, which is actually a wrapper around Robin's original Java implementation, and which is used by camxes.lojban.org. mi'e la mukti mu'o -- You received this message because you are subscribed to the Google Groups "BPFK" group. To unsubscribe from this group and stop receiving emails from it, send an email to bpfk-list+unsubscribe@googlegroups.com. To post to this group, send email to bpfk-list@googlegroups.com. Visit this group at http://groups.google.com/group/bpfk-list. For more options, visit https://groups.google.com/d/optout. ------=_Part_2822_30463946.1413775522461 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable

On Sunday, October 19, 2014 9:19:33 PM UTC-3, guskant wrot= e:

La jbovlast= e allows it based on peg morphology, and la gleki and I don't agree to the = new rule. Concretely, an experimental cmavo "jie'e'e" is now discussed:
= http://jbovlaste.lojban.org/comment= s.html?valsi=3D27527;natlangword=3D0;commentid=3D0;definition=3D0=

A note to clarify why, when and = how jbovlaste came to use the camxes BPFK morphology.

<= div>In March a bu= g was filed against jbovlaste, citing problems with words that wer= e being rejected on the basis of not being approved by "vlatai" -- a tool w= hich is part of "jbofihe"= , and which jbovlaste previously used to validate words. Since jbofihe/vlat= ai was not maintained, in April I posted a message to the lojban list = ;proposing to replace it with camxes, and asking for feedback. I ran al= l of the words in the database through camxes, and only a small number of <= a href=3D"https://github.com/lojban/jbovlaste/issues/39">cmevla an= d fu'ivla&nbs= p;were classified differently by camxes than they had been by jbofihe/vlata= i. I published a list of those words for confirmation.

In June I posted a follow-up, announcing that jbovlaste had been = updated to use a pyth= on implementation of camxes, and detailing the attendant reclassificati= ons: Out of 21,940 words, only about 100 were affected by the change in the= morphological verifier. Most of the effected words were non-conforming fu'= ivla.

All of the implementations of the implementa= tions of camxes distributed at that time, including the the one added to jb= ovlaste, were using an older version of the camxes BPFK morphology: Probabl= y version 108, from = November 2005. The most recent version of the morphology had been completed= in June 2008.

The fact that an older morpho= logy was being used came up in discussions of {relmast}, which had been per= mitted by jbofihe/vlatai, but was forbidden by camxes. After it became appa= rent that camxes was not using the latest BPFK morphology, I updated camxes/vlatai to do= so. 20 cmevla that were previously disallowed were reinstated.

<= br>

Since that time, the various implementations of camxes have b= een updated to use the 2008 camxes BPFK morphology, including Ilmen's camxes.js implementation, which powers the camxes bot in IRC, and python-camxes, which is actually a wrapper around R= obin's original Java implementation, and which is used by camxes.lojban.org.

mi= 'e la mukti mu'o

--
You received this message because you are subscribed to the Google Groups &= quot;BPFK" group.
To unsubscribe from this group and stop receiving emails from it, send an e= mail to bpfk-list= +unsubscribe@googlegroups.com.
To post to this group, send email to bpfk-list@googlegroups.com.
Visit this group at ht= tp://groups.google.com/group/bpfk-list.
For more options, visit http= s://groups.google.com/d/optout.
------=_Part_2822_30463946.1413775522461--