[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [lojban] Response to Robin's "Essay on the future of Lojban"
Robin Lee Powell wrote:
On Fri, Apr 09, 2010 at 08:11:41AM -0400, Bob LeChevalier wrote:
I call upon everyone involved in this discussion to reread the
policy http://www.lojban.org/tiki/Official+Baseline+Statement
I think the success of that policy speaks for itself.
I think it has been quite successful in achieving a stable language.
Obviously not in getting the baseline done.
Pretty sure it's called "the baseline statement", so I call that a
total fail.
The purpose of a baseline (and a baseline statement) is to reassure
people about the importance of a stable language. That part was
successful. I agree that finishing the job was not.
With regard to the stuff that is presently covered by the baseline
documents, I can live with your position, which is easier to sell
to computer people and harder to sell to linguists. I am also
sure that if it gets to be a problem, you are a reasonable person
and would reconsider.
What I want is to understand what you want, right now, to fix the
deadlock.
What I *want* is not something I can have, so at this point I am not
worrying about it %^) It would include John Cowan, who had the skills
to resolve deadlocks merely by the force of his personality and his
credibility in the language.
Doing the minimum necessary differently to eliminate the perception that
there is a deadlock (because I think it is a perception and needs to be
changed in ways that perceptions are changed).
Because your frustration has hit the level it has, and because YOU are
critical to any solution as jatna, a key priority for me is to find a
solution that reduces that frustration while getting the job done. This
means NOT relying on the medal performance you refer to below.
It looks like it might be sufficient if you just expand on the last
two paragraphs: where do you think we agree?
Most importantly, that the status quo is unacceptable, and that you are
going to take action which will make things more acceptable. NOW (which
I actually understand as zipu'o rather than ca, if I remember my tense
stuff)
what does "declaring the language" look like to you?
Completing something the community will be satisfied with as a
prescriptive baseline.
Coming up with a way to satisfy those who want more than the baseline
statement requires in terms of question answering or continued logical
specification, when it arises after the baseline is done. But this is
not an immediate priority.
What does "I can live with your position" mean, *exactly*?
I THINK it means that, having reached the point we are at, I accept that
some things will have to be done that I disapprove of, but that I can
tolerate if it a) gets the job done and b) again makes you happy to be
doing the job. (I'm not sure that "happy" is the right word there, but
someone carrying the responsibilities you have assumed deserves to get
some positive feelings out of fulfilling them.
What this means is that for those aspects of what you wish to do that
affect the short term job of getting the baseline done, I want to speak
my piece, but otherwise don't intend to fight you.
I am willing to help in the ways that I am good at helping. This
demonstrably does not include writing sections. I am not sure what it
does include, but I think we can come up with some things that will make
dealing with me more positive than negative %^). Maybe I can even find
a way to get Nora involved. Or maybe I can just stay out of the way,
and as president, handle more of the non-byfy stuff, so you can do more
of what I cannot do.
-------
On the other hand, Nora just said something that I think profoundly
reflects my misgivings. She's tried (far more than I have) to
understand xorlo, but she just can't get her mind around it. That makes
her feel incompetent at the language, which is the strongest demotivater
for actually using it, much less doing byfy work.
I actually hit that phenomenon several years ago. I barely understood
Nick's formalization of sumti-raising. I still cannot wrap my mind
around lambda calculus. People can explain it over and over, and the
next day I remember nothing and understand even less.
If Nora and I are at our limit, and we're reasonably bright and were at
one time among the most skilled of Lojbanists, then the concept of byfy
continuing to refine and formalize things even more than they are (and
more than minimally necessary to USE the language) stands to make our
investment in learning the language a waste of time.
And a lot of people aren't nearly as dedicated to the language as we've
been. They hit their limit, and give them something new, and they
leave. Indeed for some people, to say that CLL isn't enough belies the
claim that Lojban is a "simple, easy to learn, language" (because they
don't realize that English described at the CLL level would be several
times as long and have many more holes).
Arguing with And Rosta gets me upset because it seems clear to me that
he isn't much interested in whether the language is something people can
and will use to communicate - he wants more formalization and logic.
And your essay suggests that the "activists" of the community may be
closer to his camp. I am less certain where you personally are, but you
have considerable credibility with me because you can and do use the
language competently as it is (as does xorxes)
"I can live with your position" therefore means that at this point I
trust you to not change the language so far out of my recognition that
even Nora and I give up, and I have a reasonable hope that by being
reasonable and helpful myself, we can achieve something we can both be
satisfied with.
I'm glad you think I'm reasonable; I'm not *feeling* reasonable
right now. :)
Hey, you have put up with reading several of my endless posts now, and
have answered them politely, and considerately, and indeed showing
you've read them all - and you haven't blown up. I consider that
*extremely* reasonable on your part, given that I know how you dislike
any post of mine longer than a few paragraphs. %^)
And if you survive this one, we may be at the point of agreement.
The point is that defining the current state *should* be
relatively easy to get agreement on, ideally a 100% consensus.
Certainly, but who is going to do the work? *No-one* wants to do
what you're describing.
What you've said and what others have said makes me think that there is
a solution. Here, off the top of my head ... (sorry about length, of
course)
BYFY has no fixed membership. It consists of you and whoever else is
helping at the moment. It is a concept and a function rather than an
organization. At some later time post-baseline we can constitute a
Lojban Academy if the community wants one, with fixed membership. How
people perform *now* should be considered part of the qualification for
membership *then*.
There seems to be a group of relative newcomers that would be happy to
copy and/or paraphrase the CLL materials (which are now online, right?)
and the cmavo list, on each cmavo onto the web page definitions. That
is the most basic step. Some can comb through the wiki and even the
lojban list archives and find where people have raised issues about that
cmavo. I think that treating the problem more like contributing to a
wiki page on each individual word will get us pretty close to a
description of the status quo without any actual decisions needing to be
made.
You don't have to do any boring grunt work. Newcomers to the language
will do what you find boring, but it will probably be helpful for them
in learning, and we can perhaps use that to sell them on contributing.
And the issues THEY raise are the ones that really need answering,
although I'm sure xorxes will have his fepni remei to throw in.
This was done in the last day for one word by two people and xorxes
partially on the page and partially on the list. If they were doing it
on the wiki page, that word might be well on the way to being done. At
some point a "shepherd" would have to go through and smooth things out,
and maybe the set of issues to be decided goes into a "trouble report
bin" to be dealt with. But with luck shepherding wouldn't be a very
onerous job, and maybe more editorial than technical.
And by all means make use of Matt's idea for a funny hat or a merit
badge (an LLG-paid CafePress T-shirt, with appropriate design, for
someone who contributes a certain number of definitions seems like a
winner to me. I don't know much about hats %^) Maybe a
Robin-autographed LS-book containing the set of byfy definitions when
done would be a *real* unique prize, especially as a prequel of the
dictionary. If we can spend hundreds of dollars to bring people to
LogFest, I'll support spending more to promote byfy work (but not
actually paying people to do work - the business aspects of "employees"
would drive us all mad).
Someone far more competent than I am at wiki-work can try to coordinate
all this activity, point out definitions that are ignored, and maybe do
some of the page smoothing.
The point is that these are tasks many in the community know how to do,
whether they are "skilled Lojbanists" or not. And we've reduced the
individual task size to chunks that people can do in an hour or less as
they have time.
I don't know if it will get everything done as quickly as you want, but
I think it will get things moving forward again without taking too much
on your part (or mine).
(There is more to this, below, when you ask a related question).
The language is what it is, warts and all. Can we agree on what
the warts are?
We can always agree on what the warts are; what we can't always see
is what the face underneath looks like. If the CLL directly
contradicts itself, what are we to say the current state is?
"CLL says X, Y, Z. Analysis has shown that X and Z contradict each
other because .... Usage examples 1 and 2 show this problem, which is
to be resolved by byfy." Then file the issue in the problem report bin.
You or xorxes or someone else can append some possible solutions
asynchronously, but before the problem is taken up for debate. (Below I
link to a real example of what I find ideal).
The creativity comes when all the words are done, and hopefully we then
have a complete set of problems that need to be addressed. How you take
up those problems (individually, in groups, etc) will probably influence
how they are solved.
There's another big issue here: I do not recall ever understanding
that what you wanted was to define cmavo *solely* based on the cmavo
list and the CLL, without any reference to anything else. That's
really quite a surprise to me. I don't know that it makes a
difference to me, but a surprise none the less.
I'm not sure that I do. Those two are the baseline documents, and thus
the status quo. It is commonly agreed that the cmavo list is inadequate
for dictionary definitions, which is the desired product, and that we
want more examples showing how the words are used in various grammatical
contexts
(Back in the day, I wanted CLL to have examples of every selma'o in
every place it is found in the YACC grammar. But that was too much for
John to produce on his own. If something like this emerges out of the
cmavo dictionary effort, that would be great, and would probably bring
out most issues that need deciding.)
At this point, with several years of usage, I think byfy should
reasonably be able to consider if that actual usage contradicts CLL in
coming up with Task 4 changes. But a small number of usages, or all by
one person is a weak case. I was concerned back in the day that xorxes,
by writing ten times as much as everyone else combined, would
effectively define correct usage as being "what xorxes does". I am
still a little afraid of this, though less so, because I can still read
and understand what xorxes writes in Lojban, so his usage is not all
that radical even if his proposals sometimes sound like 'too much'. And
we now have other writers of the language, who I'm sure are not xorxes
clones. You certainly aren't.
And another issue: none of this is relevant to the matter at hand,
in my mind. The matter at hand is that, for better or worse, the
BPFK was trying to do description-of-current-state and changes at
the same time; perhaps that was a mistake, but it's besides the
point. At some point, there will be change proposals. The matter
at hand is: what happens to those proposals? We have been
deadlocked for *five years* because one camp believed that the
silent majority wanted the language to be as stable as possible, and
the other simply wanted the language to be awesome. I wanted to
find out which was actually true, and head in that direction.
OK. Though I think the way you asked the question did not really
produce a useful answer (and I'm not a skilled pollster to know how to
do better). I want an awesome language which is as stable as possible.
And the silent majority is still silent; many of them are like Nora,
and unable to keep up with the traffic.
What is important to me is that in the fullness of time, there will
be change proposals, and if they're going to be deadlocked, I don't
want any part of the BPFK.
I can't promise that there won't be deadlock. But ...
That expresses, I think, the heart of what I want: a way to resolve
deadlocks in the PBFK when change proposals *do* occur. It seemed
to me that changing the current ordering of the stuff in task #4,
and making it more explicitely ordered, would do the trick nicely,
so I went to the community to find out what was actually wanted.
The approach I would like to see? (Not promising that it will work)
Here is how I see consensus politics in byfy:
We have models for how to write a change proposal.
http://www.lojban.org/publications/formal-grammars/techfix.300.txt
See especially Change 20, which refers to an article in JL18. Probably
the best presentation of a problem and solution I've seen. A month
after the problem was raised, I was able to use the material to teach
the solution to a Lojban class.***
If we have a problem, someone writes it up like this and someone
proposes a solution. People add commentary. If it seems that things
are ripe for a vote, open it for voting. Leave it open. Move to the
next problem.
The jatna or shepherd monitors voting. If not enough people are voting,
post the proposal to Lojban List and ask people to vote and/or comment,
but trying to keep to one item at a time. Include a link to the prior
discussion. If there was much controversy, a couple of paragraphs
summarizing the sides is in order. Changes 28 and 41 are change
proposals that were not approved, and have suitable summary paragraphs
of the opposition.
Likewise, if things are deadlocked, post the summary to Lojban List.
Let people discuss. If nothing happens after a few days to indicate
progress, then issue some sort of "deadlock alert" giving people a week
to show progress towards a resolution. If progress is being made, allow
more time, but not too much. If not, pick your solution of choice,
write it up and flag it as a deadlock.
When all issues but the deadlocks are resolved, you get to apply lots of
pressure. You have a writeup, you have the solution you want if all
else fails. Again on Lojban List, you post the issue and your proposed
solution. People have n days to either accept your solution OR to
propose an alternative AND convince others that it is a better solution.
Feel free to use all sorts of backchannels and the list. Apply
pressure liberally. Threaten to quit if people won't come to an
agreement %^) Offer a funny hat to anyone who can produce an agreement
that others will subscribe to. And point out that if they can't reach
agreement that they lose the right to decide that question.
NO formal voting at this point. You decide whether consensus has been
reached or not, and you decide how much support means "consensus" so
long as it meets the baseline statement criterion that no one objector
defeats consensus. There will be others including me who will help you
if you don't want to decide this. But you are the jatna, and to some
extent a benevolent dictator within your jurisdiction.
If even that fails, you go to the Board, say that byfy cannot reach a
consensus, and the Board votes on whatever choices you offer them. In
all likelihood, even if you aren't on the Board (ha!) you will get
whatever you want because the Board knows that to overrule you probably
guarantees that you will quit, and would risk the antipathy of the
community and membership.
But in any case, things are decided.
The complications are when issues are dependent on the resolutions of
other issues. Ideally, all changes are decided independently, but you
in your infinite wisdom as jatna will be able to look at the issues and
say that issues X Y and Z really need to be resolved before anything
else. So you carry those issues through the full process above before
tackling the other issues.
Otherwise, the norm is that the most contentious items are saved for
last. The closer things get to the end, the more pressure there is to
get things done. In the recent health care reform, enough people who
said they would not support the bill without anti-abortion language
found a symbolic way to surrender, while holding to their principles (an
executive order that might have no real effect)
Now recall that I am committed to consensus politics. I may oppose a
change every step of the way, but when we get to the final crunch, and
you are putting the pressure on, then I have no choice but to put up or
shut up. To continue to fight when we are at the final buzzer, is to
admit that I was wrong about consensus being achievable. Moreover, if I
were willing to destroy the chance for a consensus baseline, which is so
important for the language, I sincerely hope that the community would
collectively vote no confidence in me and ask me to leave. I WILL NOT
be another JCB.
And if the community can choose to ostracize me, then they also can do
so to anyone else who is obstructive. But Lojban has had plenty of
chances to destructively schism, and I don't think it's too likely. Too
many people have invested too much time to throw it all away.
--------------
***I don't know how to effectively search googlegroups's archives yet.
http://nuzban.wiw.org/archive/9207/maillist.html
has links to 2 posts (part 1 and part 2) from Colin Fine which started
what became Change 20.
In direct contradiction to what I said mere minutes ago, I'm willing
to push the BPFK to start by describing the current state of the CLL
and cmavo, even though that seems like mindless grunt work and I'm
going to hate every minute of it (I expect to be the only one doing
it, and if it works out this way *I expect a fucking *medal* for
it*),
Solid gold if I have my way.
But I think you have the leadership skills and the respect of the
community enough that we collectively won't allow you to waste so much time.
but if it'll make things smoother, I'll do it.
What I am *not* willing to do:
1. Publish *anything* before a dictionary.
2. Publish a dictionary before all non-trivial outstanding change
proposals are completed.
3. Continue working on the BPFK for even one more *second* under
the prior conditions, which were that Broca and you and Nora would
block any change short of a direct internal contradiction *just
because it was a change*.
I agree, without reservation.
The above discussion shows one way you can prevent me from blocking
anything, with my advance blessing. That you got broca's public support
for your platform/poll suggests that he is easier than I am %^)
> I want the BPFK aligned behind a
set of goals *for the language*, and committed to acheiving them,
before I will do *any* further work on it.
I'd rather you wait on this one, in part because my suggestions above
kinda obviate a formal BPFK that needs to be aligned.
If the community won't do the job and you are having to do it yourself
and earning that solid gold medal, then you yourself are the byfy, and
you are aligned behind any set of goals you can agree with yourself on %^)
If such a set of prioritized goals is needed at the last stage to
resolve a set of deadlocks, I think it will emerge by itself (or it can
be a technique you use to resolve them. Under goals XYZ, deadlocks ABC
are resolved such and how. Rather than presenting goals in the abstract
as you have, you are showing how you think they should apply to the
specific problems we are facing. People will agree or disagree, but the
process will get their de facto alignment, not by the BPFK per se, but
by the active community acting as the BPFK auxiliary.
I think we can turn that into a middle-ground proposal. Your turn.
I hope I said yes enough and offered enough positives that no more turns
are needed.
If so, then my position becomes no more than having byfy vote that
the part 1's accurately reflect the status quo before trying to
finalize agreement on the solutions. Ideally, we'd have a
complete set of task 1 description before we start formally trying
to approve the changes, but I can bend there.
I'm not at all sure I get why that (re-writing the CLL in the form
of cmavo definitions before anything else) is *so* important to you,
but if that's really all you want in the short term, that's quite
reasonable.
Its getting all the grunt work out of the way so that ALL that is left
is the hopefully small percentage of issues that are controversial.
We need to do *SOMETHING* now.
Agreed. If only because you, as byfy jatna, want something done now.
No. We need to do something now because the BPFK has been stuck for
five years, and I'm not sure I can count on the fingers of only one
hand the major, active Lojbanists that are ready to walk out.
I agree because YOU want something done now. That others want it is
icing on your cake. You alone deciding that it must be NOW is
sufficient for me to agree NOW. I'll try to influence what is done NOW,
but I won't stop you from doing *SOMETHING*
If you don't get from that, the importance I feel YOU, and your
good-feeling are right now to the community, I don't know what I can say.
The problem with documentation in Lojban is that it's totally
useless for someone learning the language; changes need to make
their way back into the CLL and L4B and the dictionary and such.
However, as long as someone's willing to step up and turn our
in-Lojban discussion into changes to those documents, I *love* the
idea of the BPFK being a Lojban-only forum. It'll be all
me-and-xorxes, all the time. :D
One hopes that, post-baseline, there will be others who will be willing
and qualified. But I'm willing to risk it being you-and-xorxes if
absolutely necessary.
You seem to be turning this around, suggesting that if there are
no rules people tend to say nothing, but ask for the rule,
That's my experience on IRC, yes.
I think that is the nature of the medium, and the willingness of people
to answer in English with the rule, and perhaps the sort of people who
habituate IRC as well.
We did not have that in Lojban conversation sessions here. Several
people would try, sometimes needing correction in order to make sense,
and in most cases the conversation didn't get all that sophisticated,
but I remember modeling a three-level deep abstraction bridi with at
least one relative clause. I had to repeat it slowly a couple of times
before people understood it, but they did.
In sessions of The Phone Game, asking someone else was pretty much
against the rules. People made mistakes and learned from them when
there was commentary afterwards..
I don't remember exactly what happened at Logfest several years ago, but
xod, who was almost certainly not at xorxes skill level or your present
skill level, seemed to be willing to go without checking the rules.
(Whatever happened to xod, anyway?)
If we had been forced to conduct annual meetings in Lojban, and we
provided translators like Shoulson for the incompetent, I think the time
pressure would have eliminated requests for the rules (people might have
done word lookups, but that is not the same thing).
It wouldn't necessary be awesome Lojban, but it would work. And awesome
will come from lots of non-awesome no matter how we get there.
I think that rather than having a fixed time for freezing, we should
simply say: "the language is currently good enough; future
discussion of changes in Lojban only". I think that would slow down
the rate of change *quite* enough for your purposes.
Probably, but then, as you said above, it might just be you-and-xorxes,
at least at first, and the two of you almost certainly can act quickly
with no one else involved in the decision. (Not saying you would, of
course, but you have the competence)
Again, I don't think a freeze will make any difference to
popularity, because nothing about a freeze will advertise the
language to people who wouldn't otherwise find it.
We haven't talked about advertising and promotion, but up till now it
has been word of mouth since 1989 Worldcon, and maybe a couple of other
cons where we've handed out brochures.
A 2nd edition CLL, dictionary, and the other materials, available both
in print and on the web, and we can comfortably do some real marketing
in multiple communities. I have some ideas but really this is where
Matt should be asked (as well as anyone else with real marketing
experience).
If there is a freeze then that is the answer to anyone who asks whether
the materials are up-to-date. We have experience of this. The key
papers that went into CLL were largely written in 1994. 1994-1998, the
community stagnated, JL ceased, discussion on Lojban List closely
resembled your deadlocked byfy threads, with limited participation. The
debates themselves drove people away. No matter how many baselines we
declared, having people argue about the language in great volume and
length drove away most everyone but the diehards.
CLL came out, sold several dozen copies the first couple of months, and
Lojban List and the community size and activity exploded. Then the
reports came in about the random blank pages in the book due to printer
incompetence, and Nora and I effectively shut down for a few months to
try to solve the problem, some of the steam went out of the growth, book
fulfillment went to hell and never came back.
Even so, language use kept growing, even without a dictionary, and I
don't think there was much talk about changing the language - as I said,
la .alis was "usage deciding" - until I raised the issue 5 years later
with the baseline statement.
If Matt had been marketing and order fulfilling in starting in 1997, the
Lojban community would likely have been 10 times its current size by
2002, and MUCH larger if we had gotten the dictionary done, since that
has been the key obstacle to growth.
It *may* make a difference to fluency, but I'm not even sure about
that.
The commitment to our product makes a difference to some people's
motivation to make the time investment.
Of course, just committing CLL to paper had enormous effect. On-line
source documents are too ephemeral for some people. That heft TOME
meant Lojban was real and here to stay (which is why whatever is issued
for a baseline needs to include some sort of hardbound, even if we have
cheaper softbound as well)
Can you produce a rewording of task 4 (or a statement of byfy
policy/procedures that would override the text) that eliminates
this apparent ordering of requirements that I don't think we
intended, and which would end the deadlock?
If you can do so, will you try it?
That's pretty much what my plan was anyways (see mail entitled
"That's *not* my formal proposal").
I understand (I may not have seen that when I wrote the above.) There
certainly was nothing in your essay that I saw, suggesting that you had
a different proposal in mind. I responded to what you said, and what
you asked people to read.
For resolving the issue, this seems inherently *much* smaller than
something like xorlo, and I'd likely go along with anything that
was clear and not egregiously weird. I doubt that there is enough
usage of .ainai for a usage or relearning argument to count for
much.
I agree, but Broca was blocking it simply because it was a change.
This was the core of the issue.
One person alone cannot block, per the baseline statement.
I don't know if it is good enough, but I imagine a couple dozen
people could do something similar for all the remaining cmavo in a
few weeks, even if some of them only do a couple of words.
I do to, but I have no idea where to find those people. I'll try
again after the mailing list is updated, though.
Post what you want people to do to Lojban List (and maybe even beginners
list). If the tasks are small, and require no expertise, and especially
if you have a Matt style reward system, which SEVERAL people agreed
would be very motivating, they'll emerge from the woodwork.
And maybe Matt will be willing to work at keeping them motivated and
recruiting new people, especially after his con ends. Some signs of
positive movement might encourage him again.
--
Bob LeChevalier lojbab@lojban.org www.lojban.org
President and Founder, The Logical Language Group, Inc.
--
You received this message because you are subscribed to the Google Groups "lojban" group.
To post to this group, send email to lojban@googlegroups.com.
To unsubscribe from this group, send email to lojban+unsubscribe@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/lojban?hl=en.