On Thu, Apr 8, 2010 at 10:48 AM, Bob LeChevalier
<lojbab@lojban.org> wrote:
I have a separate response to Robin's essay in draft form. I think the following makes sense on its own, so I am posting it first, but it exists within the context of the other posting that I am not quite ready to post.
Matt Arnold wrote:
I have been drifting out of the community for the past couple of years
through discouragement. I used to have such grand dreams for the
project, but most of those plans turned out to be dependent on the
BPFK finishing Lojban.
I still have those grand dreams.
I thought work was proceeding apace, behind the scenes, and that the
only problem was generating enough work. It now turns out to have been
actively held back by a dispute between description of usage, and
prescription through centralized planning.
And it shouldn't have been, because until the byfy work is done, centralized planning is required.
> In a community this size, usage is a statistically insignificant sample.
On most of the issues remaining undecided, I would probably agree.
We have no known means to measure or prove anything about usage.
I'm not sure about this. Jorge, for example, argued that xorlo is pretty much transparent in terms of actual usage, a claim that could not be made if your statement were true.
It is also a self-contradicting authority.
The point is that it isn't an "authority". Most natural languages do quite well without any sort of "authority" deciding correctness. In the one noteworthy exception, French, the attempts to exercise that authority are largely ignored by actual speakers of the language.
"Let usage decide" was a recognition of this inevitability. It was a desired LONG TERM state. The byfy work of formalizing the baseline was understood as a PRECONDITION to building the usage base to the level where it becomes possible for "usage to decide".
Somehow or another, it got turned around so that people think that byfy cannot decide anything UNLESS usage exists sufficient to decide.
That certainly was not *my* intent. I wanted byfy to document the status quo, whatever it is, with any conflict resolved in favor of what people call the "conservative" prescription, UNLESS there is significant usage (or unless something was demonstrably "broken", such as the typos in CLL that are up for correction).
If there isn't usage enough, and the prescription is vague, then of course byfy was to be deciding things by fiat, in order to eliminate that vagueness where possible.
The bias for conservatism is intended as a bulwark against arbitrary change, and especially that based on aesthetics which is both transient and heavily English-biased, so long as there is more talking *about* Lojban than there is talking *in* Lojban.
Of course, for some issues, the best way to demonstrate the need for change is through usage. But Cowan and I had for several years demonstrated how you can have controlled change under the baseline system. For 4 years or so, every change to the formal grammar was documented in a change proposal which stated the status quo and the reason for change, and the resulting formal change to the documentation.
I had intended that byfy do the same, documenting the status quo, and accumulating formal proposals to be decided (separately from the documentation effort). But not one change proposal, even xorlo, has ever stated what would change in CLL or the "cmavo list" as a result of that proposal, even though those are the only two manifestations of the baseline that seem relevant. (The byfy product will presumably become a third, but we cannot formally change what hasn't yet been written).
> Usage in the wild has no mechanism with
which to resolve disputes with other usage.
In real life, it does, obviously, because you and I and others (including several non-native speakers) are communicating in English and making ourselves understood, and yet we are not speaking the same way Chaucer or Shakespeare did. The language has evolved, slowly, because "usage has decided".
But I agree that Lojban isn't at that point yet, and cannot reach that point without byfy completing its work.
>No wonder we were in a permanent bottleneck.
And we will remain so, unless we can complete the byfy project in some sort of mutually acceptable way, because people demand a prescription as a starting point.
(It is NOT clear that people will demand a later re-prescription after usage of the baselined language is well-established. I have tried to have us wait until then to decide and then have the decision made by the people actually using the language.)
We let talented people go to waste for nearly a decade. This has been
a disaster.
I don't think it has been a disaster. The community has continued to grow, and maybe now we finally have enough people to get the job done (I hope).
I do not accept it.
But you *have* accepted it up until now, and you are wearing one of the key hats in the organization. I stepped aside several years ago because people perceived that I was in the way of getting things done. Robin took over some of my hats, and has gotten an immense amount done. You've taken over order fulfillment and turned that around, for which I have much appreciation.
But if you don't accept the status quo in the byfy work, and think that talent is going to waste, what are *you* doing about it?
The "let usage decide" policy was put in place to prevent a recurrance
of the James Cooke Brown failure mode--
No. "letting usage decide", had nothing to do with JCB.
The policy of having a formal baseline, and being resistant to change imposed from on high was a response to one aspect of JCB's failure - the inability to get people to stop talking about the language (and endlessly proposing changes) and start using it. We've largely accomplished that much. People use the language, and ask "how to say it" questions.
> a failure mode which is no longer possible in the current environment.
I wish.
I can easily imagine a huge number of people deserting the project if, for example, a sufficiently radical orthography change was proposed by someone and adopted by the byfy. Whether that would kill the project is hard to say, but any mass exodus in response to an official change from 'on high' would be precisely "the James Cooke Brown failure mode".
> We need a policy that
mitigates the failure mode we're seeing, not the one that threatened
us decades ago. Now our failure mode is lack of decisiveness. It is
the responsibility of centralized authority to break the impasse that
happens when you let usage *totally* decide, rather than influence.
.oicai
Until byfy gets its job done, "let usage decide" is NOT the policy, so you and others should stop blaming it. byfy HAS the authority to *totally* decide.
http://www.lojban.org/tiki/Official+Baseline+Statement
is the policy.
The reference to "let usage decide" incorporated from the 1997 baseline statement is preceded by these words:
The following material from the 1997 baseline declaration is reiterated in the current statement, with a couple of notes as indicated. These statements apply after the language baseline is declared complete and the freeze is again imposed.
Note that second sentence. "Let usage decide" applies ONLY after byfy finishes.
Robin's plan is now well underway and it will succeed.
Which plan is that? He wrote an essay which is being debated. There is a proposal at the end that doesn't in fact deal with the current byfy paralysis, so far as I can tell, but rather talks about what will happen at some future time when byfy has actually gotten its original job done.
If Robin's essay needs a formal change to the policy, it will be up to the member's meeting to approve it, since they approved the current policy.
> I have confidence that the tide is overwhelmingly on our side.
On whose side? What other side is there?
In consensus politics, there shouldn't be "sides" in the sense you seem to be using the term. The goal is agreement, not winning a battle.
> That will
prevent me going away. But I will not remain in a project that is
committed to an indefinite holding pattern.
Why are we in an indefinite holding pattern? And what specifically do you want to *do* to change that pattern?
Redefining byfy's long-term role in the language community doesn't seem likely to change anything NOW. It won't get a cmavo dictionary done. It won't get CLL revised to incorporate xorlo and other decisions that have been made, etc.
And if we can't get people to do the necessary work, then stagnation will continue no matter how much the policy is fiddled with.