jbovlaste import canonical forms of lujvo lang en: Difference between revisions

From Lojban
Jump to navigation Jump to search
mNo edit summary
 
mNo edit summary
Line 1: Line 1:


Contains an implementation in Python of the gismu scoring algorithm as discussed in CLL 4.14
The standard lujvo scoring formula made by Lojbab and Nora ([http://www.lojban.org/tiki/tiki-download_wiki_attachment.php?attId=194#s12]) favours apostrophes at the cost of consonant clusters. I don't think that this is necessarily the best scoring method for canonical lujvo forms, and think that the old one should be adjusted, or a new one formulated. -[[jbocre: jbovlaste import: tsali lang en]]


(Or will, once Broca figures out why he can't upload it.)
I prefer a more radical change in lujvo scoring. The canonical forms should never include CV'V rafsi. Those rafsi look like cmavo, and have as many syllables as the easily-recognized long rafsi. CVV rafsi should be restricted to final position, where they never require an r-hyphen. Thus canonical lujvo would never contain apostrophes or r-hyphens. -[[jbocre: jbovlaste import: brtais lang en]]

Revision as of 16:53, 4 November 2013

The standard lujvo scoring formula made by Lojbab and Nora ([1]) favours apostrophes at the cost of consonant clusters. I don't think that this is necessarily the best scoring method for canonical lujvo forms, and think that the old one should be adjusted, or a new one formulated. -jbocre: jbovlaste import: tsali lang en

I prefer a more radical change in lujvo scoring. The canonical forms should never include CV'V rafsi. Those rafsi look like cmavo, and have as many syllables as the easily-recognized long rafsi. CVV rafsi should be restricted to final position, where they never require an r-hyphen. Thus canonical lujvo would never contain apostrophes or r-hyphens. -jbocre: jbovlaste import: brtais lang en