[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [lojban] New year's resolution: Finish the dictionary
Am Mon, 20 Jan 2014 15:42:44 +0400
schrieb Gleki Arxokuna <gleki.is.my.name@gmail.com>:
> I saw this message in #jbopre channel
> "New year's resolution: Finish the dictionary"
Well, I’ll do my very best! ;-)
<http://jbovlaste.lojban.org/personal/Wuzzy?extra=valsi>
But the big questions are: When is the dictionary considered
“finished”? Can it even ever be finished?
I think the dictionary can never be possibly finished because language
is an ongoing process, even for Lojban. You can express basically any
kind of relation with the brivla and concepts come and go. The amount
of possible lujvo—for example—is infinite since you can always attach
*another* rafsi.
Even in natural languages terms are invented every day. Especially when
innovations in technology and science happen, new terms are neccessary
to communicate. Since technology is not static, we can’t just make a
bunch of words and say we’re “finished”. Just wait 10 years and you
will see a lot of techincal innovations for which the “finished
vocabulary” hasn’t the right terms for.
So the maybe the stated goal of “finishing” the dictionary is flawed or
at least very vague. A more concrete goal would probably help more.
But this is just opinion, feel
free to add your own into this discussion. :-)
I also want to stress yet again that I still wait for the selma'o
zi'evla (Lojban words for selma'o) to be added to the dictionary:
<http://mw.lojban.org/index.php?title=selma%27o_zi%27evla>
I have posted a CSV file containing definitions, notes and etymology
earlier on the mailing list. Twice. Nobody cared. The words already have
been discussed and I think the resulting zi'evla are ready to be added
into jbovlaste. If the CSV file got somehow lost, I can re-upload it;
just ask me.
> Is it a top secret or there is some info of the plan, ToDo list etc.
> for us?
It is interesting you ask this. I have recently developed a strategy
for systematically creating new brivla én masse. But this was more like
personal ideas, still not very well developed. But I currently try to
stick to this strategy when looking for new words to create. So: Yes,
there is a “plan”, and you may even consider it a “secret” because it
was just private ideas of me.
So probably I should make it public then:
= How to create a lot of brivla systematically =
1. lujvo
1.1. direct/“mechanical” derivations from gismu
1.1.1. from alternating/broad definitions
1.1.2. from notes
1.1.3. eliminate annoying places with “zi'o”
1.2. semantic derivations from gismu
1.2.1. Think of any gismu–gismu combinations which would make sense
1.2.2. cmavo + gismu
1.2.3 standad lujvo conventions from lujpap.3 (gasnu, rinka, zanru, …)
1.2.4. “to'e” and “na'e” variants, when it makes sense
1.3. fu'ivla/cmavo/lujvo/cmevla + fu'ivla
1.4. cmevla + fu'ivla/lujvo/(cmavo)
2. fu'ivla
2.1. stage 3
2.1.1. semantic derivations from gismu if lujvo is not possible
2.2. stage 4
2.2.1. short established, often used stage-3 fu'ivla (if possible)
2.2.X. other than that, try to prefer stage-3 before stage-4
2.3. any stage
2.3.1 “root” fu'ivla: A new fu'ivla which is created as a base for a
broad range of new lujvo (see 1.3.)
This is only a rough stragegy I have worked out for my personal work.
But one main benefit of it is that minimizes bias from other languages
(at least I hope so), especially the mechanical gismu derivations.
Shoehorning words from other languages into lujvo does not always work
well in Lojban. However, this strategy is still young and may change.
To help you understand what I have just written down, here I give you
some examples:
1.1.1. “pinji” is defined as “penis/cliotris” → This leads to two new
lujvo, one explicitly for “penis” (“nakpinji”) and
“cliotris” (“fetpinji”)
1.1.2. The idea is whenever a certain (undefined) lujvo is
mentioned in the notes, you have a new lujvo idea. One example is
“tsapyvu'i”, mentioned in “vrusi”; it is now properly defined. :-)
1.1.3. “zilkei”: x1 is a toy. But still problematic for now; there is no
real broadly used convention how to kill a certain place. This is
because the “zi'o” behaves grammatically strange because it acts like a
sumti. There is no convention except for killing x1 (“zi'o brode”),
which would be “zilbo'e”. But what about killing other places?
Other gismu which certainly would deserve to get their places killed
with “zi'o” include: “latna” (x3, x4), “botpi” (x4).
1.2.1. Most words I created on jbovlaste recently fall into this
category. I looked at “balre” (blade) for example and came up with
“jincyba'e” (“blade” of scissors). I basically go through the gismu
list alphametically and look on gismu X and start to ask myself:
“Which variants of this gismu are the most obvious to me?”.
1.2.2. “te'o zei dugri”: natural logarithm (literally e-type of
logarithm” :-)). “rarna dugri” sounds too artificial to me.
1.2.3. “fazgau”: annoy (non-agentative), “fazgau”: annoy (agentative)
1.2.4. “taske” (“thirsty”) lead me to “toltaske” (“quenched”) and
“nortaske” (roughly “neither thirsty nor quenched”). Similar for
“xagji”.
1.3. “xy. zei kantu”: X-ray (suggested in CLL, no entry in jbovlaste
yet), but such words would be probably rare
1.4. This is still hypothetical, could be used for concepts which are
“named” after a person in other languages. One possibility would be
“Turing machine”. If we think of the hypothetized lujvo “sucyskami” as
“abstract machine”, then “Turing machine” could be “turing. zei
sucyskami”. All of this is hypothetical. Note that when one wants to
combine a person name with a /gismu/, one could consider creating a
stage-3 fu'ivla insted. Example: “Graham’s number”
could be lojbanized as “gry'ym zei namcu” or “namcrgra'ama” (which
destroys the original name a bit).
2.1.1. basically most of flora and fauna … Need I say more? Other
people are much more skilled in this area than I am.
2.2.1. Hypothetized example: “brodrgrolisi” → “grolisi”; same
definition as the original. I have never seen this in practice yet. :-(
Strange, as this strategy is outlined in the CLL. Most people start
right away with a stage-4 fu'ivla, even if it’s about a uncommon
concept.
2.3.1. Very hypothetical; I could imagine this becoming useful for
specialist subjects which have many words sharing a single concept.
I hope you will now understand my new strategy. You can see it as a
“ToDo list” if you wish to. ;-)
Now back to the New Year’s resolution. Now that I have presented my
rough brivla creation stragety, then I could state some clearer goal:
Define all lujvo which would result from 1.1.1. and 1.1.2.
The lujvo which would result from this would make a) more vocabulary
(obviously) and b) probably a solid base for even more complex concepts.
--
Wuzzy
XMPP: Wuzzy2@jabber.ccc.de
E-Mail: wuzzy2@mail.ru
--
You received this message because you are subscribed to the Google Groups "lojban" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lojban+unsubscribe@googlegroups.com.
To post to this group, send email to lojban@googlegroups.com.
Visit this group at http://groups.google.com/group/lojban.
For more options, visit https://groups.google.com/groups/opt_out.