alot of english terms for lojban words, this program could automatically hold IM conversations for me because I normally just intentionally misinterpret things and make bad puns out of them. It's gold.
You could also potentially describe more complex ailments of natural language like connotations and misspellings. Maybe it could also be used (down the line) to automatically explain bad buns or political buzzwords to people who don't speak the language the pun/propaganda is written in.
mi'e ku'us .i fa'oSpecifying natural language ambiguity in Lojban translation?
PierreAbbat · 2009-12-18 14:54 UTC · Tiki post #9329 · reply to Tiki post #9327
> Is there a feature of lojban for this? For example, lets say I have three > bad puns: > > pa What did the fish say when it hit a wall? Dam > re The blind man picked up his hammer and saw. > ci Why did the vampire snowman die? Frostbite > > Let's say we also had a comprehensive English -> Lojban dictionary that > could recognize (1) homophones, (2) multiple word senses and (3) possible > novel compounds. If we then also had some sort of meta-markup that could > tag one or more lojban tokens as correlating to an English word or sound, > these types of plays on words could be described. > > If there isn't part of the language that does it already, here's what I > imagine it looking like functionally: > > { .i zo'e zo'e <?lang:en lemma="dam" sampa="d{m"?> zo'e zo'e }
If I understand you right, you can do this with a type-1 fu'ivla: "me la'o gy. dam .gy". But I'd rather make puns in Lojban than try to translate English puns, just as if I'm speaking French or Spanish, I make up puns in those languages. L'aveugle traverse la route et la voie. The difference in Lojban is that it doesn't have, except in cmene, two identical-sounding words with completely different meanings, and it doesn't have two words spelled the same, such as "axes", which are pronounced differently. But there are words that sound similar but have different meanings (in Alice, I often confused "noltruni'u", the Queen, and "noltroni'u", the Duchess). Speaking of axes: lu li'o le nu le terdi cu carna le jendu cu cacra li revo li'u i lu lo'e nu catra zo'u sei la noltroni'u cu cusku ko vimcu le stedu ti li'u
mu'omi'e .pier.
-- When a barnacle settles down, its brain disintegrates. Já não percebe nada, já não percebe nada.
To unsubscribe from this list, send mail to [email protected]
with the subject unsubscribe, or go to http://www.lojban.org/lsg2/, or if
