Re: [lojban.org #92] Re: Your lujvo records in Jbovlaste

[Nick Nicholas <opoudjis@optushome.com.au>]

 Taking a break while the Level 0 is compiling to chime in.


>  From: Bob LeChevalier <lojbab@lojban.org>
> > However, the dictionary output format is not fixed, and the final
> > arbiter, at this point, is Nick, although I'm sure Bob could overrule
> > him.
>
> Bob cannot overrule Nick on anything within the scope of the byfy, and 
> Bob
> would not want to.
>
> It is not clear whether "the" dictionary output format is something to 
> be
> decided by the byfy or by anyone at this point.

Bob is right that the format of the dictionary is outside the BPFK's 
mandate. Since I don't want the board deciding this either, clearly 
there will need to be someone or someones designated at some stage to 
make these decisions. (And a single body will need to make those 
decisions.) It makes a lot of sense for that body to include me; and 
since jbovlaste is being engineered towards producing a dictionary or 
dictionaries, the jbvovlaste developers should be involved too. If this 
can wait, I'd rather it wait; if it cannot, then I'll ask the board to 
call for a Dictionary Editorial committee. (The understanding is that 
such a committee presides over format, and over how much goes in; but 
it does not decide the content that goes in, that is obviously the 
BPFK's job.)

I am very concerned that the decision making the BPFK makes *not* be 
constrained by the feasible size of a dictionary. Therefore, where 
grammatical issues will be resolved, I now think a supplement to CLL 
makes sense, over and above a dictionary. And the decision of what goes 
into which volume is by no means urgent.

> There have been many different views over the years as to how people 
> would
> like dictionary definitions to read, with no clear preference given to 
> any
> of them.  Before this year, I would never have contemplated that a 
> change
> in format would in any way imply a baseline change, if the information 
> was
> not changed.

Inasmuch as the lujvo list was never baselined anyway, I don't think it 
counts as a baseline change either; but a consistent format does need 
to be elaborated. I'll chime in on what I think when I get a free 
moment already.

> Most of my own work has been invested in the KWIC format used for
> English-to-Lojban definitions as in the draft dictionary files. It is 
> not
> clear how that work ties into jvovlaste, which I admit that I haven't
> looked at.

If you look at it, of course, it might become clearer to you. ;-)

Though it is hypocritical of me to say, because I have only spent a few 
hours with it, I proclaim unto you that it is vital for Lojbanists to 
play with jbovlaste now during development, to guarantee that it is 
usable as a platform to anchor dictionaries onto, and to forestall any 
catastrophes when half the word stock is already in there.

> I don't see a strong reason why lujvo definitions should be in the 
> exact
> same format as gismu definitions.  cmavo definitions will necessarily 
> look
> different; lujvo have additional information (source etymology) that 
> is not
> relevant to the gismu, while gismu have the word-making etymology that 
> no
> other words have (and I suspect that only gismu will have the much 
> debated
> "metaphorical" aspect to their definition, which I agree needs to be 
> more
> clearly defined so as to rule out polysemy).

This all is true. The main problem I see, though, is how to shoehorn 
cmavo definitions in there; they will necessarily be much more 
discursive, although a CLL supplement would forestall at least some of 
that.



--
Dr Nick Nicholas, French & Italian, University of Melbourne, Australia.
http://www.opoudjis.net                   nickn@unimelb.edu.au
"Most Byzantine historians felt they knew enough to use the optatives
 correctly; some of them were right." --- Harry Turtledove.