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

Re: [lojban] Balningau: The Great Update



This is too long, but when I try to shorten it, it gets longer.

On 5/24/2014 4:16 AM, Gleki Arxokuna wrote:
    You seem to have rather low standard for "encouraging and reaffirming".

    The Lojban community numbers in the several hundreds if not thousands.

    Maybe 1000 copies of CLL have been sold.  People who bought it
    expect that the language will not lightly be changed under them.

    You and your few others (relative to that 1000) of course do not
    respect that, which is your choice.

This certainly shows disrespect for CLL.

How so?

Until a new baseline is approved, the CLL (along with a few other documents) is the de jure baseline. It is of course the process of creating a new baseline that has been the problem.

But what's really interesting is how did xorlo proposals pass then?

Simply put, it was approved by the BPFK. My personal opinion is not decisive. Later, at a member's meeting, Robin and other members presented an argument that LLG stop teaching the old gadri system when most new people were learning the already approved new system. It was further argued that adoption of xorlo would not have drastic effect: little existing Lojban usage would be invalidated, but new people would understand how things in the language work differently.

 Was it a lost battle to you?

Yes and no.

Personally yes, especially in the sense that I have never understood xorlo. People have explained it to me, but it goes over my head, and I cannot apply it to any usage problems, and I forget any explanation by the time the topic comes up again.

I trust xorxes' claims that my old usage is still acceptable Lojban. (But I have found it very hard to motivate myself to use the language except in ad hoc conversation, not knowing how far my understanding deviates from the usage of the rest of the community. I want to return to my effort ancient effort translating an arbitrary chunk of Burton's Arabian Nights (I started this back in 1989); I'd also like to try some Pushkin, which would also exercise my studies of Russian. But anything I write won't use xorlo, unless I do it by accident.

On the other hand, it was a strong victory for the process I devised for language evolution (a conservative form of "let usage decide"). We took a long time arguing out the change, recording the arguments so that people could later examine them (adequate definitions of the status quo were produced in the process and argued out, so we achieved my requirement of documenting the existing language.) No changes were being made to CLL at that time.

My responsibilities as Founder and President are to the process and the community of users. I think the community won, and thus in a *professional* sense, I won.

(It also demonstrated to the community that I will in fact let strong community sentiment override my personal choices; I am not invested in asserting personal power as the guy who started the project - something JCB was too prone to doing with TLI Loglan, and historically the main reason for artificial languages failing to survive their founders).

Here is the minutes version of that official recognition of xorlo

http://www.lojban.org/tiki/LLG+2007+Annual+Meeting+Minutes
Robin Lee Powell moved that the membership recognize xorlo. Specifically, I
move that the membership assert that xorlo is more correct than the CLL at this
time, that whatever the BPFK come up with xorlo will be part of it, and that
new Lojbanists should be taught it as soon as possible.

...

1) The following procedures are added to the extant BPFK procedures:

    "Any proposal which at least half of the BPFK membership has
    voted on in a tentative vote with none voting against, may be
    submitted by the BPFKJ to the general membership as a possible
    piece of the zasni gafyfantymanri ("interim baseline", herein
    after referred to as the ZG).  Such a proposal requires a
    two-thirds majority of those voting to vote in favor of it at
    the general membership meeting in order to pass.

    Voting something into the ZG has the following effects:

    1. The proposal will be considered correct Lojban until such a
    time the complete new baseline is established and approved by
    the membership. Usage according to the CLL standard will not be
    considered incorrect, but usage according to the ZG will be
    preferred.

    2. The BPFK will recognize that such a vote indicates a desire
    by the membership for the proposal in question to be included,
    in modified form if necessary, when the new baseline is
    finished. Such a desire will not be considered binding in any
    way.

    3. The membership is encouraged to use the ZG standard in all
    pedagogical contexts, and in all Lojban conversation.

    The ZG will last only until the entire new baseline is written
    by the BPFK and approved by the membership."

In addition, xorlo was added to the zasni gafyfantymanri

To allow the above to be understood, I briefly recapitulate the process as I understand it (Robin has assumed dictatorial powers in order to get the ZG done, so in theory these procedures can be overridden):

1. The BPFK produces a collection of language description documents reflecting a new baseline, based on the current language, as you can find on the web site. Each document is approved separately in a series of "checkpoints", each of which requires consensus.

The original form of this presented by Nick Nicholas in
http://mw.lojban.org/index.php?title=Mini-dictionary (see especially 2.2)
http://mw.lojban.org/index.php?title=Mini-dictionary_To-do

Note that BPFK can consider change proposals to be submitted at any time, but the primary basis for any such discussion has to make reference to actual usage as found in the accumulated corpus of usage. The BPFK final result will nominally reflect "current usage" and not the 1997 language.

But we found that any contentious issue could not be decided finally. And actual work and completed checkpoints turned into a series of interminable discussions that I and others could not keep up with, much less provide timely rational counter-arguments and counter-proposals as required. The limited time of BPFK workers was occupied by trying to keep up with discussions of the work of very few others. Volunteers rarely finished anything, and tended to drop out because they couldn't spend the time needed to keep up.

I finally said that I personally would not agree to any change proposals at the time of checkpoint voting, but that I would also not exercise an anti-consensus veto to any checkpoint, reserving my personal approval for the package as a whole (at which point it might be easier to see any interactions and conflicts between proposals).

I got an informal commitment from Robin that the final baseline package would include a set of change pages to the published CLL, implicitly providing a simplified explanation for what, if anything, was being changed. (This effectively required that change pages for corrections of typos and the like be created, so that we have a good starting point for examining change proposals.), and that a reasonably long period of review would be allowed for this whole package (I suggested 6 months to a year) before a final vote. (Typically 1 week was the review time for checkpoint votes, and I couldn't get through all the discussions that quickly, much less produce any counter-argument or counter-proposal).

The need for that set of change pages is what drives the statement that we first document the status-quo-ante language (CLL with change pages for typos and detected errors), followed by approval of a new baseline which might include changes reflecting current usage (and possibly some changes that go beyond current usage, as reflected in experimental cmavo). At the time, xorlo and dotside were such changes. The 2007 vote merely recognized that xorlo would probably be accepted in the final package and that we shouldn't be teaching a contradictory set of gadri.

The longer the process has continued incomplete, the less has been the willingness to consider changes before getting SOMETHING done that can be called a baseline. If people had kept plugging away, we would have had a new baseline years ago, and proposals would be dealt with in a timely manner.

-------------

The following, by the way, is the official policy approved by the members which I am committed to supporting, by which I labelled this new group schismatic. We cannot as an organization cooperate with such a group acting outside of and independent of the BPFK.

Lojbanology:

MOVED: That any efforts by individuals or groups to develop a new
language or version of the existing language other than under the
direct auspices of the byfy is specifically disowned by LLG and shall
not be associated with LLG - PASSED without objection

Friendly Amendment: That membership in such an effort does not
disqualify one from membership in the LLG or the BPFK - PASSED
without objection


On the other hand, as noted in the amendment, individuals participating in such an effort are still entitled to a voice in the community (and if they seek membership, a vote). It is possible for a group such as this new one to "take over" LLG by adding enough of themselves as members. In effect, Robin gained his considerable power and authority by leading the last such anti-lojbab group. He then found out that I was correct on so many of the issues we disagreed on, especially the process questions that are to me more important than specific design points.


gleki:
xorlo invalidates large portions of CLL thus suggesting that those 1000 copies of CLL are to be burned.

I'm not into book-burning as a solution to any problem.

The commitment is not that the language never change, but rather that changes be carefully considered and not made unnecessarily. And well-documented. People with an old CLL and a set of formally approved change pages will know what the current language is. (I had hoped that the set of change pages would be small, but it has been 17 years rather than the originally intended 5).

Note also that the new baseline, when done, is intended to be the last language prescription from "on high". In the future, after the baseline is done, documents produced by LLG are intended to have a more descriptive function rather than prescriptive. A successor to BPFK might attempt to maintain the baseline documents in accordance with any changes in current usage, and might certify textbooks or other works as to whether they are consistent with the current official documents. But no one will have the authority to go off and revise the gismu list. They can choose to start using the language differently from the standard, but only if the bulk of the community chooses to adopt these variations over time in actual usage, would the changes have a chance be recognized in some official way. Hopefully the language community will be large enough that groups such as selpa'i leads will be too small to have much effect without years passing.

Do you have a solution to this problem? Because I feel that xorlo was de
facto a mistake (at least partially).

I can't argue, since I don't understand xorlo. Every time I try, I get mentally tangled up.

Shall we adapt xorlo to what is described in CLL?

The final baseline will (presumably) incorporate xorlo in CLL as a set of change pages (along with any other changes). But if there are problems with xorlo, then alternatives could be proposed to be included in the final baseline. Likely they would have to well-documented and supported with actual usage by some Lojbanists.

    My position requires me to oppose yours, and to consider your
    efforts schismatic.

They would certainly be schismatic if they invalidate many things
described in CLL.

The are schismatic because they are outside of the regular procedures for considering changes, procedures that are necessarily conservative out of respect for the people who have committed years of effort to get us to where we are today, as well as the need for language change to be slow and evolutionary rather than prescriptively revolutionary.

However, for now we don't know what they are suggesting since their work
hasn't even started.

It doesn't matter. They have explicitly denounced the BPFK process, which makes them schismatic.

If their intent was to produce a proposal to be considered by the BPFK, whenever it finally gets around to looking at gismu issues, and NOT to be promulgated in any official way until then (what people choose to do in their own conversations and writing is not "official").

I revised gimste myself injecting gua\spi style specifications of sumti
interactions and found very few problems in gimste. So I can't even
imagine what they are suggesting. Probably they don't know themselves.

Note that the original BPFK charter included the possibility of re-examining gismu, especially in light of actual usage. But it still has to go through the process, and years of non-accomplishment mean that anything unconnected to the cmavo problem is completely off of the priority charts. (I hope your revisions are kept distinct from the official baseline list, which should not be changed at all.)

Anyway, how can it be schismatic if the only reliable source for a
dictionary are definitions that we can learn from CLL?

The CLL is the baseline document. The collected BPFK checkpoints (the ZG) are (in the field of cmavo/selma'o definition) a draft proposed revision to that baseline. The draft is "official" in the sense of being produced by the officially recognized producer of new documents.

In the software development sense, the checkpoints are a pre-beta version. When the complete set is done, the ZG will be a beta version of the new baseline, released after several months of review as a new "product" hopefully with a new edition of CLL or something similar, and some sort of summary of changes in addition to a complete set of change pages for the old CLL.

Is there another official source apart from CLL? If yes then why wasn't
it printed?

Printing is expensive and hard to take back, and lasts a long time (as evidenced by the 16+ years CLL has been on the market without selling out - we did a large print run in part to keep the price down).

It also takes time to do it, time that is taken away from other tasks. While HTML and other smart-tagging have made it easier to format a book, it still takes a lot of time to do the markup, and even longer if you want an index (CLL took roughly a year to markup and index, me doing the markup and Nora doing the index, while John Cowan finished writing and made corrections. Probably at least a full time person-year, and it would have been at least twice that if we had been farming out tasks rather than doing it all ourselves).

We do have baseline gismu and cmavo and rafsi lists which are not in CLL, that were printed and published separately. These paper documents are in theory still available, although no one has ordered a copy in years, probably because the order form is some obscure place on the website:
http://www.lojban.org/old-style/publications/brochures/ordernet.html
(If I ever started getting orders, I would probably have to modify the order form because the prices do not reflect current postage costs).

Even after they were baselined, I maintained those documents personally, and sometimes corrected typos and made clarifications (this was before BPFK). But the membership rebuked me for doing so. If the document was baselined, then I didn't have the authority to change it without a vote (then a vote by the membership), which was impractical.

In theory, no one should have made ANY changes to those baseline documents as they appear on-line. Alas, I have little confidence that the people working on the wiki have followed that absolute constraint, so I keep an official baseline version on my own computer, and the BPFK will eventually have to formally approve any differences between that version and what is on-line.

(And I did produce a first working-draft dictionary file covering the gismu and some regular lujvo. But that isn't quite at the level of baseline, and is probably routinely ignored in favor of jbovlaste, which I have never found especially usable.)

Well, enough for now. Sorry this is so long.

lojbab


--
You received this message because you are subscribed to the Google Groups "lojban" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lojban+unsubscribe@googlegroups.com.
To post to this group, send email to lojban@googlegroups.com.
Visit this group at http://groups.google.com/group/lojban.
For more options, visit https://groups.google.com/d/optout.