Received-SPF: neutral (google.com: 2001:470:8:42:2c0:f0ff:fe3b:b5b5 is neither permitted nor denied by best guess record for domain of phma@bezitopo.org) client-ip=2001:470:8:42:2c0:f0ff:fe3b:b5b5;
From: Pierre Abbat <phma@bezitopo.org>
To: lojban@googlegroups.com
Subject: Re: [lojban] criteria for the dictionary
Date: Mon, 14 Jan 2013 13:11:47 -0500
Message-ID: <2881164.qORC9BWpb7@caracal>
User-Agent: KMail/4.8.5 (Linux/3.2.0-35-generic; KDE/4.8.5; x86_64; ; )
In-Reply-To: <88bd2464-6efd-4252-9ff6-7c4f68f79e89@googlegroups.com>
References: <88bd2464-6efd-4252-9ff6-7c4f68f79e89@googlegroups.com>
MIME-Version: 1.0
Reply-To: lojban@googlegroups.com
Precedence: list
Mailing-list: list lojban@googlegroups.com; contact lojban+owners@googlegroups.com
Sender: lojban@googlegroups.com
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain; charset=ISO-8859-1
X-Spam_score: -0.0
X-Spam_score_int: 0
X-Spam_bar: /

On Monday, January 14, 2013 04:55:12 jongausib wrote:
> If there's someday is going to be a complete, official lojban dictionary,=
 I
> think there's a need for some criteria for what jbovlaste should contain
> and in which form.
>=20
> Right now the dictionary is rather finite, but with more contributors it
> could expand to an extreme extent.
> I think it's a good idea to discuss this issue now, so I don't contribute
> with a lot of valsi now, and then a few years later someone delete a lot =
of
> my work, because they don't fit into some future official template or lis=
t
> of criteria.
>=20
> *Vocabulary*
> 1. Should we try to add lujvo for all places of each gismu as distinct
> valsi, like {seldri}, {selbai}, {terni'i} etc?

Only if they are glossed as different words, such as "tervecnu". We don't n=
eed=20
lots of entries for "species of <animal>".

> 2. What kind of cmene/cmevla should be added? (with no restriction this s=
et
> could be extremely large)

Names of countries, cities, oblasti, cantons, etc.
Names of diseases.
Names of well-known people (including Lojbanists who are well-known among=
=20
Lojbanists but may not be well-known to the world).
Given names (there may be more than one form of a given name).
Probably some others. I just added "relcibjolmib", a few days after hearing=
 la=20
.camgusmis. talk about it (in English).

A few names can't be entered because they're two or more words (e.g.=20
"kot.divu=E1r").

> For example, we could add recommendation that you only should add cmene
> that could be regarded as having a lexicographical value, like the most
> common names of persons, companies, geographical entities etc. Not the na=
me
> of the street where you are living and shit like that.
>=20
> 3. What kind of fu'ivla should be added?
>=20
> With ALL names of species and chemical substances and other large sets, w=
e
> are going to have a very huge dictionary.
> I've been trying to translate some names of species into lujvo (the
> solution I prefer), but the latin names are often not very descriptive
> and/or logical, so I think one of the better solution (at least for names
> of species etc, you use relative often) is to just lojbanize the latin
> names into fu'ivla.
>=20
> You'll probably already discussed this a lot, but it would be nice to hav=
e
> some guidelines documented somewhere about standards. I believe lojban
> standards about biology, chemistry, music theory and other scientific
> disciplines, doesn't belong to the official grammar of lojban (as little =
as
> Oxford style manual is normative for ALL kind of English language), but
> still it would be nice to have such guidelines (on a level below the
> official language). Especially jbovlaste need such guidelines if we don't
> want to have an inconsistent dictionary with a dukse of words in a possib=
le
> future.

You'll have trouble entering many species names, because they're at least=
=20
three words, such as "cionmau la barda" or "maxri la .durum.". Go for gener=
a=20
and up. But don't try to enter every single genus of fish until there are=
=20
Lojbanic ichthylogists who would need to use them. I've added a few such as=
=20
"skomberu".

Languages and ethnicities: there are only a few thousand of these, so enter=
ing=20
a few hundred wouldn't overload the dictionary. I'd enter "pintupi" (which=
=20
I've mentioned) and maybe "olkola" (which is a valid fu'ivla), but not=20
"Oykangand" (a language closely related to Olkola) until someone figured ou=
t=20
the right way to Lojbanize it.

Chemicals: we need to figure out the proper way to Lojbanize IUPAC before=
=20
entering IUPAC lujvo. Simple-named chemicals, such as geosmin (derpanxu'i) =
or=20
capsaicin (xumrkapsiku), can be entered already. 1,1,1-trichloro-2,2-di(4-
chlorophenyl)ethane will have to wait. Numbers are used three ways in chemi=
cal=20
names (the other is to indicate an oxidation state or valency), and we have=
=20
just the one set of numbers to use in lujvo. I've proposed some experimenta=
l=20
gismu for use in chemical names, such as "xudvu" (aldehyde).

> 4. When is it ok to add a stage-4 fui'vla in the dictionary?
>=20
> I know some lojbanist don't like stage-3 fu'ivla. I do like stage-3. The
> prefix in the stage-3 fu'ivla help you understand a little what this
> foreign word is about. And you could make distinctions easily between for
> example {spatrvanila}, {grutrvanila} "vanilla pod", {tsijrvanila} "vanill=
a
> seed" and {xukmrvanila} vanillin.
>=20
> The only stage-4 fu'ivla I add are those which are very cultural specific=
,
> not easily constructed as a lujvo and/or which doesn't easily fit into so=
me
> cathegory. Stage-4 fu'ivla should also be useful. CLL says: "[stage-4] ar=
e
> used where a fu'ivla has become so common or so important that it must be
> made as short as possible."
>=20
> But as long as you don't add stage-4 without cause (what's the cause of
> making {konjaku} a stage-4 for example? I've never heard of this species
> before), I think those fu'ivlas could really give a good flavor to the
> language, even if this at the same time means that we're going to learn a
> lot of inconsistent words just like learning natlangs. But stage-4 fu'ivl=
as
> could be really cool, my favourites are {iklki} and {fi'ikca}.

Konjac is a common ingredient in Japanese cuisine.

For words that fit into the type-4 format without too much squishing, I don=
't=20
see anything wrong with using type 4 to begin with, except where the type-4=
=20
could easily be interpreted as two unrelated things. For instance, I wouldn=
't=20
use the word "malpigi", as it could equally well mean acerola (rutrmalpigi)=
 or=20
an insect's excretory organ (ragrmalpigi), both named for Marcello Malpighi=
.=20
This turned out to happen with "konjaku" (someone thought it's cognac, whic=
h=20
is koinka), but I didn't find out until after I entered it, as I was thinki=
ng=20
of the Japanese word, which is unrelated to the French word.

I think "tcigaso" should be used as type-4 already. Most people with cars u=
se=20
it.

> *Form*
> I think jbovlaste should have a consistent format before publishing a
> printed version. Some poor fellow would therefore have to read through al=
l
> jbovlaste and edit it into a consistent format just before printing. But =
if
> we would have guidelines from now on already, and we all add valsi in the
> same way, there are going to be less work for someone in the future.
>=20
> 1. Form of definition
> Which format do you think should be standard?
>=20
> {nerkla}:
> a. n1=3Dk1 enters n2=3Dk2 from origin k3 via route k4 using means/vehicle=
 k5
>=20
> b.x1=3Dn1=3Dk1 enters x2=3Dn2=3Dk2 from origin x3=3Dk3 via route x4=3Dk4 =
using
> means/vehicle x5=3Dk5
>=20
> c. x1 enters x2 from origin x3 via route x4 using means/vehicle x5

a for lujvo where the arguments are in order, b where they are not in order=
, c=20
for fu'ivla.

> 2. Etymology
> I suggest that we don't add etymology info in the notes, but use the "add
> etymology"-link in jbovlaste.

I agree. That's what it's there for.

> I think etymology should be mandatory for cmevla and fu'ivla, so you can
> discuss which language to borrow from.
> This is a paranthetical but important question if lojban has ambition to =
be
> as cultural neutral as possible.
> So one recommendation could be that you always use latin for names of
> species, the language most related to the specific cultural
> phenomena/object (or a derivate of languages if many cultures share the
> same phenomena/object, or in that case maybe esperanto).

I wouldn't always use Latin for species. "skomberu", "polgosu", "sperlanu",=
=20
"merlanu", and "merluci" are all from Greek, Latin, or some descendant ther=
eof=20
(though "sperlanu" has a Germanic root), but for the capelin, an important=
=20
forage fish that circles Iceland, I picked the Icelandic word as a source.

> 3. How much info in the notes?
>=20
> And also a final question: Is it possible for a user to edit another user=
's
> notes in jbovlaste, to add info?

It is possible for one user to edit another's definition, but should be don=
e=20
sparingly. Jbovlaste isn't Wiktionary.

Speaking of Wiktionary, there are Wiktionaries in English, French, Lojban, =
and=20
other languages. The English Wiktionary, for any modern language (including=
=20
Lojban), requires that three different people agree on a word at least a ye=
ar=20
ago, or it appear in some well-known work. The Lojban Wiktionary doesn't, a=
nd=20
the French Wiktionary accepts Tsolyani words, which the English Wiktionary=
=20
doesn't. You can enter phrases like "lo xamgu ko li'i" in Wiktionary, but n=
ot=20
jbovlaste.

Pierre
--=20
La sal en el mar es m=E1s que en la sangre.
Le sel dans la mer est plus que dans le sang.

--=20
You received this message because you are subscribed to the Google Groups "=
lojban" group.
To post to this group, send email to lojban@googlegroups.com.
To unsubscribe from this group, send email to lojban+unsubscribe@googlegrou=
ps.com.
For more options, visit this group at http://groups.google.com/group/lojban=
?hl=3Den.