Difference between revisions of "gismu != primitives"

From Lojban
Jump to navigation Jump to search
m
 
m
Line 1: Line 1:
  
=== Nick's dictionary notes dated April 22, 2001. ===
+
This represents one of the classic Lojban fallacies, and even [[jbocre: James Cooke Brown CB|James Cooke Brown CB]] believed it for a time.
  
'' Commentary by NN Sept. 2002 in [[jbocre: brackets|brackets]]''
+
There is nothing inherently "primitive" about most of the gismu, nor are they intended to be so. As an obvious example, ''nanmu'' means "male human", and could have been represented as "nakni je remna".  But it isn't, because it is a sufficiently high-frequency concept that assigning a [[jbocre: gismu|gismu]] for it made sense even in the [[jbocre: Elder Days|Elder Days]].
 
 
==== Gismu - almost done ====
 
 
 
*[[jbocre: Not sure what we meant by that: ongoing tidying up? The various secondary word lists, such as roget's and oblique places, generated out of gismu?]]
 
 
 
==== Cmavo - based on cmavo list; maybe later combo-freq's ====
 
 
 
[[jbocre: That is, the priority now is to define the cmavo as they appear in the cmavo list -- including such combinations of cmavo as appear there. The notion of exploring which cmavo frequently cooccur as a unit might be considered later for addition as items to be defined. I will add here that not all Lojbanists '''want''' ''lenu'' included as a separate entry --- I certainly don't.|That is, the priority now is to define the cmavo as they appear in the cmavo list -- including such combinations of cmavo as appear there. The notion of exploring which cmavo frequently cooccur as a unit might be considered later for addition as items to be defined. I will add here that not all Lojbanists '''want''' ''lenu'' included as a separate entry --- I certainly don't.]]
 
 
 
* Lojban -> English:
 
** Keyword(s) if appropriate
 
 
 
** Short definition
 
***[[jbocre: As is now widely acknowledged, the keyword has failed as a definition mechanism, and this is the time for the short prose definition. Not the page-long definition and not for the corpus example, as they belong in the maxi-dictionary, not the mini-dictionary. I currently envisage an upper limit of 200 words. I envisage this being the major intellectual labour of the mini-dict. This will inevitably produce argument; such argument must be veered off the mini-dict and onto the main mailing list, at least until the advent of the Elephant.]]
 
 
 
** [[jbocre: Selma'o|Selma'o]]
 
** Grammar reference
 
 
 
*** [[jbocre: i.e. What page/sections a particular cmavo is defined/discussed in CLL]]
 
* English -> Lojban:
 
 
 
** First work from L -> E; also JCB; look for other list
 
***[[jbocre: The issue is, where shall the English to Lojban cmavo entries come from. They should be derived from such keywords as shall be refined from the Lojban to English cmavo definitions; and supplemented to comparable work done in Loglan. I remember mention made of compendia of English function words being looked up as well.]]
 
 
 
** Words or phrases that might be looked
 
***[[jbocre: i.e. how to identify entries to include|i.e. how to identify entries to include]]
 
 
 
** Reference to grammar reference
 
***[[jbocre: The mini-dict must include cross-references to CLL|The mini-dict must include cross-references to CLL]]
 
 
 
** Lojban equivalent(s) to consider, where appropriate
 
***[[jbocre: Not clear on this. I presume it refers to multiple alternative renderings into Lojban|Not clear on this. I presume it refers to multiple alternative renderings into Lojban]]
 
 
 
==== Plants/Animals ====
 
 
 
* Proposed Linnaean genus in ablative case
 
** see Interlingua
 
 
 
**[[jbocre: Referring to my [Borrowing words from Graeco-Latin obbyhorse|Borrowing words from Graeco-Latin obbyhorse]]
 
 
 
==== Cultural words - [[jbocre: fu'ivla|fu'ivla]] ====
 
 
 
* Countries
 
**[[jbocre: As featured [jbocre: Cultural fu'ivla here|Cultural fu'ivla here]]
 
 
 
* Languages (> x speakers)
 
**[[jbocre: There needs to be a minimum number of speakers to guarantee equity; say, 5 million]]
 
 
 
* Others - just samples
 
**[[jbocre: The mini-dict is not in my opinion the time or place for a major fu'ivla creation project. It should only include conservative samples.|The mini-dict is not in my opinion the time or place for a major fu'ivla creation project. It should only include conservative samples.]]
 
 
 
==== Lujvo ====
 
 
 
* Much weeding - freq, badly [[jbocre: illegible word|illegible word]], obvious
 
**[[Badly most-things, really. The lujvo were found as is, and I imposed place structures on them|Badly most-things, really. The lujvo were found as is, and I imposed place structures on them]]
 
 
 
* Keywords
 
**[[jbocre: In fact, only insert lujvo for which one can work out a keyword|In fact, only insert lujvo for which one can work out a keyword]]
 
 
 
* Prioritize
 
* Look-up usage until sick of it, in priority order
 
 
 
**[[Both the above points = More frequently used lujvo should be worked up before less frequently used. That way, when we run out of time, we have something useful to run with.]]
 
* Regular types
 
 
 
**[[jbocre: There are identified regular types of lujvo in the version of my paper that appeared as the lujvo chapter in CLL; these should be exploited]]
 
** se/te - keywords for other places - DONE
 
 
 
***[[jbocre: If these even deserve a spot in a lujvo list (I am not convinced they do), their keywords are already known, because the x2-x5 places of the gismu list have been keyworded]]
 
** mau, tol, nal, etc.
 
 
 
*** places - automatic
 
*** keywords - manual
 
 
 
***[[jbocre: That is, the place structures of -mau, tol-, nau- lujvo can be generated automatically --- and should. The keywording will still need to be manual|That is, the place structures of -mau, tol-, nau- lujvo can be generated automatically --- and should. The keywording will still need to be manual]]
 
* Place structures
 
 
 
**[[jbocre: Subtext: there is no way in hell we'll get all the place structures of all the lujvo done '''and''' vetted. How do we get as much done usefully as possible?]]
 
** start #2 for others, once prioritized
 
 
 
***[[jbocre: I presume this means start assigning place structures only after lujvo have been put in a priority order|I presume this means start assigning place structures only after lujvo have been put in a priority order]]
 
** code "name-like, so not continuing to look"
 
 
 
** verblike - attempt full definitions
 
***[[jbocre: I proposed a distinction between noun-like lujvo, likelier to be used as sumti --- for which it is not a priority to derive place structures, because their places are less likely to be used; and verb-like lujvo, likelier to be used as selbri --- for which it is a priority]]
 
 
 
** gau, sim, etc. templates, place structures from NSN's rant
 
***[[jbocre: My [http://www.lojban.org/files/papers/lujvopap.3 riginal lujvo paper]] contained more proposed templates on forming place structures than had made it into CLL. These can be exploited]
 
 
 
* Marking for place sureness
 
** Pre '94 - because ''complete'' attempt
 
 
 
** Post '94 - not complete.
 
** Checked against guidelines.
 
 
 
**[[jbocre: How to eliminate subjectivity and tendentiousness in the place structures to be included? By including typographical indication of the degree of certainty for the structures. I was vociferous that, to get this done, we should concentrate on the first 1 or 2 places wherever possible ("name-like lujvo"), and indicate this so the reader knows that there might be more,  trailing places, but only these ones will be made official. "Verb-like lujvo", for which place structures have been fully elaborated, are indicated with a different asterisk or whatever. The source of the lujvo must be noted. The pre 94 lujvo claimed to set all places for any given lujvo; so they won't have expected trailing places. The old lujvo may need to be completely reconsidered, and the voting mechanism on jbovlaste may render moot the indication of provenance = reputability]]
 
 
 
==== Other ====
 
 
 
[[jbocre: Yes, ladies and germs, this was a to-do list I set Lojban Central :-)]]
 
 
 
* [[jbocre: lojbo karni|lojbo karni]] - when level 0 ready
 
**[[jbocre: The next LK to be issued when the level 0 was ready for distribution. In my book it's been ready for a year, but of course the flurry of activity on POD means its day is on hand.]]
 
 
 
* JL - where from here?
 
** JL
 
 
 
** Lang
 
**[[jbocre: So, distinct language section to JL, I presume. And that the next JL should ask readers what they want to see in it|So, distinct language section to JL, I presume. And that the next JL should ask readers what they want to see in it]]
 
 
 
* Mail Excel worksheet on Robin Lesson topics to Nick's
 
**[[jbocre: Yeah, this I got. :-)]]
 
 
 
* People to do Adventure into Lojban
 
**[[jbocre: My old Adventure game translation. *shrug* presumably someone will want to make a programming project of it; see discussion [jbocre: Lojbanic Software lsewhere|Lojbanic Software lsewhere]]
 

Revision as of 16:50, 4 November 2013

This represents one of the classic Lojban fallacies, and even James Cooke Brown CB believed it for a time.

There is nothing inherently "primitive" about most of the gismu, nor are they intended to be so. As an obvious example, nanmu means "male human", and could have been represented as "nakni je remna". But it isn't, because it is a sufficiently high-frequency concept that assigning a gismu for it made sense even in the Elder Days.