"Bob LeChevalier, President and Founder - LLG" <lojbab@lojban.org>: Sep 09 01:30PM -0400
On 9/7/2014 5:45 PM, Alex Burka wrote:
> Dustin for doing so. I do think it will take rather longer than we
> perhaps thought when this thread began.
> I still think that the software-development model is the way to go.
We already used a software development model. That is where the
"baseline" concept came from. The problem is that software needs to be
properly documented, and normally you don't go changing it without
documenting what you already have. The other problem is that the
"development" is supposed to already be done. Long done. And for a lot
of people, the idea that they might have to go to Microsoft Lojban 8.0
from 7.0 is enough to make them throw up their hands in disgust and turn
away from the language. They might accept small tweaks to fix bugs in
"Lojban XP", but they don't want to relearn anything.
> Among other things, it lets wild experimentation coexist safely with the
> carefully curated “master branch” — I mean, this is already happening in
> the community, but it’s a mess and a hodgepodge of projects,
That is what experimental language is like, and should be like. Someone
else posted today on Buckminster Fuller's usages which are a form of
experimental English. For anyone who did not know about those usages,
they would be "a mess and a hodgepodge"
> exactly
> because there’s no accepted process, and it doesn’t help that people who
> propose one tend to get shouted down.
Because we still haven't documented the baseline "languageware".
> Languages change when they are actively used.
Not generally by fiat. Look how many approaches exist for English third
person pronouns which are gender-neutral. All are "changes" to English,
but none have really caught on.
> this will just result in everyone leaving (which seems to put me in
> fundamental disagreement with those who think that if we change
> anything, then everyone will leave).
This project is some 60 years old and we have a lot of history of people
explicitly leaving because of changes imposed from on-high. More
importantly, we have the history of dozens if not hundreds of conlangs
whose usage has not spread because people wouldn't stop fiddling with
the language design. Those that have survived and spread have stopped
being developed, and are simply USED. Adding computer terminology to
Esperanto as it becomes needed isn't a "language change".
Systematically changing the endings of gismu to fit some new schema is a
drastic change that would likely cause many to leave or schism.
People might have tolerated running across some new word on an IRC
channel and looking it up; we deal with learning new vocabulary all the
time in natural language. But they don't like someone telling them that
the old way to do something is wrong, and there is a new and better way.
> But I also think there’s a middle
> ground. We can allow some changes (with strict review, li’a sai),
> without destroying the language.
At some point, you have to stop allowing changes EXCEPT by *natural*
language processes (which aren't so much "reviewed" as "documented after
the fact".
> the one about documentation of experimental grammar being hard to find):
> in my mind, there is no way Lojban can be "considered DONE as an
> engineering effort”.
Then we are fundamentally at odds. It MUST be "done" at some point.
Engineering must stop, and we move to usage.
Certainly the publication of the CLL and everything
> considered “almost done”, but the mere existence of the BPFK and the ZG
> are confirmation that the final word (whatever that means when applied
> to a living language, see above) has not been said.
The intent of the BPFK was to finish the documentation of the language,
making corrections to CLL as necessary. The ZG acceptance of xorlo as a
major bug-fix because the community felt that the original design COULD
NOT be properly documented.
> to deny that changes have happened and become accepted by large portions
> of the community since the publication of the CLL: the BPFK morphology.
> xorlo was adopted into the ZG. Nearly everyone uses dotside for names.
"Dotside" is not really an engineering change, but rather using one rule
(the dot-pause, which is allowed between any two words) to allow another
rule to be ignored. Indeed it has been accepted because the listener
likely won't even notice that it is being used. And likely a dotside
user will not have a problem with the speech of a non-dotside speaker,
provided that the latter follows the pre-dotside rules.
xorlo was proposed and discussed within the BPFK as a solution to
problems in documenting the gadri. If the documentation had been
finished earlier, there would have been no need for the ZG. (Probably
true for dotside as well.)
> Modern {ka} with {ce’u}.
should be documented as part of the baseline. ce'u predates said baseline.
The experimental gismu {kibro}
never heard of it.
and cmavo
> {di’ai}. vu’o po’onai.
vu'o and po'onai should both be part of the baseline (not that I
remember what the latter means; I am sure it was discussed back in the
90s). I have no idea what di'ai is. That is the problem with
experimental usages. They aren't documented, and people like me would
have no idea what to do with the word if we run across it in text. Some
might try to figure it out like Jabberwocky (but the coinages in
Jabberwocky pretty much all correspond to brivla and not cmavo).
I am least interested in getting bogged down
> in legalese and bylaws if such things turn out to obstruct the use of
> Lojban and encourage it to stagnate (which is what’s arguably happened
> in the last decade).
It hasn't prevented anyone from using the language as they see fit so
far. You won't hear me complaining because you use "di'ai". I just
won't understand you. If we're on IRC I can ask you what it means, but
if it is in text, I will have no clue.
The "stagnation" seems to me that people want to document and approve
new stuff without documenting the older stuff. I don't see any easy way
to resolve this impasse, but if there is, it would require someone to
write CLL sections covering new material to replace or add to the
existing text. If someone were doing this and collecting it somewhere,
AND the original baseline documentation was being finished, we could
deal with it. We do need a new edition of CLL, probably next year, and
the argument becomes what to put into it. Likely the decision will
effectively be made by those that do the writing.
> That being said, yes, I’d like to join the LLG.
Noted.
> that there’s a way to get there, but I don’t claim to have all the
> answers of course. I am glad that there will be a meeting soon, because
> that’s at least potential progress.
There is a meeting every year. But there hasn't been progress every
year because most years, no one has any issues to bring up. The coming
meeting looks like it will resemble those of 2002-2003 when Robin and
others stepped in and took over. I stepped down as president, only
resuming the job when Matt resigned in 2010 (and I'll probably step down
again gladly if I get the feeling that someone is capable and willing to
fulfill the organizational responsibilities of the President. And some
day someone will have to take over the Virginia representative spot,
which exists solely for legal reasons).
lojbab
|
And Rosta <and.rosta@gmail.com>: Sep 09 09:50PM +0100
Bob LeChevalier, President and Founder - LLG, On 09/09/2014 18:30:
> throw up their hands in disgust and turn away from the language. They
> might accept small tweaks to fix bugs in "Lojban XP", but they don't
> want to relearn anything.
It seems to me you're setting up a largely false dichotomy. Most of what remains to be done is to complete the design where it is incomplete. So the choice is whether to do that explicitly or leave it to usage. Not much relearning entailed by that. There is the additional choice of whether to make simplifications that require a handful of individuals to do some relearning now, for the benefit of making the task for all future learners much simpler, but that is a separate debate.
> the time in natural language. But they don't like someone telling
> them that the old way to do something is wrong, and there is a new
> and better way.
Are there still many that feel thus? I wonder if it's a myth that gets perpetuated because you propagate it so insistently.
In old usage, "le" was standardly not used in a baseline-compliant way; cf how "le nu", "le ka", "le du'u" used to be default in usage. In old and new usage (for new usage, I'm relying on Selpa'i's observation), logical scope of syntactic clausemates is generally ambiguous. How many people are going to want to preserve old ways that aren't baseline-compliant or are rampantly logically ambiguous?
> This project is some 60 years old and we have a lot of history of
> people explicitly leaving because of changes imposed from on-high.
Not in the history of Lojban proper, of course, because changes haven't been imposed from on-high. So all the folk leaving for the last 27 years have been leaving for other reasons; disgruntlement at the unfinished design and the political sclerosis that prevents its completion must be the major reason why people leave Lojban, out of all reasons that have to do with some sort of disaffection with Lojban.
> More importantly, we have the history of dozens if not hundreds of
> conlangs whose usage has not spread because people wouldn't stop
> fiddling with the language design.
I think you'd be hard-pressed to identify these dozens if not hundreds of conlangs whose usage would have spread if people had stopped fiddling with the language design.
>> engineering effort”.
> Then we are fundamentally at odds. It MUST be "done" at some point.
> Engineering must stop, and we move to usage.
Especially in the case of a language like Lojban, one expects that there will always be a strong strand of prescriptivism, in areas where usage deviates from the official design or from logic. Prescriptivism is a form of engineering. It has a bad name in the domain of natlangs, mostly because actual prescriptivists tend to be foolish, but to people attracted to Lojban by its explicit definition and ostensible logical basis, rational prescriptivism is likely to be welcome.
> vu'o and po'onai should both be part of the baseline (not that I
> remember what the latter means; I am sure it was discussed back in
> the 90s).
It was discussed back in the 90s, but is it in CLL? I can't find a way to search CLL online (-- there must be one, but googling doesn't bring it up). It's not in CLL Ch 13 where po'o is introduced.
> I have no idea what di'ai is. That is the problem with
> experimental usages. They aren't documented, and people like me would
> have no idea what to do with the word if we run across it in text.
I went to the humungous effort of looking kibro and di'ai up in jbovlaste. To find jbovlaste, one googles "jbovlaste". Or, even quicker, google "jbovlaste kibro" and you get the answer in one step. For users of handheld devices, Gleki has made an android jbovlaste app -- it's excellent! And it's dead easy to use even for those of us who are weary at having to learn new technology.
--And.
|
Alex Burka <durka42@gmail.com>: Sep 09 05:43PM -0400
Thanks, And, I agree with nearly all of what you said, though I may write more later. One quick clarification, I really didn’t mean {vu’o po’onai} as another example of experimentalism — just meant to say “etc”. Sorry for the sloppy jboglish.
On Tuesday, September 9, 2014 at 4:50 PM, And Rosta wrote:
|
Gleki Arxokuna <gleki.is.my.name@gmail.com>: Sep 10 09:51AM +0400
>> and better way.
> Are there still many that feel thus? I wonder if it's a myth that gets
> perpetuated because you propagate it so insistently.
No, this is true.
> of syntactic clausemates is generally ambiguous. How many people are going
> to want to preserve old ways that aren't baseline-compliant or are
> rampantly logically ambiguous?
The task is to adapt theory to facts, i.e. usage, not adapt reality to
facts provided this doesn't lead to syntactic ambiguity which is a defining
feature of lojban.
> I think you'd be hard-pressed to identify these dozens if not hundreds of
> conlangs whose usage would have spread if people had stopped fiddling with
> the language design.
He won't. I can confirm his words.
I've got a lot of people from Russian group who immediately stopped
learning Lojban when they learnt that CLL was no longer valid.
With regret I have to acknowledge that Lojbab's task of creating a stable
language failed when the not well thought out change called "xorlo"
invalidated the refgram. May be it's still not too late to go back to
pre-xorlo
and formalize quantification problems so that they match CLL example as
close as possible.
Another example is Quenya. When new Tolkien's stuff was published the
community shrinked fivefold.
Yet another example is Loglan.
That's why any changes to basic gismu, to common usage is a way to the
final destruction of the language as it happened to other conlangs.
> To find jbovlaste, one googles "jbovlaste". Or, even quicker, google
> "jbovlaste kibro" and you get the answer in one step. For users of handheld
> devices, Gleki has made an android jbovlaste app -- it's excellent!
Huh?
|
And Rosta <and.rosta@gmail.com>: Sep 10 02:16PM +0100
Gleki Arxokuna, On 10/09/2014 06:51:
> The task is to adapt theory to facts, i.e. usage, not adapt reality
> to facts provided this doesn't lead to syntactic ambiguity which is a
> defining feature of lojban.
There are three forces that potentially shape and define what is to be deemed correct:
1. usage
2. official codification
3. logic (mapping between phonological and logical forms), consistency, regularity, unambiguity, integrity
& possibly a fourth:
4. unofficial consensus of opinion (or of influential opinion)
(4) is important for English, maybe not for Lojban.
All can conflict. Which trumps which? For me it's 3>2>1. For Bob I hope (because it's a position I can respect) it's 1>2>3. What do you think it is?
At the time I ceased active involvement with Lojban I had come to the view that that the community was wedded to 1>2>3 or 2>1>3 with immutable 2, but now I see that there are currents of opinion -- much stronger than ever in my time -- unwilling to accept either of those. Surely the only foreseeable outcomes are that the ultraconservative camp withers or that there is schism.
> He won't. I can confirm his words. I've got a lot of people from
> Russian group who immediately stopped learning Lojban when they
> learnt that CLL was no longer valid.
Bob was talking about conlangs not conlangers.
I suppose your Russian drop-outs must have been fervent devotees of {1|immutable2} > 3. What was it attracted them to Lojban in the first place, such deviation from that ranking quenched their interest?
> That's why any changes to basic gismu, to common usage is a way to
> the final destruction of the language as it happened to other
> conlangs.
You seem to have a strange notion of what language destruction is. You seem to think that a language exists if and only if it has speakers in our world. I accept that that's not a nonsensical view, tho I do think it's utterly wrong, but it's hard to have rational discussion if we use the same set of terms with such fundamentally different and incompatible senses.
> For users of handheld devices, Gleki has made an android jbovlaste app -- it's excellent!
> Huh?
There is a Google Play app maker called Vorgoron who made the app and gave it a description that says "Author - Gleki Arxokuna", which had misled me into thinking you were Vorgoron. Looking at Vorgoron's other apps, it seems likely that Vorgoron is Russian, so likely somebody you know.
--And.
|
Gleki Arxokuna <gleki.is.my.name@gmail.com>: Sep 10 05:35PM +0400
> (4) is important for English, maybe not for Lojban.
> All can conflict. Which trumps which? For me it's 3>2>1. For Bob I hope
> (because it's a position I can respect) it's 1>2>3. What do you think it is?
For me it's 2. codification > 3. logic > 4. consensus > 1. usage
although 4. defines 1. and partially 3.
>> Russian group who immediately stopped learning Lojban when they
>> learnt that CLL was no longer valid.
> Bob was talking about conlangs not conlangers.
Usage of conlangs depends on users i.e. conlanger (although i preferred to
use "conlanger" for the term "inventor of conlang")
> I suppose your Russian drop-outs must have been fervent devotees of
> {1|immutable2} > 3. What was it attracted them to Lojban in the first
> place, such deviation from that ranking quenched their interest?
There is CLL which is the reference grammar.
When someone says (and proves) that the refgram is no longer valid the
language stops to exist.
This way Lojban loses one of its selling points: the most
complete/described human language ever.
CLL is the first (and the best imo) book teaching Lojban.
What others offer instead of CLL? Nothing. Just waiting for Robin to do
something instead of them.
>> the final destruction of the language as it happened to other
>> conlangs.
> You seem to have a strange notion of what language destruction is.
Destruction is exactly when you ignore or invalidate one thing and not
providing alternatives.
Those tinkerers (including xorlofiers back in 2003) threw CLL away and thus
imo destroyed the original plan of a stable language and doomed the
language to the fate of Loglan. Only the lack of the third alternative (a
new loglang) prevented this community from complete dying.
> our world. I accept that that's not a nonsensical view, tho I do think it's
> utterly wrong, but it's hard to have rational discussion if we use the same
> set of terms with such fundamentally different and incompatible senses.
The existence of speakers doesn't matter. If you have selling points you
will get speakers.
The second selling point is monoparsing.
But they removed the first selling point.
> it a description that says "Author - Gleki Arxokuna", which had misled me
> into thinking you were Vorgoron. Looking at Vorgoron's other apps, it seems
> likely that Vorgoron is Russian, so likely somebody you know.
Oh, I see. I just offered him to include a dictionary into their
distributions. I don't even remember who is the original author of that
dictionary.
|
Alex Burka <durka42@gmail.com>: Sep 10 10:31AM -0400
Gleki, you’re not making any sense. In one breath we’re holding a bonfire and torching CLLs, while in the next we’re sitting twiddling our thumbs “waiting for Robin to do something”. Obviously, neither is true. And Lojban is still here, in contradiction to what you keep saying, though we disagree on the reasons why it languishes. I wish we could have this argument without hurling insults.
mu’o mi’e la durka
On Wednesday, September 10, 2014 at 9:35 AM, Gleki Arxokuna wrote:
|
Alex Burka <durka42@gmail.com>: Sep 10 10:32AM -0400
I think it’s 3>1>2 for me, although I’m not exactly sure of the distinction you’re drawing between “usage” and “consensus”. And to keep 3 above 1, you need 2 to be able to slowly adapt to 1/4… so it’s intertwined.
mu’o mi’e la durka
On Wednesday, September 10, 2014 at 9:16 AM, And Rosta wrote:
|
Gleki Arxokuna <gleki.is.my.name@gmail.com>: Sep 10 06:34PM +0400
> is still here, in contradiction to what you keep saying, though we disagree
> on the reasons why it languishes. I wish we could have this argument
> without hurling insults.
1. You don't provide any alternative. Where is your amended CLL? There
isn't one. Thus you are waiting for someone, what else can I assume?
Yes, torching CLLs and doing nothing constructive.
2. In what form does Lojban exist? It exists despite these invalidations of
CLL and continues to avert other people from contributing just because of
xorlo and other changes.
|