WikiDiscuss

Posted by rlpowell on Fri 19 of Nov, 2004 19:48 GMT posts: 14214

Two things:

"The tag modifies another selbri" — IMO, it modifies a bridi, not a selbri.

My parser can do ".i fi'o broda bo" without requiring fe'u.

-Robin

Posted by xorxes on Sat 20 of Nov, 2004 03:21 GMT posts: 1912

Robin:
> "The tag modifies another selbri" — IMO, it modifies a bridi, not a selbri.

The usual story has been that tags add a place to the selbri,
but I tend to agree it makes more sense to consider that they
modify a bridi. In fact that's what I do when expanding
for example {broda ki'u ko'a} as {ko'a krinu lo nu broda}.

> My parser can do ".i fi'o broda bo" without requiring fe'u.

What does it do with {i fi'o broda bo brode}?

mu'o mi'e xorxes

__
Do you Yahoo!?
The all-new My Yahoo! - Get yours free!
http://my.yahoo.com

Posted by rlpowell on Sat 20 of Nov, 2004 03:21 GMT posts: 14214

On Fri, Nov 19, 2004 at 02:42:55PM -0800, Jorge Llamb?as wrote:
>
> Robin:
> > "The tag modifies another selbri" — IMO, it modifies a bridi,
> > not a selbri.
>
> The usual story has been that tags add a place to the selbri, but
> I tend to agree it makes more sense to consider that they modify a
> bridi. In fact that's what I do when expanding for example {broda
> ki'u ko'a} as {ko'a krinu lo nu broda}.

OK>

> > My parser can do ".i fi'o broda bo" without requiring fe'u.
>
> What does it do with {i fi'o broda bo brode}?

First pass: text=( I=( i ) FIhO=( fi'o ) BRIVLA=( broda ) BO=( bo ) BRIVLA=( brode ) )
Second pass: text
text1
|- I: i
|- tenseModal
|- FIhO: fi'o
|- selbri6
|- BRIVLA: broda
|- BO: bo
|- BRIVLA: brode

Hmmm. That's interesting. If a sumti follows bo, that ends the
fi'o, but it will keep snarfing bo connected selbri into the fi'o.
Which is probably the right thing.

On the other hand:

Second pass: text
text1
|- I: i
|- tenseModal
| |- FIhO: fi'o
| |- BRIVLA: broda
| |- FEhU: fe'u
|- BO: bo
|- BRIVLA: brode

So yeah, fe'u is probably better.

(I've split up the morphology in the parser, by the way.)

-Robin

Re: BPFK Section: sumtcita Formants
Two things:

"The tag modifies another selbri" — IMO, it modifies a bridi, not a selbri.

My parser can do ".i fi'o broda bo" without requiring fe'u.

-Robin

On 11/4/07, arj <wikidiscuss@lojban.org> wrote:
>
> Why does the table in this overlap with other BPFK sections?

So as to have all "short fi'o-forms", i.e. all (single word) sumtcita
collected in one place. The table is still incomplete though.

>The columns for se, te, etc. are empty for the lower half of the table.

Yes, that's a quirk of the grammar that should be fixed. There's no
reason not to allow {se ca'u} along with {ti'a} for example, just as we
have {crane} along with {se trixe}.

>What is then the point in establishing a brivla equivalent for these sumtcita?

It's useful to have definitions of Lojban words in Lojban, not just in English.

mu'o mi'e xorxes

Posted by lojbab on Mon 05 of Nov, 2007 18:19 GMT posts: 162

Jorge Llambas wrote:
> On 11/4/07, arj <wikidiscuss@lojban.org> wrote:
>
>>Why does the table in this overlap with other BPFK sections?
>
>
> So as to have all "short fi'o-forms", i.e. all (single word) sumtcita
> collected in one place. The table is still incomplete though.
>
>
>>The columns for se, te, etc. are empty for the lower half of the table.
>
>
> Yes, that's a quirk of the grammar that should be fixed. There's no
> reason not to allow {se ca'u} along with {ti'a} for example, just as we
> have {crane} along with {se trixe}.
>
>>What is then the point in establishing a brivla equivalent for these sumtcita?
>
> It's useful to have definitions of Lojban words in Lojban, not just in English.

It may be worth noting that the original intent was that the "brivla
equivalents" of the BAI words were etymological mnemonics rather than
definitional. I understand that this intent may have been long
overtaken by usage.

ca'u, so far as I know, is still in a separate selma'o and not in BAI,
so it not a "quirk of the grammar" that causes it to be different from
BAI words in its grammar. Rather, it is a novelty to try to generalize
all of the one word sumtcita as necessarily being conceptually
equivalent because they can share that grammar role.

Still, I won't rule out accepting such a proposal at the end of the
game, provided that it is submitted as a distinct proposal, and not
something that is part of the documentation of the old baseline (and
preferably not as something that is adopted in the middle of the game
like xorlo).

lojbab

On 11/5/07, Robert LeChevalier <lojbab@lojban.org> wrote:
>
> ca'u, so far as I know, is still in a separate selma'o and not in BAI,
> so it not a "quirk of the grammar" that causes it to be different from
> BAI words in its grammar. Rather, it is a novelty to try to generalize
> all of the one word sumtcita as necessarily being conceptually
> equivalent because they can share that grammar role.

That's what I meant by "a quirk of the grammar". The meaning and
generalization of the use of SE from BAI to non-BAI tcita is obvious,
and it is thus just an accident of the historical development of the
language that it is not part of the grammar.

> Still, I won't rule out accepting such a proposal at the end of the
> game, provided that it is submitted as a distinct proposal, and not
> something that is part of the documentation of the old baseline (and
> preferably not as something that is adopted in the middle of the game
> like xorlo).

It is already proposed somewhere else, it is not mentioned in this
page. The only thing mentioned here is that for example {fi'o selcra}
and {ca'u} have the same meaning.

mu'o mi'e xorxes

Posted by arj on Tue 06 of Nov, 2007 08:05 GMT posts: 953

On Mon, Nov 05, 2007 at 09:47:28AM -0300, Jorge Llambías wrote:
> On 11/4/07, arj <wikidiscuss@lojban.org> wrote:
> >
> >What is then the point in establishing a brivla equivalent for these sumtcita?
>
> It's useful to have definitions of Lojban words in Lojban, not just in English.

No. Any word must have only one official definition. Anything else would be opening the door wide open for contradiction.

Granted, we can have such a list that is informative, not normative, but why should the BPFK make such a list?

--
Arnt Richard Johansen http://arj.nvg.org/
7% unemployment is no problem, according to 93% of the population.

Posted by arj on Tue 06 of Nov, 2007 08:07 GMT posts: 953

On Mon, Nov 05, 2007 at 03:34:32PM -0300, Jorge Llambías wrote:
> On 11/5/07, Robert LeChevalier <lojbab@lojban.org> wrote:
> >
> > ca'u, so far as I know, is still in a separate selma'o and not in BAI,
> > so it not a "quirk of the grammar" that causes it to be different from
> > BAI words in its grammar. Rather, it is a novelty to try to generalize
> > all of the one word sumtcita as necessarily being conceptually
> > equivalent because they can share that grammar role.
>
> That's what I meant by "a quirk of the grammar". The meaning and
> generalization of the use of SE from BAI to non-BAI tcita is obvious,
> and it is thus just an accident of the historical development of the
> language that it is not part of the grammar.

I agree with Bob here. Stag unification may be obvious, but it was never intended nor hinted to by the founders. It should be taken to its own proposal.

--
Arnt Richard Johansen http://arj.nvg.org/
7% unemployment is no problem, according to 93% of the population.

On 11/6/07, Arnt Richard Johansen <arj@nvg.org> wrote:
> On Mon, Nov 05, 2007 at 09:47:28AM -0300, Jorge Llambías wrote:
> > On 11/4/07, arj <wikidiscuss@lojban.org> wrote:
> > >
> > >What is then the point in establishing a brivla equivalent for these sumtcita?
> >
> > It's useful to have definitions of Lojban words in Lojban, not just in English.
>
> No. Any word must have only one official definition. Anything else would
> be opening the door wide open for contradiction.

And English has to be the language of official definitons of Lojban words?

> Granted, we can have such a list that is informative, not normative, but
> why should the BPFK make such a list?

Because it's useful. Any reason why not?

mu'o mi'e xorxes

On 11/6/07, Arnt Richard Johansen <arj@nvg.org> wrote:
>
> I agree with Bob here. Stag unification may be obvious, but it was never
> intended nor hinted to by the founders. It should be taken to its own proposal.

See:

<http://www.lojban.org/tiki/tiki-index.php?page=Internal%20grammar%20of%20tags>
http://www.lojban.org/tiki/tiki-index.php?page=Internal%20grammar%20of%20tags

There is no mention of this on the "BPFK section: sumtcita formants" page.

mu'o mi'e xorxes

Posted by arj on Tue 06 of Nov, 2007 17:28 GMT posts: 953

On Tue, Nov 06, 2007 at 11:46:15AM -0300, Jorge Llambías wrote:
> On 11/6/07, Arnt Richard Johansen <arj@nvg.org> wrote:
> > On Mon, Nov 05, 2007 at 09:47:28AM -0300, Jorge Llambías wrote:
> > > On 11/4/07, arj <wikidiscuss@lojban.org> wrote:
> > > >
> > > >What is then the point in establishing a brivla equivalent for these sumtcita?
> > >
> > > It's useful to have definitions of Lojban words in Lojban, not just in English.
> >
> > No. Any word must have only one official definition. Anything else would
> > be opening the door wide open for contradiction.
>
> And English has to be the language of official definitons of Lojban words?

Yes, unfortunately, since some BPFK commissioners cannot read Lojban at an advanced level.

> > Granted, we can have such a list that is informative, not normative, but
> > why should the BPFK make such a list?
>
> Because it's useful. Any reason why not?

In short, because it is not the BPFK's business to do that.

Quoting from the baseline document:

"It is intended that this effort ie. brief definitions of the cmavo shall take priority over work on other tasks charged to the byfy."

The rump BPFK (ie. after the completion of the language design) may sanction texts as baseline-compliant, but I can't see any mention in this document that the BPFK should provide teaching aids or the like.

--
Arnt Richard Johansen http://arj.nvg.org/
The savvy DXer will usually bet a pair of nickels he can pinpoint the DXCC
country of the "chopity-chow-pit chow-chow-pi-chow" even before he hears the
call sign. -John F. Lindholm, QST vol. 66 no. 3 p. 83

Posted by arj on Tue 06 of Nov, 2007 17:35 GMT posts: 953

On Tue, Nov 06, 2007 at 11:51:23AM -0300, Jorge Llambías wrote:
> On 11/6/07, Arnt Richard Johansen <arj@nvg.org> wrote:
> >
> > I agree with Bob here. Stag unification may be obvious, but it was never
> > intended nor hinted to by the founders. It should be taken to its own proposal.
>
> See:
>
> <http://www.lojban.org/tiki/tiki-index.php?page=Internal%20grammar%20of%20tags>
> http://www.lojban.org/tiki/tiki-index.php?page=Internal%20grammar%20of%20tags
>
> There is no mention of this on the "BPFK section: sumtcita formants" page.

Okay, I may be misusing terminology here, but it was never the intention that all sumtcita (with the possible exception of those in BAI) should be strictly equivalent with fi'o + brivla.

--
Arnt Richard Johansen http://arj.nvg.org/
<Nixon> XP kjennes ... sprengt.
<Nixon> Som om noe har eksplodert der.

On 11/6/07, Arnt Richard Johansen <arj@nvg.org> wrote:
> On Tue, Nov 06, 2007 at 11:46:15AM -0300, Jorge Llambías wrote:
> >
> > And English has to be the language of official definitons of Lojban words?
>
> Yes, unfortunately, since some BPFK commissioners cannot read Lojban
> at an advanced level.

Anyone for whom "fi'o broda" is too advanced should probably resign
as a BPFK commissioner.

> > > Granted, we can have such a list that is informative, not normative, but
> > > why should the BPFK make such a list?
> >
> > Because it's useful. Any reason why not?
>
> In short, because it is not the BPFK's business to do that.
>
> Quoting from the baseline document:
>
> "It is intended that this effort ie. brief definitions of the cmavo shall take
> priority over work on other tasks charged to the byfy."

These are brief definitions of cmavo, albeit not of the cmavo of this section
nor in the favourite language of the BPFK.

I still think brief definitions in Lojban are not a bad thing for the BPFK to
produce, and collecting all sumtcita in one table is a useful thing to have.

If the table is an impediment to voting on this section, I will move it
somewhere else when the vote is called.

mu'o mi'e xorxes

On 11/6/07, Arnt Richard Johansen <arj@nvg.org> wrote:
>
> Okay, I may be misusing terminology here, but it was never the
> intention that all sumtcita (with the possible exception of those
> in BAI) should be strictly equivalent with fi'o + brivla.

Whether it was intentional or not, the fact is that they are. Or are
you saying that some sumtcita have some particularity that makes
them unexpressible as "fi'o broda" (for some suitable broda)? One
may dispute whether the chosen broda is precisely the correct one
or not, but hardly the principle that it will be "fi'o broda" for some
broda.

mu'o mi'e xorxes

Posted by bancus on Tue 06 of Nov, 2007 21:00 GMT posts: 52

On 11/6/07, Jorge Llambías <jjllambias@gmail.com> wrote:
> > No. Any word must have only one official definition. Anything else would
> > be opening the door wide open for contradiction.
>
> And English has to be the language of official definitons of Lojban words?

FWIW, I certainly support that, ultimately, the official definitions
of Lojban be *in* Lojban. After all, we're looking for cultural
neutrality, are we not?

mu'o mi'e bancus

Posted by arj on Tue 06 of Nov, 2007 21:00 GMT posts: 953

On Tue, Nov 06, 2007 at 12:49:04PM -0800, Theodore Reed wrote:
> On 11/6/07, Jorge Llambías <jjllambias@gmail.com> wrote:
> > > No. Any word must have only one official definition. Anything else would
> > > be opening the door wide open for contradiction.
> >
> > And English has to be the language of official definitons of Lojban words?
>
> FWIW, I certainly support that, ultimately, the official definitions
> of Lojban be *in* Lojban. After all, we're looking for cultural
> neutrality, are we not?

It seems that we have a volunteer for translating the sections that are already done into Lojban.

--
Arnt Richard Johansen http://arj.nvg.org/
Public Notice as Required by Law: Any Use of This Product, in Any
Manner Whatsoever, Will Increase the Amount of Disorder in the
Universe. Although No Liability Is Implied Herein, the Consumer Is
Warned That This Process Will Ultimately Lead to the Heat Death of the
Universe. --Susan Hewitt and Edward Subitzky

Posted by arj on Tue 06 of Nov, 2007 21:15 GMT posts: 953

On Tue, Nov 06, 2007 at 05:46:11PM -0300, Jorge Llambías wrote:
> On 11/6/07, Arnt Richard Johansen <arj@nvg.org> wrote:
> >
> > Okay, I may be misusing terminology here, but it was never the
> > intention that all sumtcita (with the possible exception of those
> > in BAI) should be strictly equivalent with fi'o + brivla.
>
> Whether it was intentional or not, the fact is that they are. Or are
> you saying that some sumtcita have some particularity that makes
> them unexpressible as "fi'o broda" (for some suitable broda)? One
> may dispute whether the chosen broda is precisely the correct one
> or not, but hardly the principle that it will be "fi'o broda" for some
> broda.

I most certainly do dispute that.

What about composite tenses? What about ZA? What about VEhA? What about KI?

It will take some quite intricate machinery to make a rigorous equivalence between brivla and the imaginary journey model, and frankly I don't see the point.

--
Arnt Richard Johansen http://arj.nvg.org/
Public Notice as Required by Law: Any Use of This Product, in Any
Manner Whatsoever, Will Increase the Amount of Disorder in the
Universe. Although No Liability Is Implied Herein, the Consumer Is
Warned That This Process Will Ultimately Lead to the Heat Death of the
Universe. --Susan Hewitt and Edward Subitzky

On 11/6/07, Arnt Richard Johansen <arj@nvg.org> wrote:
> > One
> > may dispute whether the chosen broda is precisely the correct one
> > or not, but hardly the principle that it will be "fi'o broda" for some
> > broda.
>
> I most certainly do dispute that.
>
> What about composite tenses? What about ZA? What about VEhA?

What about them? "ve'a" is something like "fi'o canlu" and "va" something
like "fi'o tersei" (those are the ones I used in jbovlaste, but they
can certainly
be improved). For "puzu" we need a selbri that means "x1 happened long
before x2", and so on. Of course for long compounds it may be hard to
come up with a good selbri, but nobody is proposing to define every
possible compound anyway (obviously).

> What about KI?

I'll grant you that one, because it does have a special function not
included in "fi'o" or any selbri. The other special one, "cu'e", on the
other hand is easy: "fi'o mo".

> It will take some quite intricate machinery to make a rigorous
> equivalence between brivla and the imaginary journey model, and
> frankly I don't see the point.

And nobody is proposing to do that.

mu'o mi'e xorxes

Posted by lojbab on Tue 06 of Nov, 2007 23:38 GMT posts: 162

Jorge Llambas wrote:
>>It will take some quite intricate machinery to make a rigorous
>>equivalence between brivla and the imaginary journey model, and
>>frankly I don't see the point.
>
>
> And nobody is proposing to do that.
>
> mu'o mi'e xorxes

Actually, you are, if there is strict equivalence between the fi'o
selbri and the cmavo.

We interpret a series of tense cmavo according to the imaginary journey
model.

We have no semantic rules for interpreting a collection of fi'o modals
which are by implication adding new sumti into the predication based on
their order. fi'o modals currently are order-independent; tense cmavo
are not. Perhaps the imaginary journey model itself can be translated
into predications, but we are not obliged to solve that problem because
the existing paradigm is good enough for people to understand and use
the language even if it may be lacking in the sort of logical rigour
that exact equivalences may suggest.

lojbab

On 11/6/07, Robert LeChevalier <lojbab@lojban.org> wrote:
> Jorge Llambías wrote:
> >>It will take some quite intricate machinery to make a rigorous
> >>equivalence between brivla and the imaginary journey model, and
> >>frankly I don't see the point.
> >
> > And nobody is proposing to do that.
>
> Actually, you are, if there is strict equivalence between the fi'o
> selbri and the cmavo.

Dictionary definitions are never strict equivalences. You can't figure
out the composite meanings of compound tags out of the English
definitions of the components either. In any case, the "fi'o broda"
definition will be much closer than anything in English.

> We interpret a series of tense cmavo according to the imaginary journey
> model.

Yes.

> We have no semantic rules for interpreting a collection of fi'o modals
> which are by implication adding new sumti into the predication based on
> their order. fi'o modals currently are order-independent; tense cmavo
> are not.

That doesn't matter for the purposes of defining each sumtcita as a
"fi'o broda". It might well be that for compound sumtcita, the corresponding
selbri is not a simple function of the selbri corresponding to the
components. That doesn't mean that the compound sumtcita doesn't
have a correspondig selbri too.

> Perhaps the imaginary journey model itself can be translated
> into predications, but we are not obliged to solve that problem because
> the existing paradigm is good enough for people to understand and use
> the language even if it may be lacking in the sort of logical rigour
> that exact equivalences may suggest.

And nobody is working on that harder problem anyway (at least for now).

mu'o mi'e xorxes

Posted by JohnCowan on Wed 07 of Nov, 2007 00:51 GMT posts: 149

Jorge Llambías scripsit:

> That doesn't matter for the purposes of defining each sumtcita as a
> "fi'o broda". It might well be that for compound sumtcita, the corresponding
> selbri is not a simple function of the selbri corresponding to the
> components. That doesn't mean that the compound sumtcita doesn't
> have a correspondig selbri too.

--
Possession is said to be nine points of the law, John Cowan
but that's not saying how many points the law might have. cowan@ccil.org
--Thomas A. Cowan (law professor and my father)