WikiDiscuss

WikiDiscuss


On Fri, 11 Feb 2005 11:44:53 -0800, webmaster@lojban.org
<webmaster@lojban.org> wrote:
> !! Proposed Definition of zoi
>
> ;zoi (ZOI): Non-Lojban quotation. Creates a delimited
> non-Lojban quotation. The result treated as single Lojban word.
> "zoi" uses the following word as a delimiting word and quotes
> all further text until the same word is repeated.

Does "word" need to be defined for this purpose? For example, would
{zoi ganai this work? I expect not. ganai}

Or does it refer to "word" which is defined elsewhere?

> The delimiting word is separated from the quoted text by
> pauses, and must not be found in the written text or
> spoken phoneme stream.

Is this true for {la'o} quotes as well? It's not mentioned:

> "la'o" uses the following word as a delimiting word and quotes
> all further text until the same word is repeated. Both delimiter
> words lose their usual grammatical function, as do any Lojban
> words that might be inside the quote.


Also, how does {zoi ci mi klama lezarci le briju ci} parse? as "mi
klama lezarci le briju"? Or as {"mi klama lezar" le briju ci}? Or as
invalid text? (More generally, what if the delimiting "word" occurs
within the text but not with a pause in front?

What about {zoi zar mi klama le zarci le briju zar}? (where there is a
pause but the word is longer than the delimiting word)

mu'o mi'e .filip.
--

Philip Newton <philip.newton@gmail.com>



On Sunday 13 February 2005 11:26, Philip Newton wrote:
> Does "word" need to be defined for this purpose? For example, would
> {zoi ganai this work? I expect not. ganai}

I read that as "would \"nai this work? I expect not. \" nai", and thought
"Would it nai? What does that mean?"

> Also, how does {zoi ci mi klama lezarci le briju ci} parse? as "mi
> klama lezarci le briju"? Or as {"mi klama lezar" le briju ci}? Or as
> invalid text? (More generally, what if the delimiting "word" occurs
> within the text but not with a pause in front?
>
> What about {zoi zar mi klama le zarci le briju zar}? (where there is a
> pause but the word is longer than the delimiting word)

Based on the example "zoi gy gyrations gy" in chapter 19 of the Book, I think
{"mi klama lezarci le briju"} and {"mi klama le" ci le briju zar}.

phma
--
Sans lunettes, je ne distingue même pas les odeurs...
-Les Perles de la médecine


posts: 1912


> > ;zoi (ZOI): Non-Lojban quotation. Creates a delimited
> > non-Lojban quotation. The result treated as single Lojban word.
> > "zoi" uses the following word as a delimiting word and quotes
> > all further text until the same word is repeated.
>
> Does "word" need to be defined for this purpose? For example, would
> {zoi ganai this work? I expect not. ganai}
>
> Or does it refer to "word" which is defined elsewhere?

A word here is any member of a selmaho (including selmaho BRIVLA
and CMENE).

cmavo written together do not form a single word. {ganai} is
always two words.

The only exception that does count as one word is {y bu}, even when
written separately.

> > The delimiting word is separated from the quoted text by
> > pauses, and must not be found in the written text or
> > spoken phoneme stream.
>
> Is this true for {la'o} quotes as well? It's not mentioned:

Yes, the rules for {la'o} and for {zoi} are identical, because
the syntax only sees selmaho ZOI. The syntax generally only looks
at the selmaho, never at the members of the selmaho. The only
exception is precisely the delimiter words for ZOI, where the
word itself is considered.

> Also, how does {zoi ci mi klama lezarci le briju ci} parse? as "mi
> klama lezarci le briju"? Or as {"mi klama lezar" le briju ci}? Or as
> invalid text?

For the PEG parser, it's the first one. In this case there is no
reasonable alternative, because the ci in zarci can't be a separate
word without a pause, but there are more iffy cases such as:

> (More generally, what if the delimiting "word" occurs
> within the text but not with a pause in front?

For example {zoi ci mi klama reci zarci ci}? In this case, PEG
_will_ see the ci in reci as a separate word. (Whether or not this
is what we want is debatable.)

> What about {zoi zar mi klama le zarci le briju zar}? (where there is a
> pause but the word is longer than the delimiting word)

Again, zar in zarci cannot be a word, so it is not seen by the parser
as a word.

mu'o mi'e xorxes





__
Do you Yahoo!?
Yahoo! Mail - You care about security. So do we.
http://promotions.yahoo.com/new_mail


On Sunday 13 February 2005 15:00, Jorge "Llambías" wrote:
> --- Philip Newton wrote:
> > > ;zoi (ZOI): Non-Lojban quotation. Creates a delimited
> > > non-Lojban quotation. The result treated as single Lojban word.
> > > "zoi" uses the following word as a delimiting word and quotes
> > > all further text until the same word is repeated.
> >
> > Does "word" need to be defined for this purpose? For example, would
> > {zoi ganai this work? I expect not. ganai}
> >
> > Or does it refer to "word" which is defined elsewhere?
>
> A word here is any member of a selmaho (including selmaho BRIVLA
> and CMENE).

What about {zoi kau'i ka'ia ki'i kau'i}?

> cmavo written together do not form a single word. {ganai} is
> always two words.
>
> The only exception that does count as one word is {y bu}, even when
> written separately.

What does {zoi ybu abu boi denpa bu ybu} parse as?

> > (More generally, what if the delimiting "word" occurs
> > within the text but not with a pause in front?
>
> For example {zoi ci mi klama reci zarci ci}? In this case, PEG
> _will_ see the ci in reci as a separate word. (Whether or not this
> is what we want is debatable.)

The parser shouldn't be lexing text inside a zoi-quote; it should only be
looking for the delimiter. If it sees {ci} as a separate word in {reci}, what
does it do with {zoi ci mi qlawa reci zarci ci}?

> > What about {zoi zar mi klama le zarci le briju zar}? (where there is a
> > pause but the word is longer than the delimiting word)
>
> Again, zar in zarci cannot be a word, so it is not seen by the parser
> as a word.

What about {zoi cire cilce carce jarco cu jarco reci cilce carce}?

phma
--
lo nu punji lo sovda be lo cipnrkuku kukuku
zvati lo zdani be lo na'e cipnrkuku kuku


posts: 1912


> On Sunday 13 February 2005 15:00, Jorge "Llambías" wrote:
> > --- Philip Newton wrote:
> > > > ;zoi (ZOI): Non-Lojban quotation. Creates a delimited
> > > > non-Lojban quotation. The result treated as single Lojban word.
> > > > "zoi" uses the following word as a delimiting word and quotes
> > > > all further text until the same word is repeated.
> > >
> > > Does "word" need to be defined for this purpose? For example, would
> > > {zoi ganai this work? I expect not. ganai}
> > >
> > > Or does it refer to "word" which is defined elsewhere?
> >
> > A word here is any member of a selmaho (including selmaho BRIVLA
> > and CMENE).
>
> What about {zoi kau'i ka'ia ki'i kau'i}?

I should have said "including selmaho BRIVA, CMENE and UNASSIGNED-CMAVO.
Yes, kau'i will work as a delimiter.

> > cmavo written together do not form a single word. {ganai} is
> > always two words.
> >
> > The only exception that does count as one word is {y bu}, even when
> > written separately.
>
> What does {zoi ybu abu boi denpa bu ybu} parse as?

"abu boi denpa bu" is zoi-quoted. {y bu} behaves in all respects as
a member of selmaho BY.

> > > (More generally, what if the delimiting "word" occurs
> > > within the text but not with a pause in front?
> >
> > For example {zoi ci mi klama reci zarci ci}? In this case, PEG
> > _will_ see the ci in reci as a separate word. (Whether or not this
> > is what we want is debatable.)
>
> The parser shouldn't be lexing text inside a zoi-quote; it should only be
> looking for the delimiter. If it sees {ci} as a separate word in {reci}, what

> does it do with {zoi ci mi qlawa reci zarci ci}?

"mi qlawa re" zarci ci

{qlawa} is parsed as a non-lojban word, but the rest are all
lojban words.

The ci in {cireqlawa} would not be seen as a separate word.

> > > What about {zoi zar mi klama le zarci le briju zar}? (where there is a
> > > pause but the word is longer than the delimiting word)
> >
> > Again, zar in zarci cannot be a word, so it is not seen by the parser
> > as a word.
>
> What about {zoi cire cilce carce jarco cu jarco reci cilce carce}?

"re cilce carce jarco cu jarco re" cilce carce

mu'o mi'e xorxes





__
Do you Yahoo!?
Meet the all-new My Yahoo! - Try it today!
http://my.yahoo.com




posts: 1912


> The parser shouldn't be lexing text inside a zoi-quote; it should only be
> looking for the delimiter.

That would be much more complicated, because to recognize a delimiter
it would have to do the full syntactic parsing rather than merely
morphological. It's not enough to detect a zoi or a la'o, because
these could be affected themselves by other magic words.

It's probably doable, but I don't think it is worth the complications
it introduces. I prefer word lexing to be strictly separated from
syntactic parsing. The morphology algorithm should be able to break
any stream into words (including non-lojban words) without having any
clue as to the meaning or grammar of any of the words.

Any string separated by spaces that contains a unrecognized character
is considered a non-lojban word. More generally, any string separated
by spaces (or pauses in speech) that cannot be _fully_ broken into valid
Lojban words is considered a single non-lojban word.

mu'o mi'e xorxes




__
Do you Yahoo!?
The all-new My Yahoo! - What will yours do?
http://my.yahoo.com


posts: 14214

Ummm, this section has already been checkpointed. Did you not get
the mails?

On Sun, Feb 13, 2005 at 05:26:05PM +0100, Philip Newton wrote:
> On Fri, 11 Feb 2005 11:44:53 -0800, webmaster@lojban.org
> <webmaster@lojban.org> wrote:
> > !! Proposed Definition of zoi
> >
> > ;zoi (ZOI): Non-Lojban quotation. Creates a delimited
> > non-Lojban quotation. The result treated as single Lojban word.
> > "zoi" uses the following word as a delimiting word and quotes
> > all further text until the same word is repeated.
>
> Does "word" need to be defined for this purpose? For example,
> would {zoi ganai this work? I expect not. ganai}

"ga nai" is two words in Lojban. There is no confusion on this
point that I am aware of.

> > The delimiting word is separated from the quoted text by pauses,
> > and must not be found in the written text or spoken phoneme
> > stream.
>
> Is this true for {la'o} quotes as well? It's not mentioned:
>
> > "la'o" uses the following word as a delimiting word and quotes
> > all further text until the same word is repeated. Both
> > delimiter words lose their usual grammatical function, as do any
> > Lojban words that might be inside the quote.

Yes, it's true for {la'o} as well. That's a bit of a bug.

> Also, how does {zoi ci mi klama lezarci le briju ci} parse? as "mi
> klama lezarci le briju"? Or as {"mi klama lezar" le briju ci}? Or as
> invalid text? (More generally, what if the delimiting "word" occurs
> within the text but not with a pause in front?

It is ignored.

> What about {zoi zar mi klama le zarci le briju zar}? (where there
> is a pause but the word is longer than the delimiting word)

It is ignored.

"seperated by pauses" means on both sides.

-Robin

--
http://www.digitalkingdom.org/~rlpowell/ *** http://www.lojban.org/
Reason #237 To Learn Lojban: "Homonyms: Their Grate!"
Proud Supporter of the Singularity Institute - http://singinst.org/


posts: 14214

On Sun, Feb 13, 2005 at 03:26:01PM -0500, Pierre Abbat wrote:
> On Sunday 13 February 2005 15:00, Jorge "Llamb?as" wrote:
> > --- Philip Newton wrote:
> > > > ;zoi (ZOI): Non-Lojban quotation. Creates a delimited
> > > > non-Lojban quotation. The result treated as single Lojban
> > > > word. "zoi" uses the following word as a delimiting word and
> > > > quotes all further text until the same word is repeated.
> > >
> > > Does "word" need to be defined for this purpose? For example,
> > > would {zoi ganai this work? I expect not. ganai}
> > >
> > > Or does it refer to "word" which is defined elsewhere?
> >
> > A word here is any member of a selmaho (including selmaho BRIVLA
> > and CMENE).
>
> What about {zoi kau'i ka'ia ki'i kau'i}?

kau'i is a member of the psuedo-selmaho "CMAVO", and works with no
trouble.

> > cmavo written together do not form a single word. {ganai} is
> > always two words.
> >
> > The only exception that does count as one word is {y bu}, even
> > when written separately.
>
> What does {zoi ybu abu boi denpa bu ybu} parse as?

One big zoi quote.

> > > (More generally, what if the delimiting "word" occurs within
> > > the text but not with a pause in front?
> >
> > For example {zoi ci mi klama reci zarci ci}? In this case, PEG
> > _will_ see the ci in reci as a separate word. (Whether or not
> > this is what we want is debatable.)
>
> The parser shouldn't be lexing text inside a zoi-quote; it should
> only be looking for the delimiter.

That's not possible without lexing it.

> If it sees {ci} as a separate word in {reci}, what does it do with
> {zoi ci mi qlawa reci zarci ci}?

Umm, the same thing: zoi (ci mi qlawa reci) zarci *error*.

> > > What about {zoi zar mi klama le zarci le briju zar}? (where
> > > there is a pause but the word is longer than the delimiting
> > > word)
> >
> > Again, zar in zarci cannot be a word, so it is not seen by the
> > parser as a word.
>
> What about {zoi cire cilce carce jarco cu jarco reci cilce carce}?

The quote ends at the ci in reci.

You know, if you had so much to say about this section, it would
have been nice if you had said it sometime in the past *month* when
the section was open.

-Robin

--
http://www.digitalkingdom.org/~rlpowell/ *** http://www.lojban.org/
Reason #237 To Learn Lojban: "Homonyms: Their Grate!"
Proud Supporter of the Singularity Institute - http://singinst.org/


On Sunday 13 February 2005 15:55, Jorge "Llambías" wrote:
> --- Pierre Abbat wrote:
> > The parser shouldn't be lexing text inside a zoi-quote; it should only be
> > looking for the delimiter.
>
> That would be much more complicated, because to recognize a delimiter
> it would have to do the full syntactic parsing rather than merely
> morphological. It's not enough to detect a zoi or a la'o, because
> these could be affected themselves by other magic words.
>
> It's probably doable, but I don't think it is worth the complications
> it introduces. I prefer word lexing to be strictly separated from
> syntactic parsing. The morphology algorithm should be able to break
> any stream into words (including non-lojban words) without having any
> clue as to the meaning or grammar of any of the words.
>
> Any string separated by spaces that contains a unrecognized character
> is considered a non-lojban word. More generally, any string separated
> by spaces (or pauses in speech) that cannot be _fully_ broken into valid
> Lojban words is considered a single non-lojban word.

The problem with that is that a single foreign word could look like two Lojban
words each of which is also a foreign word. Thus {zoi gy i got to go to togo
gy} parses as "i got to go to to go".

phma
--
1 m = 3*3*5*7*47*44351/73/293339 * Cs133


posts: 1912


> > (More generally, what if the delimiting "word" occurs
> > within the text but not with a pause in front?
>
> It is ignored.

Not necessarily.

The rule is that anything that _can_ be recognized as a Lojban word,
will be.

If it is embedded in a non-lojban word, then it can't be recognized, but
if it is embedded in a Lojban string of words, and this string is
separated by pauses from non-lojban words, then it will be recognized.

The morphology algorithm knows nothing about the syntax of words. For
the morphology, zoi is just like loi, words that come after zoi
are trated like any other. The syntax parse will of course reject
any non-lojban word outside of a ZOI-quote, but for the morphology
non-lojban words are simply one more kind of word.

mu'o mi'e xorxes





__
Do you Yahoo!?
Yahoo! Mail - You care about security. So do we.
http://promotions.yahoo.com/new_mail


posts: 1912


> On Sunday 13 February 2005 15:55, Jorge "Llambías" wrote:
> > I prefer word lexing to be strictly separated from
> > syntactic parsing. The morphology algorithm should be able to break
> > any stream into words (including non-lojban words) without having any
> > clue as to the meaning or grammar of any of the words.
>
> The problem with that is that a single foreign word could look like two
> Lojban
> words each of which is also a foreign word. Thus {zoi gy i got to go to togo
> gy} parses as "i got to go to to go".

Oh, I don't mind how the parser ends up presenting the results.
A good rule of presentation might be to present everything within
zoi quotes just as it was received. But that's different from lexing.

The issue here is what counts as a word for delimiting purposes. You
could not use the word {to} or the word {go} as delimiters to quote
the foreign word "Togo", because the *morphology* pass will see
"Togo" as two lojban words, not as a non-lojban word.

mu'o mi'e xorxes





__
Do you Yahoo!?
The all-new My Yahoo! - What will yours do?
http://my.yahoo.com


On Sunday 13 February 2005 16:27, Jorge "Llambías" wrote:
> Oh, I don't mind how the parser ends up presenting the results.
> A good rule of presentation might be to present everything within
> zoi quotes just as it was received. But that's different from lexing.
>
> The issue here is what counts as a word for delimiting purposes. You
> could not use the word {to} or the word {go} as delimiters to quote
> the foreign word "Togo", because the *morphology* pass will see
> "Togo" as two lojban words, not as a non-lojban word.

Fine with me. I avoid using a delimiter that sounds or looks like some part of
what I'm quoting, even if it wouldn't be seen by the parser as the delimiter.
E.g. I wouldn't say {zoi tok toki pona tok} or {zoi ly Lythraceae ly}.

phma
--
..i toljundi do .ibabo mi'afra tu'a do
..ibabo damba do .ibabo do jinga
..icu'u la ma'atman.


posts: 1912


>
> Fine with me. I avoid using a delimiter that sounds or looks like some part
> of
> what I'm quoting, even if it wouldn't be seen by the parser as the delimiter.
>
> E.g. I wouldn't say {zoi tok toki pona tok} or {zoi ly Lythraceae ly}.

I think that's a good rule of style, even though the parser
woldn't be confused in those cases.

mu'o mi'e xorxes





__
Do you Yahoo!?
Meet the all-new My Yahoo! - Try it today!
http://my.yahoo.com




posts: 14214

On Sun, Feb 13, 2005 at 12:00:37PM -0800, Jorge Llamb?as wrote:
> --- Philip Newton wrote:
> > (More generally, what if the delimiting "word" occurs within the
> > text but not with a pause in front?
>
> For example {zoi ci mi klama reci zarci ci}? In this case, PEG
> _will_ see the ci in reci as a separate word. (Whether or not this
> is what we want is debatable.)

Actually, it's not debatable: the Quotations sectios says it must be
delimited by pauses, so this is a bug.

It shouldn't be a hard bug to fix.

My question is, does *any* space count as a pause, or only . ?

-Robin


On Sunday 13 February 2005 16:08, Robin Lee Powell wrote:
> You know, if you had so much to say about this section, it would
> have been nice if you had said it sometime in the past *month* when
> the section was open.

At the time, I was spending a bit too much time at work and going to work for
my brain to be fresh enough to tackle the intricacies of {zoi} and other
magic words. As the morphology algorithm I've thought about more, I should be
able to handle discussing it, though I will be moving sometime soon.

Anyway, Jorge and I agree that such examples are bad *style*, even if they're
not bad grammar.

phma
--
S Fa1>+/- !TM Ng- M K H T-- t? AT++ SY Te- SC- FO- D P !Tz E++ L
Am I Ha- hc-- FH+++ IP?


posts: 14214

On Sun, Feb 13, 2005 at 01:19:31PM -0800, Jorge Llamb?as wrote:
>
> --- Robin Lee Powell wrote:
> > > (More generally, what if the delimiting "word" occurs within
> > > the text but not with a pause in front?
> >
> > It is ignored.
>
> Not necessarily.
>
> The rule is that anything that _can_ be recognized as a Lojban
> word, will be.

By the morphology section, yes. But the section that actually
processes ZOI should (and can) know better.

-Robin


posts: 1912


> On Sun, Feb 13, 2005 at 01:19:31PM -0800, Jorge Llamb?as wrote:
> > The rule is that anything that _can_ be recognized as a Lojban
> > word, will be.
>
> By the morphology section, yes. But the section that actually
> processes ZOI should (and can) know better.

Yes, I suppose it's not that hard, you just need to force zoi-open
to be followed by a pause and zoi-close to follow a pause, and
zoi-word to accept words that would otherwise be delimiters
but are not preceded by pause. I don't care much one way or
the other really, if the fix is that simple.

mu'o mi'e xorxes






__
Do you Yahoo!?
Yahoo! Mail - You care about security. So do we.
http://promotions.yahoo.com/new_mail


On Sun, 13 Feb 2005 13:27:09 -0800 (PST), Jorge Llambías
<jjllambias2000@yahoo.com.ar> wrote:
>
> The issue here is what counts as a word for delimiting purposes. You
> could not use the word {to} or the word {go} as delimiters to quote
> the foreign word "Togo", because the *morphology* pass will see
> "Togo" as two lojban words, not as a non-lojban word.

I see. May not be that intuitive for people who don't understand
Lojban's morphology, but then they probably won't be using ZOI all
that much.

mu'o mi'e .filip.
--

Philip Newton <philip.newton@gmail.com>



On Sun, 13 Feb 2005 13:04:08 -0800, Robin Lee Powell
<rlpowell@digitalkingdom.org> wrote:
> Ummm, this section has already been checkpointed. Did you not get
> the mails?

I probably did.

But I didn't read all BPFK email since I'm not a member (or am I? I
thought I signed up for the list as an observer), so I didn't feel
responsibility for keeping up with what was happening--or for
understanding all that was coming, much of which seemed various
degrees of impenetrable to me.

I was mostly asking questions-for-understanding from the point of view
"I'm sure this is fine and covers all possible bases, but please
explain how X works in a bit more detail", not "I think this is wrong
because of FOO; let's get it fixed" questions.

Though this may not be the most helpful thing, nor the best time to do so.

mu'o mi'e .filip.
--

Philip Newton <philip.newton@gmail.com>



On Sun, 13 Feb 2005 13:08:37 -0800, Robin Lee Powell
<rlpowell@digitalkingdom.org> wrote:
> On Sun, Feb 13, 2005 at 03:26:01PM -0500, Pierre Abbat wrote:
> > If it sees {ci} as a separate word in {reci}, what does it do with
> > {zoi ci mi qlawa reci zarci ci}?
>
> Umm, the same thing: zoi (ci mi qlawa reci) zarci *error*.

Even without the pause?

> > What about {zoi cire cilce carce jarco cu jarco reci cilce carce}?
>
> The quote ends at the ci in reci.

and again.

mu'o mi'e .filip. noi na djica le nu jai fanza
--

Philip Newton <philip.newton@gmail.com>



posts: 14214

On Mon, Feb 14, 2005 at 06:20:15AM +0100, Philip Newton wrote:
> On Sun, 13 Feb 2005 13:04:08 -0800, Robin Lee Powell
> <rlpowell@digitalkingdom.org> wrote:
> > Ummm, this section has already been checkpointed. Did you not
> > get the mails?
>
> I probably did.
>
> But I didn't read all BPFK email since I'm not a member (or am I?
> I thought I signed up for the list as an observer), so I didn't
> feel responsibility for keeping up with what was happening--or for
> understanding all that was coming, much of which seemed various
> degrees of impenetrable to me.

  • blink*


Please accept apologies; you *are* indeed an observer. I forgot to
check.

Sorry.

> I was mostly asking questions-for-understanding from the point of
> view "I'm sure this is fine and covers all possible bases, but
> please explain how X works in a bit more detail", not "I think
> this is wrong because of FOO; let's get it fixed" questions.
>
> Though this may not be the most helpful thing, nor the best time
> to do so.

I'd much, much rather you questioned things than not; I was merely
mildly annoyed at the timing. :-)

Your point about la'o's definition has been noted and will be
handled later; the fact that zoi doesn't require pauses in the right
places in my PEG grammar will be fixed, probably this week.

-Robin

--
http://www.digitalkingdom.org/~rlpowell/ *** http://www.lojban.org/
Reason #237 To Learn Lojban: "Homonyms: Their Grate!"
Proud Supporter of the Singularity Institute - http://singinst.org/