[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: [lojban] Re: Why we should cancel the vote or all vote NO (was RE: Official Statement- LLG Board approves new baseline policy

At 07:02 PM 12/2/02 +0000, And wrote:

> >Anyway, I definitely think that nothing
> >in the never-used portion of Mex should be monosyllabic
>
> Which is a surefire way of making sure that Mex would never be used.  If it
> takes lots of syllables to say even the simplest thing, people won't be as
> likely to use it

I have the same anxieties about ordinary bridi that are logically explicit.
I don't see what is special about mex that gives it a special claim
to brevity. I'd have thought it had the least claim to brevity, given
that it is never used.


It has a claim for cohesiveness of design, so that it can be used.
To give an example from Mex, usage frequency would not justify giving the hex digits monosyllables - they aren't used as much as the base 10 digits and not as much as many of the disyllables. But as a cohesive system, the digits make sense as designed.
lau will never exist as a standalone word - it will always be paired with another word (maybe more - I can't remember the grammar off the top of my head). Thus any lau construct is ALREADY polysyllabic, with the lau serving as a common beginning to give a grammatical cue as to what is going on.
There were a lot of such considerations where design decisions were made NOT based on individual word frequency, but based on how the words would be used in a context. Overall things came out quite well, with the only gotchas coming where usage of the language uncovered a new multiword construct that had not been planned for, such as du'u or ce'u.
My personal feelings in all this, given the desire for monosyllables, is that the byfy should probably reabsorb the xVVs, and consider using the 4 monosyllables for high frequency combinations that have unexpectedly emerged. But I would not reassign any words: ce'u which would otherwise be a good candidate, should remain ce'u because otherwise several years of Lojban text, and prior learning, is invalidated.
But I don't feel strongly on this, and indeed have a good argument against it. For all that it takes 4 syllables, I have found that, in speech at least, "la'edi'u" works very well for aural parsing, just as "lenu" and "lesedu'u" and "sekaileka". These make the language easier to listen to and understand without conscious parsing (you hear "sekaileka" and think "property", not "sumti-tcita with an abstraction". If all the key constructs are monosyllables, then that means that they lack contrast, and you have "noisy environment" errors where people hear "xau" as "kau" or "gau". We've already found that the se/te/ve/xe series is troublesome in oral communication, and we changed the rafsi so that at least in lujvo there is contrast (this is one reason for using the se/te/ve/xe lujvo rather than separate words, is that it adds that oral contrast).
It is hard to argue these things in a community that thus far is mostly text-oriented, and in which some subjects are not discussed to the degree that they might be under fluent use. In the case of Mex, that entire sublanguage is a linguistic experiment: can we come up with a speakable form of unambiguous Mex corresponding to the written mathematical notation. If you design it to be unspeakable, it most certainly will never be spoken. If you design it to be speakable, it might be used, and in fact if used could provide a new application for Lojban that expands our potential community and applicability. (Lojban: The International Language of Mathematics ??? %^) 


If TLI dies, it could be that
> many of them would respond to a polite "We are reaching out to the Loglan
> community, seeking your support for a reunited Loglan/Lojban effort, one
> that will honor JCB's legacy by showing the world the fruits of his
> efforts.  We've made some provisions to make the learning-transition from
> TLI Loglan to Lojban a little less painful, and you have our demonstrated
> commitment that Lojban will not be changing so that relearning will never
> again be an issue.  We invite you to join us (and ideally: we have the
> consent and encouragement from the TLI leadership in making this overture
> to you.  They recognize that Lojban is the future of Loglan, and we want
> you to be a part of that future.)

Like Steve, I support this.
I think it costs nothing. Unlike Steve, I don't think it belongs in the baseline statement. In an earlier draft I had an explicit invitation to McIvor, but removed it precisely to try to prevent this sort of acrimonious debate over Loglan, when in fact it is tangential to the baseline process. Instead, the Board voted to privately extend a special invitation to him as an outreach to the Loglan community, which SHOULD have made the matter a non-issue for the Lojban community, but I did not expect Steve or anyone else to raise the issue. 


> >  The old guard have more of the force of a dead weight,
>
> Gee, thanks for the vote of confidence!  Should Cowan and I resign now?  pc
> already did

I've sung your praises many a time, so I won't wax sycophantic now. But
you yourself will recognize that at least in the last year or two your
contributions have mainly been nay-saying rather than leading new
initiatives.
I think that if you look back in history, my job has been "naysaying" for a lot more years than just one or two. More like a dozen. This is not necessarily my preference, but rather what I see as my obligation. Trying too many new initiatives and failing at them makes us look like failures. Waiting till the right moment has led to good results. I was non-supportive of the wiki when it started. It started anyway and did well, and I am as strong a backer of the thing as anyone for the purpose that it serves. The byfy was an impossibility 5 years ago; now it is a necessity. The job of dictionary writing did not change, but the community changed and matured and that is now the right way to produce a dictionary rather than Lojbab going off into a corner for a year. Believe me, I like the responsibility off my back. 


> >an impediment, and
> >it is the voices of the likes of Adam, Craig, Jordan, xod, Robin and
> >(horribile dictu) Jay that we should be paying the most attention, as
> >most representative of the future of Lojban. AFAIK, Michael is the
> >only Lojbanist over 40 who is actively and publically producing
> >Lojban text. Furthermore, we can see in these young Turks a dynamism
> >in trying to move Lojban forward, in sundry ways, that is in marked
> >contrast to the current inertia of the oldies who in the days of their
> >own youngturkhood founded Lojban, and I expect it won't be long before
> >the new generation take over the board and decide to forget about the
> >old guys who were forever stating the conditions under which they would
> >refuse to learn Lojban without ever showing signs of a real intention
> >to learn it
>
> Are you a parent?

One son, almost 7. I take your point; I devote an excessive amount of
time to Lojban, and am hoping to radically rein back once the BF is
done.


%^)
My lack of writing is almost certainly due to the demands of parenting. I now have my own leather bound copy of Burton's Arabian Nights, and I'd love to have time. 


> The history of the project has been one where most of the work has been
> done by college students (and recent graduates who aren't yet established
> in their careers) and retirees, because the people in the middle have
> careers and families to raise.  TLI Loglan is heavily weighted towards the
> older people and retirees, and JCB wasn't too good in later years at
> recruiting college kids, which is why they don't have much new blood.  I
> made my strongest effort at recruiting the next generation, to ensure that
> Lojban would survive me, and clearly it will.  But, God willing, I have 30
> good years left, and my kids will be grown in only a few.  Then I'll have
> the time and todays' college kids won't (though with effort we'll be
> getting new college kids)
>
> But if people don't want me to hang around ..

I'm sure you're right. Still, we don't want the people with the time and
energy to be held back by those without.
I'm the last to tell people not to write in Lojban. More! More! I'm still not satisfied! %^)
Actually the changes we've made in the Board and LogFest meeting procedures will make it easier. And the members list provides an avenue for formal input in between LogFests. In non-emergency situations, we can make more efforts to consult with the community (before making decisions), but I think we need to have this less than open-ended. The Loglan debate was not necessary to the baseline policy discussion. More importantly, changing how we make decisions is a different discussion from what the decisions should be.
What the Loglan debate shows me however is that I DO have a role, even if I seem to be a nay-sayer. Diplomatic talk seems to be lacking from these discussions, and people saying what they think has ruffled feathers unnecessarily - nothing that has been said requires any change in my policy towards TLI and Loglan. It just requires words that bridge between factions like the ones I wrote above. Consensus group work is an interesting and sometimes difficult skill. Interestingly enough, JCB did his PhD work on the subject (I've never read his thesis, but that was the subject) and then proceeded to ignore whatever he had learned. My whole Lojbanic life has been about learning from JCB's lessons, so we do better. 


> Those that refuse to learn the lessons of history are condemned to repeat
> them.  There are MANY artificial languages that have failed.  I contend
> that my vision has gotten us into a rare position of potential success, so
> one would think that people would give me a little benefit of the doubt
> that I am leading us in the right direction
>
> But if people reject my vision, I won't do as JCB did and fight to the
> death of the language to keep control

Winston Churchill lost the 1945 election by a landslide. He had almost
single-handedly saved the country during the war, but was not the right
leader for the peace.


He might have been, but the people wanted a change.


I'm *not* saying the same actually applies to you,
but I do mean to say that your leadership in the past, which I admire
hugely -- and I do give you all the very great credit for holding the
community together and fostering the tolerant pluralistic climate that
we have here -- doesn't *necessarily* guarantee that you're best placed
to lead us forward into the next era. I really mean no more than that
(but my private opinion is that I'd vote for you as Head of State of
Lojbanistan, but not necessarily as Prime Minister).


%^)
We haven't got the structure to separate the two jobs. Maybe we need to find ways to do so. For a long time, Cowan was my prime minister. Now my cabinet seems to include Robin and probably Nick and Jay in key roles, and the Board rather than Lojbab will be making most decisions. The more I delegate, the less I have to be all things to all people.
That may be sufficient to end the crisis of confidence in the community leadership. But it takes time to implement change. If the policy is backed now, people will have a chance to see whether it has borne useful fruit next summer, and that is the time to change it if it needs changing. 


> I question that in natural languages it is meaningful to say "bad for
> semantic reasons".  Natural languages as well as artificial languages are
> humpty-dumpty: words mean what we want them to mean, and if communication
> occurs then the semantics is "correct"

If a foreign learner says "I am knowing the answer", what is the nature
of the error? It is grammatical,
End of point. Find an example that is NOT a grammatical error. When I refer to grammar, I distinguish it from semantics. 


yet it is bad for more than stylistic
reasons -- it is not bad merely because native speakers wouldn't say
it. Communication is unimpaired, so it's not bad for that reason. The
problem is semantic; the meaning encoded by the sentence does not
naturally serve as a basis for inferring the meaning the speaker intends
to communicate.
"I am going down to the pub" when the pub is at the top of a hill and I am not, is a semantic error of this sort made by perfectly native speakers, yet it is not considered "bad". If indeed the relative altitude is important, the idiomatic usage will interfere with communication. Likewise someone describing how they enjoyed a "gay celebration".
> > > The whole area of alphabets and lerfu is tied up with Mex. We cannot say > > > how useful Mex will be, but it certainly will not be useful if we make it 
> > > more difficult to use
> >
> >Avoiding making mex harder to use is not a good reason for not making the
> >rest of the language easier to use
>
> I disagree

You I think are mex's sole fan.
It doesn't have a large constituency, but TLI not having one, just as it not having a proper tense grammar, kept on cropping up and biting. ("Four score and seven years ago ..." 


> >I am proposing (and I think Jordan is
> >too) that mex and other stuff that has never seen substantial usage be
> >made more longwinded so that future generations of fluent lojbanists can
> >decide where shortwindedness can most efficaciously be applied
>
> Whereas I think it likely that future generations of fluent Lojbanists will
> do the RIGHT thing and start over from scratch to design Lojban Mark II,
> based on the experience of Lojban over a couple of generations

Hmm. Do you really?


Yes.  But it won't be any of us that does it, if it is done properly.


It seems likely to me that there would be a strong
wish to stick with Lojban Mark I, because it already has a speech
community.
Certainly. But what were the reasons for creating Loglan/Lojban in the first place? If the fluent Lojbanist find that Mark I does not in fact result in a "logical language" but does come closer to that goal, they more than we will be able to spec out what Mark II needs to be, and given the nature of the community, someone undoubtedly will go build it. The existence of Loglan and Lojban has not stopped guaspi and Ceqli from being written, and no doubt if you ever work on Livagian (sp?) again, you will do so informed by what you've learned from Lojban. Why presume that native Lojban speakers won't have Tolkien's "Secret Vice" as well? 


If you really think that future Lojbanists will give up on
Mark I and work on Mark II,
Who said that they will "give up on it"? They'll probably experiment with things though, and whether they adopt them will be based on cost/benefit tradeoffs that we cannot begin to see. That was why I explicitly have left hooks in the design for future use like the experimental cmavo and Mex (Mex needed more hooks which is why it takes up so much grammar, but we have a large and flexible architecture).
There is only one thing in the language that MAY have been proven to be an unnecessary hook. Unicode standards may have made much of my alphabet switching apparatus unnecessary. But until people know how to use Unicode better, it will be difficult to write the teaching materials that might free up the alphabet switching cmavo. That is probably the only "change" I would argue for consideration by byfy in the absence of usage, if those who know Unicode (Cowan?) can explain things to us in a useful way and a streamlined design is devised that does not change too much, and I would be happier to let a proposal evolve over the next 5 years and be dealt with by Lojban speakers. 


 then maybe I should do what xod advised me
to do & be thinking of working on Mark II instead of Mark I. What do
you reckon?
Way too premature. No one will know what Mark II would need to be like until the generation that learns Lojban AFTER the baseline ends tackles the problem. At that point YOU'LL be the old fogey standing in the way of progress. If people want to make notes for things they would do in a Lojban Mark II on the wiki *IN LOJBAN*, I am sure that the generation in question will find your ideas interesting, if quaint. %^) 


> >and partly because you use it without scrupulous regard for meaning
>
> Correct.  I don't accept that debates over semantics will establish
> meaning.  Communicative usage will establish meaning.  Lojban has a defined
> grammar, but the semantics will (or at least SHOULD) be defined through usage 

I hope you realize that you are in this respect an arch-Naturalist,
very much at one extreme of the spectrum of views.
That is the necessity for a Sapir-Whorf test, IMNSHO, which is part of the design criteria. I'm not a "naturalist", but an engineer thinking like a scientist, and wanting the experimental subjects to explore that variable. 

Lojban is supposed to be metaphysically minimalist.


When you express
these views you should make it clear that you are expressing them as a private
Lojbanist, not as president of LLG. Nick is more representative
of the middle-ground on this issue.
Yet I agree with him on almost all issues, and most of the time when we disagree, I can convince him. (When I cannot reach agreement with him, it is usually because I haven't learned enough to understand what he is talking about).
And yes I am talking as President of LLG. I am defending what I see as a design principle. I so seldom stop to think about what >I< want for the language that I doubt that I could easily figure it out. I'm almost always representing a constituency, not myself. 


> >and so say things
> >like {le nu} when {lo'e du'u} would be more correct
>
> It wouldn't be "more correct" because there is no official semantics to
> that phrase, and since I do not accept your jboske debates on lo'e as
> having any meaning, and thus haven't he vaguest idea what "lo'edu'u" means

Assuming that there are others who think as you do, then we have
two different dialects: Naturalist Lojban and Formalist Lojban (to
use Nick's terminology).
We have many different dialects. You forgot Helsem's dialect, which I doubt that anyone else can speak. And Jorge has made claims that amount to speaking a different dialect.
All of them parse using more or less the same parser (if you plug in processing for experimental cmavo). Thus I am not bothered. 


We must bear in mind that when you speak
of Lojban you generally have Naturalist Lojban in mind, while when
I speak of Lojban I generally have Formalist Lojban in mind.
The only Lojban that matters is the stuff that the byfy will find when it looks at the corpus. 


Some people might think that you are being undesirably schismatic
in your dismissal of jboske,
I don't dismiss it as an adjunct of the community. I said that its debates mean nothing to me. I don't understand most of them, and I don't accept the results of those debates as binding. Now if those debates are input into the byfy, and achieve resonance, then the byfy discussion will make them relevant to me. Until then, they might as well be discussions of Mark II. The language I write in won't be informed by stuff that I haven't paid attention to. And I think that if I have time to write in Lojban then that is better time spent than reading the debates about usage that has not happened. 


but I think it is better to let each
dialect go its own way, in the hope that Naturalists will gradually
be absorbed by the Formalist dialect once the Formalist has
progressed to become a living dialect more than an object of debate.
If it becomes a living dialect, that is PRECISELY what you should strive for (and I'll probably back it if I understand it %^) 


> >But I have the impression that within Loglan/Lojban there
> >has always been the more whorfian Conservative faction and the more
> >logical/engineerist Progressive faction
[...]
> >Lojban was founded by Conservatives,
>
> No.  Lojban was founded by people not aligned to any of your factions, and
> I continue to reject them.

It is hard to accept that you reject factionalism when you doggedly
represent the most extremist position of anyone.
I represent the JCB faction, and he is dead. And I represent the not-on-line faction, and they aren't around to argue. And I represent the faction of those who have never discovered Lojban, and the faction of those who discovered it and who are waiting till it matures. I represent the people who may or may not use it for Mex, the people who may or may not conduct Sapir Whorf research, and lots of others.
I balance all of these factions in my head, pretty much subconsciously these days. Then when the community speaks I attempt to fit what they say into the balance.
Most of the time it looks like I am a radical conservative, until I turn out to be a radical at the other extreme. I think I surprised both Nick and xod by backing the position you see in the policy statement. If I ever get back to being a Lojban user, you might be surprised at how I use the language. So might I. 


> It was founded by someone who was committed to
> certain design principles.  Once beyond those design principles, I daresay
> that I was a good deal more "progressive" than most of you

Conservative = resistant to change; Progressive = in favour of change
for the better. You have said that Lojban was founded to stop Loglan
changing; and you have stuck to your guns ever since. So it is hard
for me to squint sufficiently wonkily to see you as a progressive
rather than a conservative.
On the other hand, much that is in Lojban is a radical departure from what JCB was willing to accept, so to him I was a flaming progressive. At one time YOU thought that the apostrophe was a radical idea, even though all it did was explicitly display the phonological phenomenon we wanted to see between disyllable vowel pairs. JCB never accepted it.
I am progressive when the time is ripe for change, and radically conservative or evolutionary at other times. That is three different points on the scale and I am all three depending on the problem. That is why I don't think I fit such a factional categorization. 


> >but it never advertised itself as the Conservative alternative to
> >Progressive Loglan; it advertised itself as the thriving and publicly owned 
> >alternative to the moribund and privately owned Loglan. So I don't think
> >it is fair to argue that Lojban is intrinsically Conservative
>
> It also advertised itself as committed to language stability, to the
> elimination of the "moving target", to getting the language DONE and out of
> the hands of the endless fiddlers

Those things passed me by. I found out about them only after I was
already involved. My first impression of Lojban was "look at all these
cool features we've got", so my early impression was that its main
goal was to have cool features.
Several of those cool features were introduced by me. The tense grammar was Cowan's, and this "arch conservative" embraced it almost immediately as the Right Thing because the time was ripe to consider such a change. 


> NO!!!  The first and foremost consideration of a baseline is that it is a
> snapshot of the language design at a current point.  That snapshot receives
> the full organizational support and does not change while the baseline is
> in effect.  (Note that what people actually do may differ from what the
> organization supports, depending on how much control the organization has
> over the community)

I see. I was thinking of the baseline as more like the law. The law is
published, and everybody knows what the law is, but new laws can be
added. Actually, that is compatible with what you say and with what I
said. Which bit of what I said are you saying "NO!!!" to?
A baseline indeed is like the law. It has formal procedures for change, and there is an extensive set of documents describing what the law is. But even law-abiding people sometimes break the law either accidentally or when they have good reason. Still the government enforces the law as written in theory, though in practice it winks at certain violations.
My "no" is that the baseline would leave things undefined and then define them over time. Ideally at this point there will become one last unchanging baseline, and it will never be officially amended. In practice, after the 5 years, the byfy might become like the French Academy with prescriptive power but no enforcement power.
Or more likely it would be like the computer language standards committees, which is where the terminology comes from. There is a standard for HTML (and for MIME and for COBOL), but no one implements those standards perfectly or in the same way. The standards are the baseline. (The following may be inaccurate in the details regarding HTML, but shows how baselines and languages have gone together on some projects I've worked on.) Your browser will handle some things well and it won't handle others, and other people will write pages that do or do not follow the standard as they choose. There is a negotiation process such that what people actually put on web pages may or may not conform to the baseline standard, but where it doesn't, other users may or may not adapt to the non-standard behavior at their discretion (or choose to ignore the pages of someone who is too deviant from the standard); they also have to find a browser (which corresponds to a Lojban parser) that will process their dialect. Every once in a while the HTML standard is revised, and those revisions may or may not be heeded, but the revisions are not lightly made in any event and there are numbers of proposals that have not become part of the standard floating around as well. Still, when someone wants to know what HTML is, they look at the standard, and the proposals are not even officially acknowledged until they are ripe for consideration (and written up as formal proposals). 


> > > >and instead
> > > >simply say that the mini-dictionary fixes the meaning of the cmavo it
> > > >lists. A proper syntactic parser should not have the mahoste built
> > > >in to it, but should instead take input from a community-maintained
> > > >mahoste that can be updated with cmavo not listed in the mini-dictionary 
> > >
> > > Then write one
> >
> >I have (collaboratively) written one for cmavo that are not in the
> >official mahoste. It is on the wiki. It is easily adaptable (with
> >about 1 minute's work) by anyone writing a parser to take input from
> >a mahoste
>
> Write the parser then.  No one else is likely to do so.  The official LLG
> parser is Cowan's parser

I don't have the skills to write a parser.
Neither do I. But if you can't write a parser, please don't assume it is easy to write one that does what you want. It might indeed be easy, but it is not the approach that either Prothero or Cowan or Curnow or any of the other Lojban parser writers have ever used EVEN THOUGH when they were written the cmavo list was more likely to change than it is now. This suggests that people have not felt that to be the optimal way to write such software under such conditions. It may be easier to hard code the lexer to include changes in cmavo assignments than to make it a file-lookup; neither you nor I can say. 


 I'm a formal linguist, not
a computational linguist. I'm happy to advise you on how I think the
parser should work, if you are suggesting that a new parser should be
written.
I'm suggest that people who say that the software should work a certain way need to write software to work that way, or find people on their own willing to write such software. I have trouble finding people to do priority jobs, and parser writing right now isn't one. 

lojbab

--
lojbab                                             lojbab@lojban.org
Bob LeChevalier, President, The Logical Language Group, Inc.
2904 Beau Lane, Fairfax VA 22031-1303 USA                    703-385-0273
Artificial language Loglan/Lojban:                 http://www.lojban.org