[lojban] Named multiples
Daniel Brockman
both about equally no'e certu, so it's time to bring it up here.
(I especially look forward to hearing xorxes' view on this.)
So,
Sometimes when referring by name to a multiple of things, our
natural language intuition conflicts with our Lojbanic intuition.
For example, I drive an Opel, so in Lojban I could say
.i mi klama fu lo me la .opel.
The {lo me}, which is supposed to be a no-op, comes from my
natural language intuition: I couldn't say that I "drive Opel," and
transferred to Lojban that means I can't say {klama fu la .opel.},
so {lo me la .opel.} is a lame attempt to express "_an_ Opel."
But that intuition is malglico (or malzvero): I also couldn't say
that I "drive car," but in Lojban I _can_ say {klama fu lo karce};
there's definitely no need to say {klama fu lo me lo karce}.
Lojban is ostensibly unmarked for number, but we Lojbanists are
not used to named multiples; we only ever talk about named
singles, like {la .lojban.} or {la .xorxes.}.
Because of this de facto convention, we assume that {la .opel.}
refers to a company or a *type* of car (singles), rather than to the
cars (a multiple). So in practice {la} is not number-agnostic.
But there's no ban on named multiples, so {klama fu la .opel.}
should be perfectly okay, since it's obvious from context that in
this case {la .opel.} is a multiple: "the cars named Opel."
Discuss.
--
You received this message because you are subscribed to the Google Groups "lojban" group.
To post to this group, send email to loj...@googlegroups.com.
To unsubscribe from this group, send email to lojban+un...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/lojban?hl=en.
Jorge Llambías
> (I especially look forward to hearing xorxes' view on this.)
really no reason to artificially restrict the syntactic possibilities
of CMEVLA.
> For example, I drive an Opel, so in Lojban I could say
>
> .i mi klama fu lo me la .opel.
mi klama fu lo .opel.
> Lojban is ostensibly unmarked for number, but we Lojbanists are
> not used to named multiples; we only ever talk about named
> singles, like {la .lojban.} or {la .xorxes.}.
really the "name" of your car, at least not any more than your dog's
name is "dog". "Opel" is what your car is:
lo do karce cu .opel.
> Because of this de facto convention, we assume that {la .opel.}
> refers to a company or a *type* of car (singles), rather than to the
> cars (a multiple). So in practice {la} is not number-agnostic.
mi citka la spagetis
sounds absolutely weird to me, because "spagetis" is not a "name" in
the relevant sense. (The only sense in which it is a "name" is that it
has cmevla form.)
> But there's no ban on named multiples, so {klama fu la .opel.}
> should be perfectly okay, since it's obvious from context that in
> this case {la .opel.} is a multiple: "the cars named Opel."
spaghetti. If all "la" does is "just like 'lo' except that the word
that follows ends in a consonant instead of a vowel", then we wouldn't
need "la" at all. What "la" says is that we don't care at all about
the meaning of the word that folllows, we care only about its form.
But when you say that you drive an Opel, the word Opel does have a
meaning, just like when you say you eat spaghetti, the word spaghetti
does have a meaning.
So the problem is not really with "lo me", which does (practically)
nothing, but with "la", which takes away the meaning of whatever
follows. When cmevla happen to have meaning, as in .opel. or
.spagetis., we don't really want to take it away from them.
mu'o mi'e xorxes
Daniel Brockman
>> (I especially look forward to hearing xorxes' view on this.)
>
> My view on this is that CMEVLA should be merged with BRIVLA. There is
> really no reason to artificially restrict the syntactic possibilities
> of CMEVLA.
>
>> For example, I drive an Opel, so in Lojban I could say
>>
>> .i mi klama fu lo me la .opel.
>
> With ".opel." as a predicate, this would be:
>
> mi klama fu lo .opel.
but I didn't expect this kind of almost psychedelic proposition.
I'm a little afraid to comment on this other than to say that I'm intrigued.
>> Lojban is ostensibly unmarked for number, but we Lojbanists are
>> not used to named multiples; we only ever talk about named
>> singles, like {la .lojban.} or {la .xorxes.}.
>
> Right, because we usually do give names only to singles. "Opel" is not
> really the "name" of your car, at least not any more than your dog's
> name is "dog". "Opel" is what your car is:
>
> lo do karce cu .opel.
cmevla and fu'ivla. That makes sense to me, and actually seems
similar to the intention of the language designers --- what with their
talk of "stage-2 fu'ivla" and everything.
>> But there's no ban on named multiples, so {klama fu la .opel.}
>> should be perfectly okay, since it's obvious from context that in
>> this case {la .opel.} is a multiple: "the cars named Opel."
>
> But the cars are not "named" Opel any more than spaghetti are "named"
> spaghetti. If all "la" does is "just like 'lo' except that the word
> that follows ends in a consonant instead of a vowel", then we wouldn't
> need "la" at all. What "la" says is that we don't care at all about
> the meaning of the word that folllows, we care only about its form.
> But when you say that you drive an Opel, the word Opel does have a
> meaning, just like when you say you eat spaghetti, the word spaghetti
> does have a meaning.
argue that {.xorxes.} has a "meaning". What's the difference?
Jorge Llambías
>
> In effect, what you're saying is that you want to completely merge
> cmevla and fu'ivla.
be divided in four morphological classes: gismu/lujvo/fu'ivla/cmevla,
but there is no need for the last class to have its own separate
syntax.
> I don't really see what you're saying here, unless you also want to
> argue that {.xorxes.} has a "meaning". What's the difference?
all. In any case, in "la xorxes" the meaning of the predicate is as
irrelevant as the meaning of "donri" in "la donri". "la" in any case
removes the meaning from the word that follows.
"x1 is named '...'" would probably be the default meaning for cmevla.
But other cmevla, like "spagetis" or "opel" would carry more meaning
than just "x1 is named '...'".
(In fact "spagetis" is a bad example because "spageti" is a valid the
stage-4 fu'ivla. There is no need to go through the more distorted
".spagetis." when the original is already closer to a proper stage-4
form, but that's the example CLL uses.)
mu'o mi'e xorxes
tijlan
On Fri, May 7, 2010 at 1:04 PM, Daniel Brockman <dbro...@gmail.com> wrote:Merge the syntactic classes cmevla and brivla, yes. Brivla would still
>
> In effect, what you're saying is that you want to completely merge
> cmevla and fu'ivla.
be divided in four morphological classes: gismu/lujvo/fu'ivla/cmevla,
but there is no need for the last class to have its own separate
syntax.
A predicate "xorxes" could mean "is named 'xorxes'", if anything at
> I don't really see what you're saying here, unless you also want to
> argue that {.xorxes.} has a "meaning". What's the difference?
all.
I guess the merging would involve changing the stress rules for cmevla a little bit. Firstly, no stress for monosyllabic cmevla, especially when used as a selbri. Otherwise the likes of {ko'a JAN je prenu} would be a problem. Secondly, the stress position would have to be fixed in some way. Imagine a nation whose name could be lojbanized as {nantanmin} with the original stress on the first {nan}. {ti NANtanmin cidja}, intended to mean "this is Nantanmin-type-of food", could possibly be mistaken for {ti NANtan mincidja} or even {tiNAN tanmincidja}. Doing away with the stress wouldn't solve this. But would {nantanMIN} or {nanTANmin}?
mu'o mi'e tijlan
Daniel Brockman
> little bit.
Daniel Brockman
>> cmevla and fu'ivla.
>
> Merge the syntactic classes cmevla and brivla, yes. Brivla would still
> be divided in four morphological classes: gismu/lujvo/fu'ivla/cmevla,
> but there is no need for the last class to have its own separate
> syntax.
what you're doing is essentially turning cmevla into a variant of fu'ivla.
But talking about it like that is probably misleading, since it will make
people think about morphology instead of syntax.
>> I don't really see what you're saying here, unless you also want to
>> argue that {.xorxes.} has a "meaning". What's the difference?
>
> A predicate "xorxes" could mean "is named 'xorxes'", if anything at
> all. In any case, in "la xorxes" the meaning of the predicate is as
> irrelevant as the meaning of "donri" in "la donri". "la" in any case
> removes the meaning from the word that follows.
> "x1 is named '...'" would probably be the default meaning for cmevla.
> But other cmevla, like "spagetis" or "opel" would carry more meaning
> than just "x1 is named '...'".
in a more sophisticated way than simply enumerating all its referents.
If so, then I agree: {spagetis} definitely has more meaning than {xorxes}.
Enumerating all spaghetti would be impossible; enumerating you is easy.
Anyway, I think your solution is great. Do you see any problems with it?
Jorge Llambías
> 2010/5/7 Jorge Llambías <jjlla...@gmail.com>
>> Merge the syntactic classes cmevla and brivla, yes. Brivla would still
>> be divided in four morphological classes: gismu/lujvo/fu'ivla/cmevla,
>> but there is no need for the last class to have its own separate
>> syntax.
>
> little bit.
what I'm saying. All the phonological rules accomplish is break down a
string of phonemes into a string of words. Once we have those words
identified, how they are then structured to form a text is independent
of the previous step.
> Firstly, no stress for monosyllabic cmevla, especially when used
> as a selbri. Otherwise the likes of {ko'a JAN je prenu} would be a problem.
it is stressed or not. Any string of phonemes without an intervening
pause that ends in consonant plus pause is a cmevla. Stress is
irrelevant. If you are thinking that "JAN je" could be identified as a
gismu, it can't because the cmevla must end with a pause. Just as in
"la JAN ji la JEN cu prenu", which can't be understood as "la janji la
jencu prenu".
> Secondly, the stress position would have to be fixed in some way. Imagine a
> nation whose name could be lojbanized as {nantanmin} with the original
> stress on the first {nan}. {ti NANtanmin cidja}, intended to mean "this is
> Nantanmin-type-of food", could possibly be mistaken for {ti NANtan mincidja}
"nantan" and no pause after "min", while "nantanmin cidja" requires no
pause between "nantan" and "min", and a pause after "min". In fact "la
NANtanmin cidja" is currently grammatical and cannot be mistaken as
"la NANtan mincidja".
> or even {tiNAN tanmincidja}. Doing away with the stress wouldn't solve this.
> But would {nantanMIN} or {nanTANmin}?
final consonant. That's their identifying trait.
mu'o mi'e xorxes
Daniel Brockman
Jorge Llambías
>> Anyway, I think your solution is great. Do you see any problems with it?
>
> Maybe most obvious: {la .daniel. cu xagji} now requires a {cu}.
a big problem. And the main advantage (which is being able to form
tanru with a mix of cmevla and brivla) greatly offsets that minor
inconvenience in my opinion.
The real problem I see is that half(*) of the community would claim
that this amounts to pulling the rug from under their feet as they are
trying to learn the language, and all that stuff.
(*) Give or take some random number.
mu'o mi'e xorxes
tijlan
On Sat, May 8, 2010 at 5:15 PM, tijlan <jbot...@gmail.com> wrote:
> 2010/5/7 Jorge Llambías <jjlla...@gmail.com>
>>
>> Merge the syntactic classes cmevla and brivla, yes. Brivla would still
>> be divided in four morphological classes: gismu/lujvo/fu'ivla/cmevla,
>> but there is no need for the last class to have its own separate
>> syntax.
>
> I guess the merging would involve changing the stress rules for cmevla aNo, the phonological/morphological rules are totally independent of
> little bit.
what I'm saying. All the phonological rules accomplish is break down a
string of phonemes into a string of words. Once we have those words
identified, how they are then structured to form a text is independent
of the previous step.
Yes, I understand that. What I'm failing to see in spite of my 3 years of Lojban experience is how a listener (not a reader) is supposed to recognise the 'pause after the final consonant' of a cmevla. How exactly is the pause supposed to be realised orally, so as to differentiate the sequence of sounds "min.ci" from "minci"?
Jorge Llambías
>
> Yes, I understand that. What I'm failing to see in spite of my 3 years of
> Lojban experience is how a listener (not a reader) is supposed to recognise
> the 'pause after the final consonant' of a cmevla. How exactly is the pause
> supposed to be realised orally, so as to differentiate the sequence of
> sounds "min.ci" from "minci"?
But this difficulty is independent of whether the grammar of cmevla is
extended or not. It already exists with the grammar as it is.
mu'o mi'e xorxes
Jonathan Jones
2010/5/8 Jorge Llambías <jjlla...@gmail.com>On Sat, May 8, 2010 at 5:15 PM, tijlan <jbot...@gmail.com> wrote:
> 2010/5/7 Jorge Llambías <jjlla...@gmail.com>
>>
>> Merge the syntactic classes cmevla and brivla, yes. Brivla would still
>> be divided in four morphological classes: gismu/lujvo/fu'ivla/cmevla,
>> but there is no need for the last class to have its own separate
>> syntax.
>
> I guess the merging would involve changing the stress rules for cmevla aNo, the phonological/morphological rules are totally independent of
> little bit.
what I'm saying. All the phonological rules accomplish is break down a
string of phonemes into a string of words. Once we have those words
identified, how they are then structured to form a text is independent
of the previous step.
Yes, I understand that. What I'm failing to see in spite of my 3 years of Lojban experience is how a listener (not a reader) is supposed to recognise the 'pause after the final consonant' of a cmevla. How exactly is the pause supposed to be realised orally, so as to differentiate the sequence of sounds "min.ci" from "minci"?
I would imagine they would figure it out from the speaker, you know, pausing. Kind of like, if I were speaking this, I would've paused five times now. Make that six, not including the pause just after the word "six".
--
You received this message because you are subscribed to the Google Groups "lojban" group.
To post to this group, send email to loj...@googlegroups.com.
To unsubscribe from this group, send email to lojban+un...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/lojban?hl=en.
--
mu'o mi'e .aionys.
.i.a'o.e'e ko klama le bende pe denpa bu
Daniel Brockman
>>> Anyway, I think your solution is great. Do you see any problems with it?
>>
>> Maybe most obvious: {la .daniel. cu xagji} now requires a {cu}.
>
> Yes, you would need a few more "cu"s here and there, but that's hardly
> a big problem.
So maybe it's time to capitalize on all those extra {cu}s that people
are already putting there.
> And the main advantage (which is being able to form tanru with a
> mix of cmevla and brivla) greatly offsets that minor inconvenience
> in my opinion.
>
> The real problem I see is that half(*) of the community would claim
> that this amounts to pulling the rug from under their feet as they are
> trying to learn the language, and all that stuff.
Daniel Brockman
(And it's worth pointing out again that we're talking about
a _simplification_ of the grammar --- not an extension.)
Jorge Llambías
> Another benefit is that {la .daniel. ku} would be allowed.
>
> (And it's worth pointing out again that we're talking about
> a _simplification_ of the grammar --- not an extension.)
mu'o mi'e xorxes
tijlan
I actually would very much like the merging (which I might be going to privately call "xorla", although the focus is more on cmevla than {la}), along with your another view that {gi'e} can be replaced with {gije} and {.e} with {je}.
Stela Selckiku
the official grammar where cmevla don't lose their meaning: When
attached into a lujvo with "zei". And so that's one way I'd think to
translate your sentence:
mi klama fu lo .opel. zei karce
mi'e la stela selckiku
mu'o
Jorge Llambías
> Anyway, besides all these reform proposals :), there IS somewhere in
> the official grammar where cmevla don't lose their meaning: When
> attached into a lujvo with "zei". And so that's one way I'd think to
> translate your sentence:
>
> mi klama fu lo .opel. zei karce
mi klama fu lo me zei .opel.
if you feel that adding "karce" is unnecessary or redundant.
The fact that we already can create the brivla "me zei ...." from any
cmevla is another indicator that cmevla should be brivla, like the
rest of open class words.
mu'o mi'e xorxes
purpleposeidon
that you've got a tense-simplification proposal on the Tiki.
Jorge Llambías
<purplep...@gmail.com> wrote:
> Xorxes, do you have a grammar containing all of your changes? I know
> that you've got a tense-simplification proposal on the Tiki.
http://www.lojban.org/tiki/tiki-index.php?page=zasni+gerna+cenba+vreji
Luke Bergen
I noticed something on that page that I didn't understand. Why get rid of sa? What could be more clear than "take next word and erase all previous words back to the last occurrence of the following word"? What confusions arise from this definition/usage? Or maybe I'm confusing sa with su
On May 13, 2010 8:30 AM, "Jorge Llambías" <jjlla...@gmail.com> wrote:
On Wed, May 12, 2010 at 10:48 PM, purpleposeidon
<purplep...@gmail.com> wrote:
> Xorxes, do you ...
Most of them are listed here:
http://www.lojban.org/tiki/tiki-index.php?page=zasni+gerna+cenba+vreji
mu'o mi'e xorxes
--
You received this message because you are subscribed to the Google Groups "lo...
Daniel Brockman
> sa? What could be more clear than "take next word and erase all previous
> words back to the last occurrence of the following word"?
The definition may be clear, but it's apparently a headache to implement.
Have a look for yourself in the normal grammar.
Jorge Llambías
> I noticed something on that page that I didn't understand. Why get rid of
> sa? What could be more clear than "take next word and erase all previous
> words back to the last occurrence of the following word"? What confusions
> arise from this definition/usage? Or maybe I'm confusing sa with su
much erasing to do."
CLL: "The cmavo following ``sa'' should be the starting marker of some
grammatical construct. The effect of the ``sa'' is to erase back to
and including the last starting marker of the same kind."
YACC: "If the word ``sa'' (selma'o SA) is identified, erase it and all
preceding text as far back as necessary to make what follows attach to
what precedes. (This rule is hard to formalize and may receive further
definition later.)"
EBNF: "null = utterance SA"
None of those rules are very clear. One suggested implementation was
to delete back to the preceding occurence of some word of the same
selma'o as the word that follows. Another suggested implemantation was
to select some constructs such that they are eligible for SA-deletion.
Your proposed rule may be clear, but it is difficult to implement from
a formal perspective (because the number of possible words, unlike the
number of selma'o, is infinite). It could probably be done with a
similar trick to the one used for the ZOI-delimiters.
In any case, none of the rules seem to give anything very usable from
a human perspective. You may come up with some simple examples that
may seem to give something useful, but if you explore some more you
always end up with unintuitive cases.
mu'o mi'e xorxes
--
Lindar
Leave names be names.
What's wrong with {.i mi klama fu le mi karce no'u la .opel.} if you
have to be explicit about the brand name?
I've never had a problem with {.i citka la .spagetis.} at any point.
You're eating something named spaghetti. It's the name of the dish.
You're driving something named Opel, it's the brand of the car. Stop
trying to conform Lojban to your native language.
If we make cmevla usable as selbri, then there's little to no point or
motive to create or use fu'ivla, and then we might as well just say
"fuck it" and use all English words written in Lojban phonology. This
is a horrible idea.
By your logic we may as well say {.i mi .going. .stor. .xaus.}, and
then why bother? Just speak English. >_>
To me you're now breaking the uniqueness of Lojban by essentially
making a really stupid shortcut so nobody has to use fu'ivla or lujvo
ever again.
Jorge Llambías
>
> I've never had a problem with {.i citka la .spagetis.} at any point.
> You're eating something named spaghetti. It's the name of the dish.
comes you may end up eating something named Spaghetti.
mu'o mi'e xorxes
tijlan
I do not like this idea at all.
Leave names be names.
{la cribe} is also a name. So you are talking about cmene (words that start with gadri {la}), not cmevla (words that end with a consonant). The proposed view is more about cmevla than cmene. It would not affect the grammar of cmene whatsoever, I think.
There are obvious occasions in which we want to use a (lojbanized) foreign word as a selbri where any good gismu or lujvo alternative escapes, like {ti me la spagetis} or {ti cidjrspageti}. Unlike independent {la spagetis}, which is a name, {me la spagetis} is a non-name syntactic unit, selbri, and does not semantically differ from {spagetis} as selbri in the proposed view. That is, we could simplify the form of cmevla-based selbri without changing the grammar of cmene, names. Names would be left names.
What's wrong with {.i mi klama fu le mi karce no'u la .opel.} if you
have to be explicit about the brand name?
{le mi karce no'u la .opel.} might work if the car is *actually* named {la .opel.}. But isn't {la .opel.} the name of a company? If so, {no'u} is wrong, since it suggests "my car is a company". {pe} is better. {no'u} is for identification; {pe} is for association.
If we make cmevla usable as selbri, then there's little to no point or
motive to create or use fu'ivla, and then we might as well just say
"fuck it" and use all English words written in Lojban phonology. This
is a horrible idea.
We could make a fu'ivla like "fakiti" to mean "fuck it", but we haven't. And if someone actually said {fakiti}, it wouldn't be approved by the community at large. For the same reason, a cmevla selbri like {fakit}, possible in the proposed scheme, wouldn't take root, I think.
By your logic we may as well say {.i mi .going. .stor. .xaus.}, and
then why bother? Just speak English. >_>
The official grammar already allows {mi me la going.stor.xaus}, {mi going zei stor zei xaus}, etc.
Not all Lojbanists speak English. If I want to be understood in Lojbanistan, I should bother to speak Lojban optimally. (Nevertheless, if I was an English speaker and didn't know the valsi for "store", I don't think using "stor" as a makeshif and at the same time asking for correction would be reproachable.)
To me you're now breaking the uniqueness of Lojban by essentially
making a really stupid shortcut so nobody has to use fu'ivla or lujvo
ever again.
This proposed rule for shortcut is not in itself stupid. What is stupid is the assumption that one can be a good lojbanist without learning the proper lojban valsi (gismu, lujvo, cmavo).
Warrigal
> I do not like this idea at all.
> Leave names be names.
simply helping you become right with regards to this particular issue,
I could show you the mistake you made, thereby causing you to never be
wrong the same way again? I mean, I could just give responses to the
arguments you actually made, but that would be like giving you a fish
and feeding you for a day.
Of course, you can't learn the right way until you learn that the old
way was the wrong way. Take a moment to look over what you said, and
see if you can figure out what's wrong with it. I can give you hints,
or, if you like, the answer.
--Warrigal (aka uorygl)
Daniel Brockman
If you really want to participate in the discussion, I suggest you read the
whole thread, then quote something specific and reply to that.
As it stands, your message is so full of ignorance, hyperbole, invective, non
sequiturs and vague emotional appeals that it's almost unanswerable.
Do you really think that the syntactical difference between cmevla and brivla
is a contributing factor to Lojban's uniqueness? If anything, I would guess
that merging the two would make the language *more* unique, not less.
Have you even considered whether English has such a difference?
(Hint: Think about whether you need an article in front of names.)
Luke Bergen
>> sa? What could be more clear than "take next word and erase all previous
>> words back to the last occurrence of the following word"?
>
Jorge Llambías
>
> Wait, the proposal then is to get rid of all of SA, not just sa itself?
> What would be the proposed default meaning of
> something like {ko'a cu djan ko'e ko'i}?
there is no reason for stage-two fu'ivla to have a default meaning.
I would say the most obvious meaning for an otherwise meaningless
predicate word is "x1 is named <word>". I wouldn't expect it to be a
three argument predicate.
mu'o mi'e xorxes
Luke Bergen
Jorge Llambías
> So effectively {mi djan} would be {mi me la djan}.
> And my mistake, I forgot that si,sa,su each had their own selma'o. So you
> have no qualms about si and su then?
Minimiscience
> of
> >> sa? What could be more clear than "take next word and erase all previous
> >> words back to the last occurrence of the following word"?
> >
> > (It's about selma'o, not about specific words.)
>
> Wait, the proposal then is to get rid of all of SA, not just sa itself? I
> don't think I want to live in a world without {si}.
Firstly, that's not what he meant. "{sa}" is currently defined to erase
everything back through the last occurrence of a word of the same {selma'o} as
what follows, not the last occurrence of the same exact word. Secondly, "{sa}"
is the only {cmavo} of {selma'o} SA; each of the three erasure {cmavo} belongs
to its own {selma'o}, which it shares with no other. The fact that they each
have completely different effects on the grammar should have been a tip-off to
you.
> ta'onai. So, I haven't been able to completely follow the conversation
> fully. Would the proposed merging cause cmevla to become valid selbri
> words? e.g. something like {mi spagetis. citka}?
> What would be the ramifications of this?
deal with {fu'ivla} morphology or "{me la}." However, prior texts which did
not separate {cmevla} from following {selbri} (e.g., "{la .ktulxus. senva
denpa}") would either take on a completely different meaning or become invalid.
> What would be the proposed default meaning of something like {ko'a cu djan
> ko'e ko'i}?
> "x1 is named '...'" would probably be the default meaning for cmevla. But
> other cmevla, like "spagetis" or "opel" would carry more meaning than just
> "x1 is named '...'".
mu'omi'e .kamymecraijun.
--
vo selska cu banzu
Luke Bergen
> not separate {cmevla} from following {selbri} (e.g., "{la .ktulxus. senva
> denpa}") would either take on a completely different meaning or become invalid.
Daniel Brockman
> Would it try to suck {cusku} in to the cmevla... where would a parser try to
> put the elided {ku} in that bridi under the proposed change? Or would it be
> a syntax error?
It's not difficult to understand: every cmevla would act exactly as a brivla.
Oleksii Melnyk
deal with {fu'ivla} morphology or "{me la}." However, prior texts which did
not separate {cmevla} from following {selbri} (e.g., "{la .ktulxus. senva
More possible issues:
1. so far, "la bertas marias bender beis ibragim oglys" was one name. It becomes a tanru with a vague meaning.
2. Names was allowed to have any stresses/pauses (afair). We can, probably, have an ambiguous parsing of some weird enough names, especially in spoken conversation, where the "." detection can give a lot of false positives.
--
mu'o mi'e lex
Jorge Llambías
>
> More possible issues:
>
> 1. so far, "la bertas marias bender beis ibragim oglys" was one name. It
> becomes a tanru with a vague meaning.
name, the meaning of the internal tanru is not relevant to that.
> 2. Names was allowed to have any stresses/pauses (afair).
And, like for any other word, you cannot pause in the middle of a
cmevla
> We can, probably,
> have an ambiguous parsing of some weird enough names, especially in spoken
> conversation, where the "." detection can give a lot of false positives.
mu'o mi'e xorxes
Luke Bergen
So this change assumes dotside in order to work, right?
On May 19, 2010 8:38 AM, "Jorge Llambías" <jjlla...@gmail.com> wrote:
On Wed, May 19, 2010 at 2:24 AM, Oleksii Melnyk <lame...@gmail.com> wrote:
>
> More possible issues...
Anything preceded by "la" is a name. "la plipe klupe clupa" is one
name, the meaning of the internal tanru is not relevant to that.
Any stress, yes. Any pause, no. cmevla must be surrounded by pauses.
> 2. Names was allowed to have any stresses/pauses (afair).
And, like for any other word, you cannot pause in the middle of a
cmevla
> We can, probably,
> have an ambiguous parsing of some weird enough names, especially in spoken
> ...
No, there is no change at all in the morphology rules.
mu'o mi'e xorxes
--
You received this message because you are subscribed to the Google Groups "lojban" group.
To post to...
Jorge Llambías
> So this change assumes dotside in order to work, right?
mu'o mi'e xorxes
--
You received this message because you are subscribed to the Google Groups "lojban" group.
Luke Bergen
Jorge Llambías
>>> So this change assumes dotside in order to work, right?
>>
>> No, not at all. They are two separate issues.
> plise} without dotside there to keep {la} from eating up the whole thing as
> a cmene?
change, it would just eat up a selbri (a string of cmevla would be
just another type of selbri).
This has nothing to do with dotside.
> As you said earlier:
>> "la plipe klupe clupa" is one name, the meaning of the internal tanru is
>> not relevant to that.
>
> How do you keep {la} from just sucking everything up into a name without
> dotside? Can {ku} terminate {la}? Is there some rule that gismu can be
> part of a name but cmavo can't (making {cu} the {la} killer)?
same as for LE, yes. With the change, cmevla would no longer be a
special case.
Oleksii Melnyk
Anything preceded by "la" is a name.
That is. {la mersedes bents} always was {named (mersedes bents)}. After the unification, {mersedes bents karce} can be {[named (mersedes)][named bents][karce]} or {[named (mersedes bents)][karce]}, longer names will give a lot more.
And, like for any other word, you cannot pause in the middle of a
cmevla
Not sure. IMO, given the properly stressed speech, listener can recover the accidental extra pause between the syllables of the long gismu/lujvo/fu'ivla/cmevla, with some exceptions _like_ pause before "ku" in "la SELBRIkuLIKENAMEs", which, most probably, will be detected as "parsing error".
It is impossible to avoid the extra pauses in spoken conversation (need to take a breath in the middle of a-very-long-and-difficult-foreign-name, network loose several voice packets, noise, speaker distraction). So, the change can be dangerous.
BTW, pause in the speech is annoying, so, (while it is OK for written language,) the spoken one, (trying to exploit the "isomorphism" with the text,( full of cmene,)) can become a whole mess.
--
mu'o mi'e lex
Jorge Llambías
>
> That is. {la mersedes bents} always was {named (mersedes bents)}.
> After the
> unification, {mersedes bents karce} can be {[named (mersedes)][named
> bents][karce]} or {[named (mersedes bents)][karce]}, longer names will give
> a lot more.
still have to use "la": "lo karce pe la mersedes bents".
>> And, like for any other word, you cannot pause in the middle of a
>> cmevla
>
> Not sure. IMO, given the properly stressed speech, listener can recover the
> accidental extra pause between the syllables of the long
> gismu/lujvo/fu'ivla/cmevla, with some exceptions _like_ pause before "ku" in
> "la SELBRIkuLIKENAMEs", which, most probably, will be detected as "parsing
> error".
you talking about human parsing (which uses the meaning of words to
help with parsing them) or pure machine parsing of sounds without
regard to meaning of the words? The latter is perfectly defined for
Lojban, and is not in the least affected by the merging of cmevla with
brivla.
> It is impossible to avoid the extra pauses in spoken conversation (need to
> take a breath in the middle of a-very-long-and-difficult-foreign-name,
> network loose several voice packets, noise, speaker distraction). So, the
> change can be dangerous.
change has no relevance to word parsing.
> BTW, pause in the speech is annoying, so, (while it is OK for written
> language,) the spoken one, (trying to exploit the "isomorphism" with the
> text,( full of cmene,)) can become a whole mess.
cumbersome type of word, so they don't blend well with normal Lojban
words. Simplifying their syntax would not make them morphologically
prettier, it would only make the syntax simpler.
mu'o mi'e xorxes
Oleksii Melnyk
Can you give an example? ... Are you talking about human parsing?
Yes. Let's take an example from NORALUJV.txt. I've spotted "backemselrErkru" (: it is long enough to either got out of air in the middle of it, or just need to look into the dictionary to recall the next rafsi, whatever :). Any pause after the vowel and the stressed "lrE" leaves us with "rkru" as the last rafsi, so here was some error. Any pause after the consonant will alert us about "no LA before the cmene"(not morphology level, but easy/close enough for humans). Pause after the "CVV" rafsi (not in an example), will be noticed as "no stress in previous word". Only in "la backemselrErkru" the pause between the rafsi's will go unnoticed. In the cmevla, after the required LA, we do expect the sequence of names, so breaking one into several will cause almost no harm, we'll just glue them together up to the last consonant ending word.
So, we are almost immune to the extra pauses inside the long words. Now, without the required LA, we'll get:
ba.ckemselrErkru - can fail
bac.kemselrErkru - OK
back.emselrErkru - can fail
backe.mselrErkru - can fail
backem.selrErkru - OK
backems.elrErkru - can fail
backemse.lrErkru - can fail
backemsel.rErkru - OK
backemselr.Erkru - can fail
backemselrE.rkru - can fail
backemselrEr.kru - OK, can eat the next word
backemselrErk.ru - OK
backemselrErkr.u - OK
Note, that all the "fails" only "_can_ be"; if there are the otherwise _allowed_ pause after the entire word, it will parse, giving us 2 wrong meaningful chunks.
The exact same problem exists with or without the change. The
change has no relevance to word parsing.
So, the change affects an error detection. If the humans all were the reliable electronic devices with the error detection codes in the communication channel, that would be irrelevant. However they are not, and the language was meant to be "error detecting communication channel" for them.
Why would a text be full of cmevla? As you say, cmevla are a
cumbersome type of word, so they don't blend well with normal Lojban
words. Simplifying their syntax would not make them morphologically
prettier, it would only make the syntax simpler.
I do like the idea. I just looking for the bad consequences. The result of the change would be the wider usage of cmevla. Otherwise, there are no reason to make it simpler. Trying to read them aloud would push "the live language" towards toki pona (as the best case).
"." is OK in writing. Not so in speech. We do need to think/breath/rest/etc. sometimes. That sounds just as ".......". So, using "." as the syntax marker looks bad for a human-spoken audio-video isomorphic language.
mu'o mi'e lex
Jorge Llambías
> 2010/5/19 Jorge Llambías <jjlla...@gmail.com>
>
>> Can you give an example? ... Are you talking about human parsing?
>
> Yes. Let's take an example from NORALUJV.txt. I've spotted "backemselrErkru"
> (: it is long enough to either got out of air in the middle of it, or just
> need to look into the dictionary to recall the next rafsi, whatever :).
errors in pronunciation will probably be ignored, or (most likely)
they won't recognize it, in which case they will say "ki'a" whether
there were any errors in pronouncing that monster-word or not.
> Any
> pause after the vowel and the stressed "lrE" leaves us with "rkru" as the
> last rafsi, so here was some error. Any pause after the consonant will alert
> us about "no LA before the cmene"(not morphology level, but easy/close
> enough for humans).
You could be talking about the word "zo backemselrerkru" for example.
> Pause after the "CVV" rafsi (not in an example), will be
> noticed as "no stress in previous word". Only in "la backemselrErkru" the
> pause between the rafsi's will go unnoticed. In the cmevla, after the
> required LA, we do expect the sequence of names, so breaking one into
> several will cause almost no harm, we'll just glue them together up to the
> last consonant ending word.
than one cmevla, so if you say for example: "la pip backemselrerkru",
and you pause somewhere after a consonant of the lujvo, it will be
taken (at least by a mechanical parser) as a cmevla.
> So, we are almost immune to the extra pauses inside the long words. Now,
> without the required LA, we'll get:
>
> ba.ckemselrErkru - can fail
> bac.kemselrErkru - OK
> back.emselrErkru - can fail
> backe.mselrErkru - can fail
Another weird place to stop.
> backem.selrErkru - OK
> backems.elrErkru - can fail
> backemse.lrErkru - can fail
Another weird place to stop.
> backemsel.rErkru - OK
> backemselr.Erkru - can fail
> backemselrE.rkru - can fail
Another weird place to stop.
> backemselrEr.kru - OK, can eat the next word
> backemselrErk.ru - OK
> backemselrErkr.u - OK
Another weird place to stop.
> Note, that all the "fails" only "_can_ be"; if there are the otherwise
> _allowed_ pause after the entire word, it will parse, giving us 2 wrong
> meaningful chunks.
The only realistic unintended pauses are:
ba.ckemselrerkru
bac.kemselrerkru
backem.selrerkru
backemsel.rerkru
backemselrer.kru
The last one the least realistic, because of the stress.
All the others are not reasonable places for unintended stops.
The human hearer will have to decide whether or not to take any such
pauses seriously based on the resulting meaning.
>> The exact same problem exists with or without the change. The
>> change has no relevance to word parsing.
>
> So, the change affects an error detection. If the humans all were the
> reliable electronic devices with the error detection codes in the
> communication channel, that would be irrelevant. However they are not, and
> the language was meant to be "error detecting communication channel" for
> them.
backem.selrerkru" is ungrammatical, therefore I will attempt a
possible correction to "lo backemselrerkru, which fixes the problem",
whereas with the change the human will say "lo backem.selrerkru" is
grammatical, therefore I will not attempt a correction, even though
what I'm hearing makes very little sense". Is that the point?
>> Why would a text be full of cmevla? As you say, cmevla are a
>> cumbersome type of word, so they don't blend well with normal Lojban
>> words. Simplifying their syntax would not make them morphologically
>> prettier, it would only make the syntax simpler.
>
> I do like the idea. I just looking for the bad consequences. The result of
> the change would be the wider usage of cmevla. Otherwise, there are no
> reason to make it simpler.
remove them from the language, not complicate their grammar
needlessly. (Not that their grammar is too complicated, just more
complicated than what it needs to be.) The motivation is not to use
more cmevla, but to allow things like "la cmalu djan" for a name like
"Little John". There is no reason why the change should encourage more
cmevla, since anything that can be said with the change can already
be said in some other way without the change.
> Trying to read them aloud would push "the live
> language" towards toki pona (as the best case).
would push the language towards toki pona?
> "." is OK in writing. Not so in speech. We do need to think/breath/rest/etc.
> sometimes. That sounds just as ".......". So, using "." as the syntax marker
> looks bad for a human-spoken audio-video isomorphic language.
fu'ivla or cmevla. But the proposed change doesn't really affect that.
mu'o mi'e xorxes
Oleksii Melnyk
A human parser will either recognize the word, ... or (most likely)
they won't recognize it, in which case they will say "ki'a" whether
Not so with the names. As they are "arbitrary label", they always are "known", without a recognizing.
Are you saying that with the current grammar ... I will attempt a
possible correction ... whereas with the change ... I will not. Is that the point?
That is. Under the current syntax, names are always clearly labelled, so, listener knows, that here will go something, not expected to be known in advance to understand.
After the change, listener will never be sure, if he should look up the next word in the memory, or just use the letter sequence, it hears, as the name.
So, I'm not sure, if the current syntax complexity is really needless.
I don't understand what you mean by that. Trying to read what aloud
would push the language towards...
So, when we'll got big enough user base, taught on the text with enough unlabelled names, they will tend to use that names in speech, and will tend to shorten the required pause to zero, as it annoys a lot. That will result in a lot of misunderstanding, causing a language drift.
mu'o mi'e lex
Jorge Llambías
>
> Under the current syntax, names are always clearly labelled, so,
> listener knows, that here will go something, not expected to be known in
> advance to understand.
vowel or a consonant.
> After the change, listener will never be sure, if he should look up the next
> word in the memory, or just use the letter sequence, it hears, as the name.
places they need to look it up. Just like now. In "la spagetis" all
that matters is the letter sequence, in "lo me zei spagetis", they
need to look it up. In "la donri" all that matters is the letter
sequence, and in "lo donri" they need to look it up. Word form has
nothing to do with it.
> So, when we'll got big enough user base, taught on the text with enough
> unlabelled names, they will tend to use that names in speech, and will tend
> to shorten the required pause to zero, as it annoys a lot. That will result
> in a lot of misunderstanding, causing a language drift.
consonant" and "word used as arbitrary label". In current Lojban,
words that end in a consonant are not necessarily used as arbitary
labels, and words used as arbitrary labels do not necessarily end in a
consonant. So why use the same word to talk about both?
mu'o mi'e xorxes
Oleksii Melnyk
Names still have to be labeled with "la", whether the word ends in a
vowel or a consonant.
Looks like I've misunderstood the proposal. Excuse me for stealing so much of your time.
--
fe'o mi'e lex
Adam D. Lopresto
> On Wed, May 12, 2010 at 10:48 PM, purpleposeidon
> <purplep...@gmail.com> wrote:
>> Xorxes, do you have a grammar containing all of your changes? I know
>> that you've got a tense-simplification proposal on the Tiki.
>
> Most of them are listed here:
>
> http://www.lojban.org/tiki/tiki-index.php?page=zasni+gerna+cenba+vreji
Interesting. Some of those are definitely good, some strike me as going way
too far (of course), some somewhere in between. But I'm curious about the
last one, allowing relative clauses to attach to selbri. What exactly is the
"obvious meaning" that they have? I mean, the common error is {ti gerku pe
mi} ("This is my dog."), but I don't see how you can consistently get normal
semantics, or even how to get the right meaning there (it's not the same as
{ti pe mi gerku}).
--
Adam Lopresto
http://cec.wustl.edu/~adam/
Good communication is as stimulating as black coffee,
and just as hard to sleep after.
--Anne Morrow Lindbergh
Jorge Llambías
> On Thu, 12 May 2010, Jorge Llambías wrote:
>>
>> http://www.lojban.org/tiki/tiki-index.php?page=zasni+gerna+cenba+vreji
>
> Interesting. Some of those are definitely good, some strike me as going way
> too far (of course), some somewhere in between. But I'm curious about the
> last one, allowing relative clauses to attach to selbri. What exactly is
> the
> "obvious meaning" that they have?
<selbri> <relative-clause> = me lo <selbri> <relative-clause>
The only difference is that the left hand side would have well defined
sumti places beyond x1, and the right hand side doesn't.
> I mean, the common error is {ti gerku pe
> mi} ("This is my dog."), but I don't see how you can consistently get normal
> semantics, or even how to get the right meaning there (it's not the same as
> {ti pe mi gerku}).
mu'o mi'e xorxes
Lindar
a meaning "x1 is called 'randomcmevlastring'" ? Cos if so, that's
actually a good idea. From how you were explaining it before, I
thought you were proposing to say, that ".spagetis." should be a
selbri with a meaning "x1 is spaghetti", which is a horrible idea. I
think somebody had already mentioned it, but it would make a name like
"Little John" or "Big Stu" easier to translate, and we could ascribe
qualities of people to things in tanru.
So, I'm sorry if my previous post came off as an emotional appeal.
If your proposal is to make .cmevlastrings. mean x1 is
called .cmevlastrings., then I like it.
However, that doesn't really solve the initial problem, from what I
can see. Isn't "la .bu,ik." still the company, and not the brand of
car? How would one, under this system, refer to a plate of spaghetti,
or a specific make of vehicle, using this new system versus the
current system?
purpleposeidon
first one. Continuing in that manner: That using cmeselbri would turn
{la .djan. klama} into a sumti makes me sad. It would make all of the
lessons invalid. What if they were forbidden in being part of a tanru?
Perhaps the grammar could look like:
selbri-6 =
tanru-unit [BO # selbri-6]
| [NAhE #] guhek selbri gik selbri-6
| CMEVLA
Jorge Llambías
> So, is the proposal that ".randomcmevlastring." is just a selbri with
> a meaning "x1 is called 'randomcmevlastring'" ? Cos if so, that's
> actually a good idea.
http://www.lojban.org/tiki/tiki-index.php?page=zasni+gerna+cenba+vreji
then no, the proposal is just syntactic, it says nothing about the
meaning of any word.
I do think that "x1 is called '......'" would be the most natural
meaning for most cmevla though.
> From how you were explaining it before, I
> thought you were proposing to say, that ".spagetis." should be a
> selbri with a meaning "x1 is spaghetti", which is a horrible idea.
and stage-four fu'ivla "cidjrspageti" and "spageti"?
> I
> think somebody had already mentioned it, but it would make a name like
> "Little John" or "Big Stu" easier to translate, and we could ascribe
> qualities of people to things in tanru.
> So, I'm sorry if my previous post came off as an emotional appeal.
why you think it's a horrible idea.
> If your proposal is to make .cmevlastrings. mean x1 is
> called .cmevlastrings., then I like it.
>
> However, that doesn't really solve the initial problem, from what I
> can see. Isn't "la .bu,ik." still the company, and not the brand of
> car? How would one, under this system, refer to a plate of spaghetti,
> or a specific make of vehicle, using this new system versus the
> current system?
but if I was forced to use ".spagetis." then:
lo va .spagetis. cu lenku
"Those spaghetti are cold."
lo va karce cu .buik.
"Those cars are Buicks."
mu'o mi'e xorxes
Lindar
> and stage-four fu'ivla "cidjrspageti" and "spageti"?
It breaks the 'phonological fluidity' of the language by putting
irregular word-forms in places where they aren't normally allowed.
While I think the idea is great in theory, it probably isn't so much
in practice. The apparent advantages don't really outweigh the
disadvantages, methinks. While it would allow names in tanru (so I can
have a xorxes-ish laugh or a camgusmis-ish smile), allowing them to
mean anything other than "x1 is called..." would just allow laziness.
Why bother learning the fu'ivla, or even coming up with one when we
can just use cmevla strings and give them any definition we want?
> As I said, I would use "spageti", not ".spagetis." for "spaghetti",
Right, but that's you. I see the possibility for a huge trend to come
of this where people -don't- bother with fu'ivla, and possibly don't
even bother with lujvo because it's easier to use selbri cmevla. Even
so, really what you're doing is forcing every .namestring. to be read
with "me la", which makes me wonder why we can't just keep doing that.
Daniel Brockman
If cmevla break the "phonological fluidity", which presumably is something
that speakers care about (otherwise why would it be a problem?), then isn't
that in itself a strong reason for people to keep coining fu'ivla and lujvo?
On the other hand, if people stop coining fu'ivla and lujvo just because we
allow cmevla to be used as brivla, wouldn't that imply "phonological fluidity"
maybe wasn't that important after all?
I'm certain that this change would increase the usage of cmevla (since it
couldn't possibly decrease it). But I'm equally certain that common words
would not remain as cmevla if people started getting annoyed with them.
By your reasoning, stage-3 fu'ivla is an even more horrible idea, since who
wants the entire language infected by these weird prefixes and r-hyphens?
And what about the fact that a lujvo can be 15+ letters long? That's a pain
in the ass. Who wants to go around saying {mitpavycinglepre}? No, we need
to put a three-rafsi limit on lujvo, or the language will eventually turn into
some monstrous German-like disaster.
Finally, it's quite ironic that you, who so abhor anything that interferes with
the "phonological fluidity", wants us to "just keep saying {me la}" everywhere.
Warrigal
On Tue, May 25, 2010 at 2:40 PM, Lindar <lindar...@yahoo.com> wrote:
>> Why is it a horrible idea? What are the meanings of the stage-three
>> and stage-four fu'ivla "cidjrspageti" and "spageti"?
>
> It breaks the 'phonological fluidity' of the language by putting
> irregular word-forms in places where they aren't normally allowed.
This argument is really vague. What does "irregular word-form" mean?
What does "normally allowed" mean? Are you saying that this change
would make there be too many pauses?
> Why bother learning the fu'ivla, or even coming up with one when we
> can just use cmevla strings and give them any definition we want?
Because if you make up your own words instead of learning the existing
ones, you won't understand anybody, and nobody will understand you.
If you're saying that this change could result in fu'ivla being
abandoned altogether, why is that a bad thing? Specifically, what is
it about cmevla-as-selbri that makes them good for the speaker but bad
for the language?
--
Cantr, a browser-based RPG: http://www.cantr.net/ Create a Lojban
character so you can practice your Lojban!
Lindar
> What does "normally allowed" mean? Are you saying that this change
> would make there be too many pauses?
lojbancanbereadasonebigstringwhichishowit'sheardinspeachsoifwesuddenlyintroducewordsthatCANendinconsonantsthenwefrackuptheabilitytopickoutwordsaccurately.
> If you're saying that this change could result in fu'ivla being
> abandoned altogether, why is that a bad thing? Specifically, what is
> it about cmevla-as-selbri that makes them good for the speaker but bad
> for the language?
The case against long lujvo is moot. We create loanwords from our own
language when they become too long. Also, mitpavycinglepre is a really
stupid example as that isn't an accurate lujvo for 'homosexual', and
I'm quite offended by the idea that we have to have a word like that
in a more enlightened age where gender and sexuality are fluid enough
to warrant "nakni cinse", "fetsi cinse" and so on. I personally am
offended by the use of "mitpavycinglepre".
That aside, my point is that we create fu'ivla of exceptionally long
lujvo in an effort to keep them short. It's like the lojbanic version
of abbreviating things.
We have a standard of phonological rules in Lojban. Everything ends in
a vowel, anything with two or more syllables has penultimate stress,
things have to have correct consonant clusters so they don't break
apart. There is an ordered and well-thought-out structure to every
single word, phrase, emphasis, and every other thing in the language.
The ONLY thing that ends in a consonant is a name, and so names become
easy to pick out due to the fact that we rarely use them in
conversation, and they sound nothing like the rest of the language. On
that mental parse tree, if we used "xorla", I would now have to stop
and question every single cmevla to check whether or not it's being
used as a selbri.
If we restricted cmevla selbri to ONLY being "x1 is named/called...",
then I would agree full-tilt with this proposal. If that doesn't work
with everybody else, then can we at least put -some- restrictions on
them? Perhaps that they're restricted to being single place selbri?
=D We can call my proposal "linla" in honour of xorxes. ((zo'o))
Daniel Brockman
>> What does "normally allowed" mean? Are you saying that this change
>> would make there be too many pauses?
>
> lojbancanbereadasonebigstringwhichishowit'sheardinspeachsoi
> fwesuddenlyintroducewordsthatCANendinconsonantsth
> enwefrackuptheabilitytopickoutwordsaccurately.
Sigh, no. One word: dotside.
>> If you're saying that this change could result in fu'ivla being
>> abandoned altogether, why is that a bad thing? Specifically, what is
>> it about cmevla-as-selbri that makes them good for the speaker but bad
>> for the language?
>
> The case against long lujvo is moot. We create loanwords from our own
> language when they become too long. Also, mitpavycinglepre is a really
> stupid example as that isn't an accurate lujvo for 'homosexual',
There's no such thing as an "accurate lujvo". But the fact of the matter
is that {mitpavycinglepre} is a well-established word. And my point is
exactly that at some point people thought it was getting too inconvenient.
So we shortened it. Rationalized it. Came up with a slicker word for it.
http://jbovlaste.lojban.org/dict/mitcinse
> and
> I'm quite offended by the idea that we have to have a word like that
> in a more enlightened age where gender and sexuality are fluid enough
> to warrant "nakni cinse", "fetsi cinse" and so on. I personally am
> offended by the use of "mitpavycinglepre".
Well, thank you for proving my point. Inconvenient words for important
concepts are always undesirable, whether they be cmevla, lujvo, or fu'ivla.
> That aside, my point is that we create fu'ivla of exceptionally long
> lujvo in an effort to keep them short. It's like the lojbanic version
> of abbreviating things.
Exactly the same can be done with cmevla.
> We have a standard of phonological rules in Lojban. Everything ends in
> a vowel,
Not cmevla, and they are part of the language.
> anything with two or more syllables has penultimate stress,
Not necessarily cmavo, and not necessarily cmevla.
> things have to have correct consonant clusters so they don't break
> apart.
That's one requirement for gismu, lujvo and fu'ivla, yes. There are others for
those classes and there are yet others for other classes. For example, cmavo
can not have any consonant clusters, and cmevla must end in consonants.
> There is an ordered and well-thought-out structure to every
> single word, phrase, emphasis, and every other thing in the language.
Well... heh. Let's agree to disagree on that.
> The ONLY thing that ends in a consonant is a name, and so names become
> easy to pick out due to the fact that we rarely use them in
> conversation, and they sound nothing like the rest of the language.
You're exaggerating how different they sound. They only use Lojban phonemes,
and most of them sound exactly like a brivla with a consonant at the end.
By your reasoning, every kind of word in Lojban sounds nothing like the
rest of the language. Yet that's a feature. It's a good thing. (In English,
all the words sound just like each other.)
The rules for fu'ivla are more relaxed than for other brivla, but I don't see
you proposing to ban fu'ivla because they can sound a bit different.
> On that mental parse tree, if we used "xorla", I would now have to
> stop and question every single cmevla to check whether or not it's
> being used as a selbri.
You don't have to "stop and question" every cmevla. That's not how
language works. Do you have to stop and question every word to
check whether or not it's being quoted? Do you have to stop and
question every brivla to see whether it's being used as a name?
No, you don't, because that's not how language works.
I guess you would prefer if we didn't have selbri names?
> If we restricted cmevla selbri to ONLY being "x1 is named/called...",
> then I would agree full-tilt with this proposal. If that doesn't work
> with everybody else, then can we at least put -some- restrictions on
> them? Perhaps that they're restricted to being single place selbri?
>
> =D We can call my proposal "linla" in honour of xorxes. ((zo'o))
(No comment.)
Adam D. Lopresto
>> This argument is really vague. What does "irregular word-form" mean?
>> What does "normally allowed" mean? Are you saying that this change
>> would make there be too many pauses?
>
> lojbancanbereadasonebigstringwhichishowit'sheardinspeachsoifwesuddenlyintroducewordsthatCANendinconsonantsthenwefrackuptheabilitytopickoutwordsaccurately.
How about learning the current Lojban morphology before attacking "changes"
that don't actually change it one iota? Lojban can be unambiguously written
without spaces only if stress and all mandatory stops are explicitly marked.
The morphology algorithm breaks words apart and determines what words are of
what type before the grammar has any say in anything. Changing where cmevla
are *grammatical* does not affect the *morphology*. The grammar is only
involved after the phoneme or text stream is broken into words by the
morphology.
>> If you're saying that this change could result in fu'ivla being
>> abandoned altogether, why is that a bad thing? Specifically, what is
>> it about cmevla-as-selbri that makes them good for the speaker but bad
>> for the language?
>
> The case against long lujvo is moot. We create loanwords from our own
> language when they become too long. Also, mitpavycinglepre is a really
> stupid example as that isn't an accurate lujvo for 'homosexual', and
> I'm quite offended by the idea that we have to have a word like that
> in a more enlightened age where gender and sexuality are fluid enough
> to warrant "nakni cinse", "fetsi cinse" and so on. I personally am
> offended by the use of "mitpavycinglepre".
And I am offended that after we come up with a huge set of words to precisely
cover every part of the Venn diagram of possible human sexuality, you come in
and express offense at the existence of a word for a particular part of that.
> That aside, my point is that we create fu'ivla of exceptionally long
> lujvo in an effort to keep them short. It's like the lojbanic version
> of abbreviating things.
That's really not done much at all (I first heard of it a few months ago), but
it is a somewhat nifty approach.
> We have a standard of phonological rules in Lojban. Everything ends in
> a vowel, anything with two or more syllables has penultimate stress,
> things have to have correct consonant clusters so they don't break
> apart. There is an ordered and well-thought-out structure to every
> single word, phrase, emphasis, and every other thing in the language.
> The ONLY thing that ends in a consonant is a name, and so names become
> easy to pick out due to the fact that we rarely use them in
> conversation, and they sound nothing like the rest of the language. On
> that mental parse tree, if we used "xorla", I would now have to stop
> and question every single cmevla to check whether or not it's being
> used as a selbri.
Again, this doesn't change the morphology at all. cmevla are defined and
identified exactly as before, they just also are grammatically valid in other
places.
> If we restricted cmevla selbri to ONLY being "x1 is named/called...",
> then I would agree full-tilt with this proposal. If that doesn't work
> with everybody else, then can we at least put -some- restrictions on
> them? Perhaps that they're restricted to being single place selbri?
>
> =D We can call my proposal "linla" in honour of xorxes. ((zo'o))
>
>
--
Adam Lopresto
http://cec.wustl.edu/~adam/
And entropy continued to increase.
tijlan
The ONLY thing that ends in a consonant is a name, and so names become
easy to pick out due to the fact that we rarely use them in
conversation, and they sound nothing like the rest of the language.
{la cribe} is a name (cmene) too, and it sounds like {lo cribe}.
The word class that ends with a consonant is cmevla. And this class has experimental usages for other than what we usually consider names. {sa'ei}, for instance, marks a cmevla as onomatopoeia.
On
that mental parse tree, if we used "xorla", I would now have to stop
and question every single cmevla to check whether or not it's being
used as a selbri.
(a) lo me la spagetis
(b) lo spagetis
With (a), you analyse {spagetis}'s relation to {la} to {me} to {lo}, resulting in the interpretation "that which is called {spagetis}".
With (b), you analyse {spagetis}'s relation to {lo}, resulting in the interpretation "that which is called {spagetis}".
In either case, you 'check' the syntax of {... spagetis}. Which syntax is simpler? In which case do you have to 'stop and question' less? (b). That's one thing the proposal would achieve.
John E Clifford
<<> That aside, my point is that we create fu'ivla of exceptionally long
> lujvo in an effort to keep them short. It's like the
lojbanic version
> of abbreviating things.
That's really
not done much at all (I first heard of it a few months ago), but
it
is a somewhat nifty approach.>>
Have y'all forgotten Zipf's law, which, while not prescriptive, is certainly predictive? If you don't like this trick, try another, but someone is going to do the cutting sometime for sure (assuming the language lives).
Lindar
Some of you are taking this a little too personally.
We're not all linguists and maths professors in the community, but I
think it's still fair to have a word in what happens to Lojban. I'm
sorry that I'm not the most educated of the group and that I don't
know all the correct terminology. I'm expressing my opinion as the
layman user, not the programmer or the logician. Mr. Lopresto, I'm not
-attacking- anything, I'm debating my side of the argument, and I
don't appreciate you talking down to me when I've been with the
community for almost a year now. I know how to speak Lojban just fine,
and I don't appreciate you insinuating that I don't.
tijlan, your argument actually makes a lot of sense. I suppose I was
taking a somewhat conservative approach to things, which I'm sure is
only normal when we're making big changes to the grammar. xorlo was
probably charged by the horns as well, wasn't it? Well, if I'm
completely outnumbered, I might as well roll with the punches and see
where this goes.
So, if everybody is adamant on changing how cmevla work in grammar,
then why don't we actually test it out? Cement things out firmly?
Again, I'm not the sharpest bulb in the tool shed, so please explain a
few things for me.
What is the semantic difference between lo .lindar. and la .lindar. ?
Can cmevla have multiple places?
Is {me} being removed from the grammar due to it not being necessary
after this change?
Could this possibly be used as a system for creating "slang" words?
Could somebody please write a few sample bridi using this new rule?
Lindar
Mr. Brockman has pointed out that I come off as "inflammatory,
ignorant, and loud-mouthed".
I'm really not trying to flame, nor am I trying to come off as
ignorant.
Like I said in my previous post, I'm not the brightest speaker around
considering most of you have a history of higher education and I
don't.
I tend to come off as rude or abbrasive to pretty much everybody in
any situation, and I apologise if it seems like I'm flaming.
I do realise that I'm the only person to oppose this idea, and that I
held a position VERY much against it, but I'm not trying to make
emotional appeals, come off as loud-mouthed, or flame anybody. Most of
the language that flies around these boards is completely over my
head, so when I try to pitch in my two cents, I seem to come off as
loud and retarded instead of just pitching in my counter-argument.
So, long story short, I'm sorry that I come off as a loud asshole. I'm
really just trying to have a pleasant debate.
Daniel Brockman
Same as between {lo broca} and {la broca}.
> Can cmevla have multiple places?
I don't see why not?
> Is {me} being removed from the grammar due to it not being necessary
> after this change?
No, definitely not.
> Could this possibly be used as a system for creating "slang" words?
Sure, why not? That's a good thing about cmevla: they're easy to make.
> Could somebody please write a few sample bridi using this new rule?
I'll pass.
Lindar
of characters) and lo .lindar. is "Thing(s) named Lindar." ? That's
nifty.
> > Can cmevla have multiple places?
>
> I don't see why not?
??
> > Is {me} being removed from the grammar due to it not being necessary
> > after this change?
>
> No, definitely not.
> > Could this possibly be used as a system for creating "slang" words?
>
> Sure, why not? That's a good thing about cmevla: they're easy to make.
Well, I'm convinced. It seems like a decent idea as long as it doesn't
get abused. Do we just start using it now, is there some kind of
ceremony we have to have first, or... what now? Does everybody agree
on exactly how it'll work?
Luke Bergen
>>> after this change?
>>
>> No, definitely not.
Luke Bergen
Daniel Brockman
> of characters) and lo .lindar. is "Thing(s) named Lindar." ? That's
> nifty.
Well, {la broca} is "Broca", but {lo broca} is just undefined.
The same would be true of {lo lindar} (although xorxes has suggested that the
most useful definition of at least many existing cmevla would be as you say).
I wouldn't focus so much on that "default meaning" stuff, as it makes the
whole concept more difficult to grasp.
>> > Can cmevla have multiple places?
>>
>> I don't see why not?
>
> .spagetis. = x1 is spaghetti with sauce x2
> ??
Sure. (Although "x1 is spaghetti of type x2" would be more likely.)
>> > Is {me} being removed from the grammar due to it not being necessary
>> > after this change?
>>
>> No, definitely not.
>
> Then under what condition would it still be used?
Well, that's a pretty broad question. Whenever you want to convert a sumti
into a selbri. i lo me zo me selbri cu plixau
> Well, I'm convinced. It seems like a decent idea as long as it doesn't
> get abused. Do we just start using it now, is there some kind of
> ceremony we have to have first, or... what now?
You can smoke whatever weird experimental stuff you can fit in your pipe,
but don't expect something like this to just suddenly change overnight.
Most Lojbanists probably haven't even read this, much less commented.
But people experiment with all kinds of stuff all the time (e.g., on IRC).
> Does everybody agree on exactly how it'll work?
I suggest you focus on the essence of it, and the essence of it is extremely
simple: We treat cmevla exactly as we treat brivla. Read that again, slowly:
We treat cmevla exactly as we treat brivla.
You know how brivla work, right? What {lo broda} and {la broda} mean?
Good. Then you know how cmevla work under xorla (i.e., exactly the same).
Lindar
> simple: We treat cmevla exactly as we treat brivla. Read that again, slowly:
>
> We treat cmevla exactly as we treat brivla.
>
> You know how brivla work, right? What {lo broda} and {la broda} mean?
> Good. Then you know how cmevla work under xorla (i.e., exactly the same).
It pleases me when people find their Lindar-filters.
((I lol'd @ talking to me like I'm a simpleton. <3 ))
We can't come up with a different name? I'm pretty sure that, aside
from brodV, we're not supposed to have gismu that are the same
excepting the final vowel. xorcme, perhaps? It's more about cmevla
than "la" itself. xorlo, xorcme, what's next?
I have a very valid question for all involved:
Will we be documenting definitions for cmevla now that they are valid
brivla, or should it be left up to context?
Luke Bergen
Daniel Brockman
> ((I lol'd @ talking to me like I'm a simpleton. <3 ))
That wasn't meant as a comment about you so much as a comment about the
simplicity of the basic proposal, though.
> We can't come up with a different name? I'm pretty sure that, aside
> from brodV, we're not supposed to have gismu that are the same
> excepting the final vowel.
These are just names. They're not gismu in the sense of "root words",
they're gismu only in the purely morphological sense.
> I have a very valid question for all involved:
>
> Will we be documenting definitions for cmevla now that they are valid
> brivla, or should it be left up to context?
You're being a little too forward with this ("now that they are valid brivla").
What we should do is just to continue experimenting and discussing.
Lindar
them?
Luke, was that you using xorcme? (it doesn't change la, so why call it
xorla?) It seemed to be a good example.
.i mi gleki be tu'a la xorcme be'o .lindar. .u'isai
So what is "loi .xorxes." then? .i .o'i mu xagji .lindar. cu zvati
le purdi
Oh! How does COI default now? It used to be that "coi .luk." was "coi
la .luk." and "coi xalbo" was always "coi le xalbo", so should it
always default to le now?
John E Clifford
(BTW names are not gismu in precisely the morphological sense (don't end in vowels for starters) and they are root words in the sense of not having an etymology within Lojban -- what that notion means in the case of gismu.)
--
Dag Odenhall
> all. In any case, in "la xorxes" the meaning of the predicate is as
> irrelevant as the meaning of "donri" in "la donri". "la" in any case
> removes the meaning from the word that follows.
{la donri} means "That named Day". {la} does not remove the meaning.
mi'e dag — donri in Swedish.
tijlan
.spagetis. = x1 is spaghetti with sauce x2
??
Generally I'm not very fond of being quick to stow places into a brivla that could otherwise be expressed with sumtcita or some other means. The fact that you and Daniel came up with different sumti for the x2 already shows the limitaion of this place's predictability. And the more we try to expand the place structure, the more we would risk to make the brivla's definition arbitrary and less neutral.
"with" can be expressed with {joi}, {jo'u}, {jo'e}, {je}, etc., and "of type" with {le'a}, {kai}, {ra'i}, etc. I suggest we start with as minimal a place structure as possible for a new word and then let usage call for addition of places if any. The place structure should be convenient, not cumbersome.
tijlan
I would say {la donri} means "That named 'donri'". {la} removes the place structure from {donri}. What's left then is just a string of letters.
Daniel Brockman
meaning from whatever comes after or if it "retains" the meaning.
In fact, there are not just two levels to this. Already, {la irc} is vague
about whether spelling or pronunciation is important. Already, {la donri}
can be translated literally into Swedish but not English. Other names are
relatively "transparent": {la dansu be fi'o kansa lo labno} can be translated
into any language with an acceptable result. Some names are strings of rafsi.
Some are brivla with dummy consonants tacked on; some are brivla with vowels
chopped off. Some, like {daniel}, are almost completely intact and can be both
phonologically and letter-wise translated into any language with an acceptable
result; others have been mangled to fit Lojban phonology and morphology and
must be "unmangled" when translated into another language.
There are many kinds of names in Lojban, and that's okay.
So it makes sense for Lojban to simply not specify one way or the other.
We should allow both "meaningful" and "meaningless" names; both names
where spelling is important and names where pronunciation is important.
In almost all cases, this will be totally irrelevant. In most other cases,
context will be sufficient. For the last few cases, we could come up with a
system for indicating the level of meaning. But {la} by itself cannot do it.
tijlan
Obviously we're never going to agree on whether {la} "removes" the
meaning from whatever comes after or if it "retains" the meaning.
I guess we actually can agree on one hand 1) that, in {la donri}, {la} may remove the place structure from {donri} so that {la donri} does not inherently/primarily mean "daytime" as in {lo donri}, and on the other hand 2) that those who are familiar with gismu {donri}'s place structure, may superficially associate {la donri} with the meaning of {lo donri}, which is "daytime", or with something else. In other words, {la donri} may have no inherent/primary meaning other than "that named 'donri'" but may have superficial/secondary meanings such as English "daytime" or Swedish "dag" or whatever people can associate with the string of letters {donri}.
Already, {la irc} is vague about whether spelling or pronunciation is important.
Well, the important thing is whether people can associate the right, intended meaning more with a certain string of letters than with other strings. If to people {irc} is generally more associative of "internet relay chat" (assuming that's the intended meaning) than {iburyc} is, then we should consider {irc} better than {iburyc} as the cmevla for "internet relay chat".
Oleksii Melnyk
--
mu'o mi'e lex
Lindar
The benfit seems to be that one can now use cmevla in tanru, quick "in
my native language" words can be used, and we can use slang... and
probably a bunch of other things. tijlan named a bunch of them.
Pierre Abbat
I do not think the change is worth it. Lojban has two open parts of speech
(cmevla, brivla) to English's four (noun, verb, adjective, adverb). (Yes I
know there are languages with two kinds of adjective, or no distinction
between verb or adjective, etc. All my native languages have those four.) It
is possible to use "la" with a brivla, or "me" with a cmene, or make a brivla
corresponding to a cmevla (e.g. sfa'ani/isfa'an). But generally the two parts
of speech serve distinct functions and should be kept separate.
Pierre
--
La sal en el mar es más que en la sangre.
Le sel dans la mer est plus que dans le sang.
Adam D. Lopresto
> la .luk." and "coi xalbo" was always "coi le xalbo", so should it
> always default to le now?
Might make more sense to make {la} always the default. Simplifies the rules,
and puts those of us with selbri names on an equal footing. But I can see
arguments either way.
mu'o mi'e la xalbo do'u mi'e xalbo je'u
--
Adam Lopresto
http://cec.wustl.edu/~adam/
One who cannot cast away a treasure at need is in fetters.
--Aragorn
Lindar
My ideas:
1. COI implies {la} ({.i coi xalbo} is {.i coi la xalbo} and never {.i
coi le xalbo}.
2. {la} is merged with LE and is terminated by {ku}.
3. lo --------- le -------- la
My idea of how {la} works now is based on a scale where {lo} is the
other end and {le} is the midpoint. {lo broda} must broda, {le broda}
doesn't necessarily have to broda, but that's what I'm calling it, and
{la broda} doesn't have to broda, but that's the proper name for it.
To address the "la erases all meaning" vs. "la keeps the meaning"
thing, my interpretation is that the meaning of the x1 is the name,
just like how it works for lo/le. Context can decide how names are
translated ("Hi, Day!" vs. "Hi, Donri!" vs. "Hi, Dag!"), but the
default is always the meaning (Hi, Day!).
4. Use of cmevla as brivla should be restricted to brands of things
({la .opel.} is the brand, {lo .opel.} is a car of that brand.) or
similar situations in which names of things are based upon names, "x1
is called/named..." definitions, simple and recognisable slang which
is appropriate for the conversation (perhaps {.i mi .lol.} or
something silly for "I lol'd." or {.i mi .dogezas.} for "I (am)
bow(ing)."), or quick borrow-words which should be corrected by others
({.i mi .going.} would be understood quickly by English speakers and
could be corrected to {cliva} or {klama}.).
My strong opinion is that this is very fun and perhaps useful, but
should not be a substitute for good fu'ivla.
Thoughts?
Jorge Llambías
>
> 1. COI implies {la} ({.i coi xalbo} is {.i coi la xalbo} and never {.i
> coi le xalbo}.
I often use "doi pendo", "doi dirba", "doi patfu", etc with no "la" implied.
I think "coi xalbo" could be "coi la xalbo" or "coi lo xalbo"
depending on context, no need to rule either out. You can always use
the gadri explicitly if confusion is possible or likely.
mu'o mi'e xorxes
Lindar
> I think "coi xalbo" could be "coi la xalbo" or "coi lo xalbo"
> depending on context, no need to rule either out. You can always use
> the gadri explicitly if confusion is possible or likely.
trying to avoid. Isn't Lojban supposed to be "completely regular and
without exception"?
Either way, I do not at all like the idea of leaving it up to context.
It needs to be one or the other. Having both be {la} is more
consistent, having selbri imply {le} makes more sense.
Luke Bergen
Daniel Brockman
> depending on context, no need to rule either out. You can always use
> the gadri explicitly if confusion is possible or likely.
+1
Obvious, simple, and good. I use {coi pendo}, etc., a lot as well.
It's always bugged me that {coi la xalbo} needs {la} but {coi daniel} doesn't.
Lindar
of logic."
Well, it -could- be la... or lo... or maybe even le... what do you
think he meant? I don't know, there isn't a regular rule that we
follow.
>_>
I really, truly, honestly believe that your idea is a bad one.
That's not regular grammar. It -is- an exception. It is ambiguous in
meaning. A computer could not parse it.
So your choice isn't an available option, as I see it.
In my opinion, you have three options:
1. It implies {lo}, because that is the generic gadri.
2. It implies {le}, because that has been the standard, and it makes
sense for greetings such as {.i coi pendo} more so than option 1.
3. It implies {la}, because we shouldn't have an inconsistency between
cmevla names and selbri names, so we can say {.i coi xalbo}.
Don't know what else to tell you. =\
Minimiscience
> I really, truly, honestly believe that your idea is a bad one.
> That's not regular grammar. It -is- an exception. It is ambiguous in
> meaning. A computer could not parse it.
Although it doesn't really affect your main point, I feel like I should point
out that interpretation of vocatives is entirely *semantic*, not
syntactic/grammatical. It *is* regular grammar; it's the same grammar as
before, except that the rules for <COI> <selbri> and <COI> <CMENE>+ have now
been merged. A computer could still parse vocatives, and the resulting syntax
tree would be the same. The {lo}/{le}/{la} ambiguity only becomes apparent if
a computer or human attempts to assign a *meaning* to a vocative, which is at
least one level of abstraction higher than the grammar. Moreover, Lojban
already contains known (and unavoidable) semantic ambiguities, such as {tanru}
(which are ambiguous almost by definition), determining whether an instance of
"{lo broda}" under xorlo truly {broda}s, and, say, referring to "{la .djan.}"
in a bathroom full of male native Anglophones accompanied by ladies of
negotiable affection (I apologize for the mental image). The question (well,
*a* question) is whether this new potential ambiguity of vocatives does not
exceed these other uncertainties.
mu'omi'e .kamymecraijun.
--
jicmu traji zifre fa loi remna lonu senpi
Lindar
(dance-r(s)) :] mi I, me)4 [is, does] <<5dansu danc-ing>>5]3 .i
`.' (6(7coi greetings lindar. [NAME])7)6 [8(9[nelci1 (like-r(s)) :]
do you)9 [is, does] <<10nelci lik-ing>>10]8
.i /`.'/ (^coi /greetings/ la / / xalbo /flippant thing(s)/ KU / /
do'u / /^) [([dansu1 (dance-r(s)):] mi /I, me/) CU /is/does/ <<dansu /
danc-ing/>> VAU / /]
.i /`.'/ (^coi /greetings/ mi /I, me/ DO'U / /^) [<<xalbo /being
flippant/>> VAU / /]
.i /`.'/ (^coi /greetings/ xalbo /flippant thing(s)/ DO'U / /^)
.i /`.'/ (^coi /greetings/ lo /any/some/ xalbo /flippant thing(s)/
KU / / DO'U / /^)
So you can see here that there is a difference in parsing. It reads
{xalbo} with an implied gadri, though it doesn't seem to specify
which.
I don't think we can just say "whatever" and leave it up to context. A
parser needs to know what the default is if one is not specified.
Daniel Brockman
> A parser needs to know what the default is if one is not specified.
No, a parser definitely does not need to know things like that.
Lindar
Daniel Brockman
Minimiscience
> So you can see here that there is a difference in parsing. It reads
> {xalbo} with an implied gadri, though it doesn't seem to specify
> which.
Jbofi'e is both a parser and a glosser, and it is fulfilling the latter role
when it gives meanings for phrases. Parsing consists solely of determining the
grammatical structure of a text (i.e., what parts are {selbri}, {sumti},
numbers, etc.) and does *not* involve determining or assigning meaning to a
text, which is done *after* parsing.
> I don't think we can just say "whatever" and leave it up to context. A
> parser needs to know what the default is if one is not specified.
No, a glosser or translator needs to know which {gadri} to use. A program that
is a parser and nothing more (regardless of how useless it may seem) only needs
to know the {selma'o} & pseudo-{selma'o} of its input and, of course, the rules
of Lojban syntax. All it generates from this is a syntax tree or equivalent
structure, which, by itself, conveys no useful information about the meaning of
a text.
mu'omi'e .kamymecraijun.
--
bu'u la .lojbangug. lo bangu cu daspo do
Lindar
guess it isn't necessary. Have we hammered out all of the details?
Daniel Brockman
always imply a particular gadri (namely, {le}). This is a change.
Lindar
> Huh? If you're still talking about COI BRIVLA, then yes, that did
> always imply a particular gadri (namely, {le}). This is a change.
(under xorla).
So, with that settled, are there any other issues which have not been
fixed?