WikiDiscuss

WikiDiscuss


Magic Words: left-to-right

posts: 14214
Use this thread to discuss the Magic Words: left-to-right page.
posts: 14214

I'm really not happy with BAhE. It's too hard to get out of using SI, and you can't return to it with SA. Furthermore, BAhE and UI are handled differently.

-Robin

posts: 14214

From the UI notes:

(including ZO-quoted words, LOhU-quotation words, ZEI-lujvo, BU-lerfu words, and ZOI-foreign words)

That's, like, child abuse of the word "word". How about UI affects "the previous word or extra-grammatical phrase"?

-Robin

posts: 1912

Robin:
> I'm really not happy with BAhE. It's too hard to get out of using SI, and
> you can't return to it with SA.

SA could return to it. That's really independent of left-to-right processing,
it's just a matter of defining what SA can see, and I don't see any reason
why SA shouldn't see it. I was confused about that before because I thought
BAhE was a magic word, but it is not, so SA should see it. What I think SA
should not see is words that are treated as "any-word" by magic words,
because those words are not really playing their role as selmaho members,
they are not fulfilling their function, and the point of the word after
SA is to replace a word that fills the same function.

As for SI, I don't find any compelling reason why BAhE should surrender
its natural (just because it comes first) precedence to it.

> Furthermore, BAhE and UI are handled
> differently.

That's inevitable, because one acts on what follows and the other on
what precedes. {zo} and {bu} present similar differences, for example,
for the same reason.

mu'o mi'e xorxes




__
Do you Yahoo!?
Check out the new Yahoo! Front Page.
www.yahoo.com




posts: 14214

On Tue, Nov 09, 2004 at 04:45:08PM -0800, Jorge Llamb?as wrote:
> Robin:
> > I'm really not happy with BAhE. It's too hard to get out of
> > using SI, and you can't return to it with SA.
>
> SA could return to it. That's really independent of left-to-right
> processing, it's just a matter of defining what SA can see, and I
> don't see any reason why SA shouldn't see it.

Because BAhE says "The construct behaves just as that word by
itself.", which implies to me that BAhE is gone for future
processing.

> > Furthermore, BAhE and UI are handled differently.
>
> That's inevitable, because one acts on what follows and the other
> on what precedes. {zo} and {bu} present similar differences, for
> example, for the same reason.

That's not what I'm talking about; BAhE gets absorbed, UI does not.

-Robin


posts: 1912

Robin:
> (including ZO-quoted words, LOhU-quotation words, ZEI-lujvo, BU-lerfu words,
> and ZOI-foreign words)
>
> That's, like, child abuse of the word "word".

"quoted word" is normal English for what ZO does.
"lerfu word" is even official for what BU does.
ZEI-lujvo may or may not be official (as in 'used in CLL', I'd have
to check) but it is normal lojbo usage for what ZEI does.

I grant that the other two are extensions, perhaps more justifiable
in the case of ZOI than in the case of LOhU.

> How about UI affects "the
> previous word or extra-grammatical phrase"?

I'm not sure extra-grammatical is the right word though.

mu'o mi'e xorxes





__
Do you Yahoo!?
Check out the new Yahoo! Front Page.
www.yahoo.com




posts: 14214

On Tue, Nov 09, 2004 at 04:53:43PM -0800, Jorge Llamb?as wrote:
> Robin:
> > How about UI affects "the previous word or extra-grammatical
> > phrase"?
>
> I'm not sure extra-grammatical is the right word though.

"grammar over-riding quote phrase"?

-Robin


posts: 1912


> Because BAhE says "The construct behaves just as that word by
> itself.", which implies to me that BAhE is gone for future
> processing.

Right. Yes, I think that's the right interpretation. After all,
the other magic words see a single word, so SA should too.

What do you rules do with {ba'e a bu}, {ba'e a zei da}?
Is it {*a* bu} or {*a bu*}? {*a* zei da} or {*a zei da*}?
Or is it a conflict of fighting over the same word?

> BAhE gets absorbed, UI does not.

Yes, that's a difference. But the reason is the left-to-right
processing. The only way for BAhE to do its thing AND not get
in the way of the word it modifies is by getting absorbed.
UI don't need to get absorbed because by the time they get to
do their thing, the word they modify has already done its own
thing. BAhE is very sui generis in the language.

mu'o mi'e xorxes




__
Do you Yahoo!?
Check out the new Yahoo! Front Page.
www.yahoo.com




posts: 14214

On Tue, Nov 09, 2004 at 07:14:25PM -0800, Jorge Llamb?as wrote:
>
> --- Robin Lee Powell wrote:
> > Because BAhE says "The construct behaves just as that word by
> > itself.", which implies to me that BAhE is gone for future
> > processing.
>
> Right. Yes, I think that's the right interpretation.

I don't.

> After all, the other magic words see a single word, so SA should
> too.

SA should be able to back to the last BAhE, IMO.

Hmmm. What does "sa zo ba'e" do in your interpretation? Or "sa zo"
in general, for that matter.

> What do you rules do with {ba'e a bu}, {ba'e a zei da}? Is it {*a*
> bu} or {*a bu*}? {*a* zei da} or {*a zei da*}?

The former in both cases.

> > BAhE gets absorbed, UI does not.
>
> Yes, that's a difference. But the reason is the left-to-right
> processing. The only way for BAhE to do its thing AND not get in
> the way of the word it modifies is by getting absorbed. UI don't
> need to get absorbed because by the time they get to do their
> thing, the word they modify has already done its own thing. BAhE
> is very sui generis in the language.

So is BU.

I really don't like SA not being able to back into BAhE, but it's a
minor thing.

-Robin


posts: 1912


> SA should be able to back to the last BAhE, IMO.

It's not a big deal. What exactly is the grammar
of SA anyway? Is it something like:

KOhA' = KOhA sa-KOhA / KOhA
sa-KOhA = SA KOhA / any-word sa-KOhA

?

> Hmmm. What does "sa zo ba'e" do in your interpretation? Or "sa zo"
> in general, for that matter.

Goes back to the last ZO. SA should see preceding magic words, it just
doesn't see the "any-word" part. (It shouldn't see any preceding SI,
SA or SU though, so {sa si}, {sa sa}, or {sa su} would take you to
the beginning of text.)

> > BAhE
> > is very sui generis in the language.
>
> So is BU.

After writing that, I realized that BAhE is not that sui generis
after all. SA has the same behaviour of BAhE with respect to
the word that follows: it works with it but does not change
its function. SA is of course more complicated than BAhE because
it also deals with preceding words, but in its dealings with the
following word they are the much same.

BU is just like ZEI in its dealings with the preceding word.

> I really don't like SA not being able to back into BAhE, but it's a
> minor thing.

I'd like to understand a bit better the formal grammar
of SA and BAhE before commiting one way or the other.

mu'o mi'e xorxes





__
Do you Yahoo!?
Check out the new Yahoo! Front Page.
www.yahoo.com




posts: 14214

On Wed, Nov 10, 2004 at 12:34:13PM -0800, Jorge Llamb?as wrote:
>
> --- Robin Lee Powell wrote:
> > SA should be able to back to the last BAhE, IMO.
>
> It's not a big deal.

What's not? SA not working with BAhE, or fixing it?

What *does* "mi za'e klama sa ba'e klama" do in LTR, anyways?

> What exactly is the grammar
> of SA anyway?

Umm, it's more than 600 lines of the current grammar.

> Is it something like:
>
> KOhA' = KOhA sa-KOhA / KOhA
> sa-KOhA = SA KOhA / any-word sa-KOhA
>
> ?

Pretty much, yeah, for every selma'o.

> > Hmmm. What does "sa zo ba'e" do in your interpretation? Or "sa
> > zo" in general, for that matter.
>
> Goes back to the last ZO. SA should see preceding magic words, it
> just doesn't see the "any-word" part. (It shouldn't see any
> preceding SI, SA or SU though, so {sa si}, {sa sa}, or {sa su}
> would take you to the beginning of text.)

<nod>

> > I really don't like SA not being able to back into BAhE, but it's a
> > minor thing.
>
> I'd like to understand a bit better the formal grammar
> of SA and BAhE before commiting one way or the other.

Ok. Let me know how I can help.

-Robin