Received-SPF: pass (google.com: domain of eyeonus@gmail.com designates 209.85.215.44 as permitted sender) client-ip=209.85.215.44;
MIME-Version: 1.0
In-Reply-To: <20120826203514.GV10684@stodi.digitalkingdom.org>
References: <20120826203514.GV10684@stodi.digitalkingdom.org>
Date: Sun, 26 Aug 2012 15:02:00 -0600
Message-ID: <CAMKwHj2Fp4qHBrk-kR6fvL5j756BwH379R0=Zsi5YhyP3bQVZA@mail.gmail.com>
Subject: Re: [lojban] carna, my turn.
From: Jonathan Jones <eyeonus@gmail.com>
To: lojban@googlegroups.com, lojban-list@lojban.org
Reply-To: lojban@googlegroups.com
Precedence: list
Mailing-list: list lojban@googlegroups.com; contact lojban+owners@googlegroups.com
Sender: lojban@googlegroups.com
Content-Type: multipart/alternative; boundary=e0cb4efe35362feb4704c831858f
X-Spam_score: -0.7
X-Spam_score_int: -6
X-Spam_bar: /

--e0cb4efe35362feb4704c831858f
Content-Type: text/plain; charset=ISO-8859-1

On Sun, Aug 26, 2012 at 2:35 PM, Robin Lee Powell <
rlpowell@digitalkingdom.org> wrote:

>
> (Jesus fuck this is long; it's *really* not as bad as it looks.
> It's certainly nowhere near as complicated as, say, the tense
> system.)
>
> This is my attempt to synthesize the thinking so far into an
> alternate definition for {carna} that shouldn't invalidate any
> current usage.
>
> Note that between this and the problem we've had with {farna}, I'm
> inclined to write up a full tutorial on directions, distances,
> locations, vectors, and so on, add it to the CLL, and have these
> words just have the basic definition and say "See chapter N".  But
> here it is anyway.
>
> The basic mechanism here, in programming terms, is polymorphism in
> the x2 and x4 places.
>
> Since this *looks* crazy complicated, I'm going to start with
> examples, to show that what comes out is quite reasonable.
>
>   {mi carna fi lo pritu}
>   I turn towards something to my right.  The distance is
>   unspecified and left up to context; it could mean you spin in
>   place many times, or that you turn just enough to face the house
>   on your right, or anything in between.
>
>   {mi carna fi lo berti}
>   I turn towards the north [again, distance unspecified].
>
>   {mi carna (mi/zo'e) lo pritu (li pi re mu/lo julra'o be li so
>   no/lo radno be li pi mu pi'i pai)}
>   I turn a quarter turn to the right.
>
>   {do carna do lo pritu li pi re mu}
>   You turned a quarter turn to your right.
>
>   {mi carna fo lo pritu}
>   I turn in some direction unspecified such that I end up facing
>   something to my right (i.e. that was on my right before I started
>   turning).  Since minimal distance is assumed, this is the shortest
>   way to say "I turn right" with almost no risk of being
>   misunderstood.
>
>   {mi carna fi lo pritu lo pritu}
>   I turn to the right.  This is the same as the previous example,
>   just being extra clear.
>
>   {do carna mi lo pritu}
>   From my point of view, you rotated clockwise [which, assuming two
>   standing humans facing each other, implies a cartwheel or similar]
>
>   {do carna do lo pritu be mi}
>   You turned towards something on my right.
>
>   {do carna le zunle tupyjanco be fi do ce'o le pritu tupyjanco be
>   fi do le zunle be mi li pi mu}
>   You rotated over your hips towards something to my left, one half
>   a turn, i.e. you are now standing on your hands or something.  A
>   situation that would make this perfectly reasonable is if you're
>   doing a flip into a handstand, and I'm your spotter, and I'm
>   standing to your right.  Note that in this case, because the thing
>   being turned towards isn't relative to the x1, the direction of
>   the x2 vector isn't important.  For the same reason, we don't know
>   if the gymnast did a front or back flip.
>
>   {do carna le zunle tupyjanco be fi do ce'o le pritu tupyjanco be
>   fi do lo zunle li pi mu}
>   Same thing, but now we know for certain that it was a back flip.
>   If you were willing to argue that the line through a person's hips
>   has a natural front, you could say {lo gapru} instead of {lo
>   zunle}, because if you imagine a person pointing an arm out of the
>   middle of their crotch, a back flip is equivalent to that arm
>   moving upwards.
>
>   {do carna mi lo zunle li pi mu}
>   If I'm your spotter and I'm on your right and you are sideways to
>   me facing to my right, this means exactly the same thing as the
>   previous sentence, and is what would be used in almost every case.
>
>   {le selkei cu carna fo so'i da}
>   The toy is spinning.
>
>   {le selkei cu carna mi lo pritu so'i da}
>   The toy is spinning clockwise from my point of view.
>
>   {le selkei cu carna (lo gapru/le drudi) lo pritu so'i da}
>   The toy is spinning clockwise if viewed from above.
>
>   {le selkei cu carna le loldi ce'o le drudi lo pritu so'i da}
>   The toy is spinning clockwise if viewed from above.
>
>   {le selkei cu carna le drudi ce'o le loldi lo pritu so'i da}
>   The toy is spinning clockwise if viewed from below.
>
> And here's the actual definition plus notes/tutorial.  This really
> is *fundamentally* complex, as we've discovered in this discussion.
> My previous version, which relied on the right hand rule, was very
> easy to define, but led to *horrendous* sentences that were
> extremely counter-intuitive.  In this version, I've sacrificed some
> initial learning curve for very easy use once you get used to it.
>
>   x1 turns about vector x2 towards direction x3, turning angular
>   distance / to face point x4
>

I think that's the best idea for fixing carna I've seen yet. I'm certainly
for making this change. I do think the actual wording of the definition
could use a little polish, but that's my only contention.


>   Notes: When reading these notes please give allowances for the
>   fact that facing in three dimensions is inherently complicated,
>   but note that the defaults make this actually very easy to use.
>   Each subsequent section gets more general and less likely to be
>   needed in practice.
>
>   Let's start with an x1 object that has both a top and a front,
>   i.e. you yourself.  Your default x2 is a vector running along your
>   body from the bottom to point out of the top of their head.
>   Imagine that you start pointing your hand in some direction, and
>   then move your pointing hand towards the x3.  The movement
>   describes a circle around you, perpendicular to the x2 vector;
>   we'll reference that later.  For this case in the x2 you could use
>   {zo'e}, or {mi}, or just skip it entirely; all of those options
>   are shorthand for "the vector running through me pointing out the
>   top of my head".  The x3 is anything you can point at.  We'll
>   cover x4 seperately.
>
>   This holds for any object with a top and a front; by default, if
>   the x2 is {zo'e} or the object itself, the x2 vector is from the
>   bottom pointing topwards, and the turning is from the front of the
>   object as though the object was a person describing eir own
>   movement.
>
>   If the object has a top but not the front, it's basically the
>   same, you just can't specify anything but {lo pritu} or {lo zunle}
>   for the x3.  If it has a front you could talk about the front
>   turning "towards the north" or whatever, but with no natural front
>   you can only talk about the direction of rotation in general
>   terms.
>
>   Note that in all such cases, which is things like "x1 carna x1 lo
>   pritu", these are equivalent to "x1 rotates clockwise if viewed
>   from above", and similarily for "lo zunle" and
>   "counter-clockwise".
>
>   If the x2 is a single object or place other than the x1, then draw
>   the vector from the x1 to the x2.  This leads to very natural ways
>   to talk about something you're looking at moving clockwise or
>   whatever; {le junla grana cu carna mi lo pritu} means "if the clock was
>   a person, with its head pointing towards me and its clock hand
>   sticking out in front of it, it would say that it is moving its
>   clock hand to the right", which happens to be what "clockwise"
>   is, as with viewing a person from above in the {do carna do lo
>   pritu} case.
>
>   Similarily, {le selkei cu carna lo gapru lo zunle} is "the toy
>   [perhaps a top] is spinning counter-clockwise when viewed from
>   above", because that's what it would look like if a human was in
>   place of that top, and standing up (i.e. head pointing to {lo
>   gapru}) and turning to what they called {lo zunle}.
>
>   For maximum specificity, you can specify the vector exactly, by
>   using "ce'o"; the vector points from the first element of the
>   "ce'o" to the last.  This is unlikely to be useful very often.
>
>   The x4 is one of a number, in which case it is the (possibly
>   fractional) number of complete turns made; an exact specification
>   of radians or degrees or whatever (i.e. {lo julra'o li so no}), in
>   which case that's how far the turn went; or an object or place, in
>   which case the turn was the minimum distance required to end with
>   the x1 facing that object or place.  That last is not OK if the x1
>   has no natural front.
>
>   Unless clearly specified otherwise, the x3 and the x4 from the point of
>   view of the circle of movement perpendicular to the x2 vector, as
>   everything was positioned before the movement started.  This is
>   almost always the same as saying that the x3 and x4 are from the
>   point of view of the x1 before the movement started.
>
> --
> http://singinst.org/ :  Our last, best hope for a fantastic future.
> .i ko na cpedu lo nu stidi vau loi jbopre .i danfu lu na go'i li'u .e
> lu go'i li'u .i ji'a go'i lu na'e go'i li'u .e lu go'i na'i li'u .e
> lu no'e go'i li'u .e lu to'e go'i li'u .e lu lo mamta be do cu sofybakni
> li'u
>
> --
> You received this message because you are subscribed to the Google Groups
> "lojban" group.
> To post to this group, send email to lojban@googlegroups.com.
> To unsubscribe from this group, send email to
> lojban+unsubscribe@googlegroups.com.
> For more options, visit this group at
> http://groups.google.com/group/lojban?hl=en.
>
>


-- 
mu'o mi'e .aionys.

.i.e'ucai ko cmima lo pilno be denpa bu .i doi.luk. mi patfu do zo'o
(Come to the Dot Side! Luke, I am your father. :D )

-- 
You received this message because you are subscribed to the Google Groups "lojban" group.
To post to this group, send email to lojban@googlegroups.com.
To unsubscribe from this group, send email to lojban+unsubscribe@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/lojban?hl=en.


--e0cb4efe35362feb4704c831858f
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

<div class=3D"gmail_quote">On Sun, Aug 26, 2012 at 2:35 PM, Robin Lee Powel=
l <span dir=3D"ltr">&lt;<a href=3D"mailto:rlpowell@digitalkingdom.org" targ=
et=3D"_blank">rlpowell@digitalkingdom.org</a>&gt;</span> wrote:<br><blockqu=
ote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc s=
olid;padding-left:1ex">
<br>
(Jesus fuck this is long; it&#39;s *really* not as bad as it looks.<br>
It&#39;s certainly nowhere near as complicated as, say, the tense<br>
system.)<br>
<br>
This is my attempt to synthesize the thinking so far into an<br>
alternate definition for {carna} that shouldn&#39;t invalidate any<br>
current usage.<br>
<br>
Note that between this and the problem we&#39;ve had with {farna}, I&#39;m<=
br>
inclined to write up a full tutorial on directions, distances,<br>
locations, vectors, and so on, add it to the CLL, and have these<br>
words just have the basic definition and say &quot;See chapter N&quot;. =A0=
But<br>
here it is anyway.<br>
<br>
The basic mechanism here, in programming terms, is polymorphism in<br>
the x2 and x4 places.<br>
<br>
Since this *looks* crazy complicated, I&#39;m going to start with<br>
examples, to show that what comes out is quite reasonable.<br>
<br>
=A0 {mi carna fi lo pritu}<br>
=A0 I turn towards something to my right. =A0The distance is<br>
=A0 unspecified and left up to context; it could mean you spin in<br>
=A0 place many times, or that you turn just enough to face the house<br>
=A0 on your right, or anything in between.<br>
<br>
=A0 {mi carna fi lo berti}<br>
=A0 I turn towards the north [again, distance unspecified].<br>
<br>
=A0 {mi carna (mi/zo&#39;e) lo pritu (li pi re mu/lo julra&#39;o be li so<b=
r>
=A0 no/lo radno be li pi mu pi&#39;i pai)}<br>
=A0 I turn a quarter turn to the right.<br>
<br>
=A0 {do carna do lo pritu li pi re mu}<br>
=A0 You turned a quarter turn to your right.<br>
<br>
=A0 {mi carna fo lo pritu}<br>
=A0 I turn in some direction unspecified such that I end up facing<br>
=A0 something to my right (i.e. that was on my right before I started<br>
=A0 turning). =A0Since minimal distance is assumed, this is the shortest<br=
>
=A0 way to say &quot;I turn right&quot; with almost no risk of being<br>
=A0 misunderstood.<br>
<br>
=A0 {mi carna fi lo pritu lo pritu}<br>
=A0 I turn to the right. =A0This is the same as the previous example,<br>
=A0 just being extra clear.<br>
<br>
=A0 {do carna mi lo pritu}<br>
=A0 From my point of view, you rotated clockwise [which, assuming two<br>
=A0 standing humans facing each other, implies a cartwheel or similar]<br>
<br>
=A0 {do carna do lo pritu be mi}<br>
=A0 You turned towards something on my right.<br>
<br>
=A0 {do carna le zunle tupyjanco be fi do ce&#39;o le pritu tupyjanco be<br=
>
=A0 fi do le zunle be mi li pi mu}<br>
=A0 You rotated over your hips towards something to my left, one half<br>
=A0 a turn, i.e. you are now standing on your hands or something. =A0A<br>
=A0 situation that would make this perfectly reasonable is if you&#39;re<br=
>
=A0 doing a flip into a handstand, and I&#39;m your spotter, and I&#39;m<br=
>
=A0 standing to your right. =A0Note that in this case, because the thing<br=
>
=A0 being turned towards isn&#39;t relative to the x1, the direction of<br>
=A0 the x2 vector isn&#39;t important. =A0For the same reason, we don&#39;t=
 know<br>
=A0 if the gymnast did a front or back flip.<br>
<br>
=A0 {do carna le zunle tupyjanco be fi do ce&#39;o le pritu tupyjanco be<br=
>
=A0 fi do lo zunle li pi mu}<br>
=A0 Same thing, but now we know for certain that it was a back flip.<br>
=A0 If you were willing to argue that the line through a person&#39;s hips<=
br>
=A0 has a natural front, you could say {lo gapru} instead of {lo<br>
=A0 zunle}, because if you imagine a person pointing an arm out of the<br>
=A0 middle of their crotch, a back flip is equivalent to that arm<br>
=A0 moving upwards.<br>
<br>
=A0 {do carna mi lo zunle li pi mu}<br>
=A0 If I&#39;m your spotter and I&#39;m on your right and you are sideways =
to<br>
=A0 me facing to my right, this means exactly the same thing as the<br>
=A0 previous sentence, and is what would be used in almost every case.<br>
<br>
=A0 {le selkei cu carna fo so&#39;i da}<br>
=A0 The toy is spinning.<br>
<br>
=A0 {le selkei cu carna mi lo pritu so&#39;i da}<br>
=A0 The toy is spinning clockwise from my point of view.<br>
<br>
=A0 {le selkei cu carna (lo gapru/le drudi) lo pritu so&#39;i da}<br>
=A0 The toy is spinning clockwise if viewed from above.<br>
<br>
=A0 {le selkei cu carna le loldi ce&#39;o le drudi lo pritu so&#39;i da}<br=
>
=A0 The toy is spinning clockwise if viewed from above.<br>
<br>
=A0 {le selkei cu carna le drudi ce&#39;o le loldi lo pritu so&#39;i da}<br=
>
=A0 The toy is spinning clockwise if viewed from below.<br>
<br>
And here&#39;s the actual definition plus notes/tutorial. =A0This really<br=
>
is *fundamentally* complex, as we&#39;ve discovered in this discussion.<br>
My previous version, which relied on the right hand rule, was very<br>
easy to define, but led to *horrendous* sentences that were<br>
extremely counter-intuitive. =A0In this version, I&#39;ve sacrificed some<b=
r>
initial learning curve for very easy use once you get used to it.<br>
<br>
=A0 x1 turns about vector x2 towards direction x3, turning angular<br>
=A0 distance / to face point x4<br></blockquote><div><br>I think that&#39;s=
 the best idea for fixing carna I&#39;ve seen yet. I&#39;m certainly for ma=
king this change. I do think the actual wording of the definition could use=
 a little polish, but that&#39;s my only contention.<br>
=A0</div><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;borde=
r-left:1px #ccc solid;padding-left:1ex">

=A0 Notes: When reading these notes please give allowances for the<br>
=A0 fact that facing in three dimensions is inherently complicated,<br>
=A0 but note that the defaults make this actually very easy to use.<br>
=A0 Each subsequent section gets more general and less likely to be<br>
=A0 needed in practice.<br>
<br>
=A0 Let&#39;s start with an x1 object that has both a top and a front,<br>
=A0 i.e. you yourself. =A0Your default x2 is a vector running along your<br=
>
=A0 body from the bottom to point out of the top of their head.<br>
=A0 Imagine that you start pointing your hand in some direction, and<br>
=A0 then move your pointing hand towards the x3. =A0The movement<br>
=A0 describes a circle around you, perpendicular to the x2 vector;<br>
=A0 we&#39;ll reference that later. =A0For this case in the x2 you could us=
e<br>
=A0 {zo&#39;e}, or {mi}, or just skip it entirely; all of those options<br>
=A0 are shorthand for &quot;the vector running through me pointing out the<=
br>
=A0 top of my head&quot;. =A0The x3 is anything you can point at. =A0We&#39=
;ll<br>
=A0 cover x4 seperately.<br>
<br>
=A0 This holds for any object with a top and a front; by default, if<br>
=A0 the x2 is {zo&#39;e} or the object itself, the x2 vector is from the<br=
>
=A0 bottom pointing topwards, and the turning is from the front of the<br>
=A0 object as though the object was a person describing eir own<br>
=A0 movement.<br>
<br>
=A0 If the object has a top but not the front, it&#39;s basically the<br>
=A0 same, you just can&#39;t specify anything but {lo pritu} or {lo zunle}<=
br>
=A0 for the x3. =A0If it has a front you could talk about the front<br>
=A0 turning &quot;towards the north&quot; or whatever, but with no natural =
front<br>
=A0 you can only talk about the direction of rotation in general<br>
=A0 terms.<br>
<br>
=A0 Note that in all such cases, which is things like &quot;x1 carna x1 lo<=
br>
=A0 pritu&quot;, these are equivalent to &quot;x1 rotates clockwise if view=
ed<br>
=A0 from above&quot;, and similarily for &quot;lo zunle&quot; and<br>
=A0 &quot;counter-clockwise&quot;.<br>
<br>
=A0 If the x2 is a single object or place other than the x1, then draw<br>
=A0 the vector from the x1 to the x2. =A0This leads to very natural ways<br=
>
=A0 to talk about something you&#39;re looking at moving clockwise or<br>
=A0 whatever; {le junla grana cu carna mi lo pritu} means &quot;if the cloc=
k was<br>
=A0 a person, with its head pointing towards me and its clock hand<br>
=A0 sticking out in front of it, it would say that it is moving its<br>
=A0 clock hand to the right&quot;, which happens to be what &quot;clockwise=
&quot;<br>
=A0 is, as with viewing a person from above in the {do carna do lo<br>
=A0 pritu} case.<br>
<br>
=A0 Similarily, {le selkei cu carna lo gapru lo zunle} is &quot;the toy<br>
=A0 [perhaps a top] is spinning counter-clockwise when viewed from<br>
=A0 above&quot;, because that&#39;s what it would look like if a human was =
in<br>
=A0 place of that top, and standing up (i.e. head pointing to {lo<br>
=A0 gapru}) and turning to what they called {lo zunle}.<br>
<br>
=A0 For maximum specificity, you can specify the vector exactly, by<br>
=A0 using &quot;ce&#39;o&quot;; the vector points from the first element of=
 the<br>
=A0 &quot;ce&#39;o&quot; to the last. =A0This is unlikely to be useful very=
 often.<br>
<br>
=A0 The x4 is one of a number, in which case it is the (possibly<br>
=A0 fractional) number of complete turns made; an exact specification<br>
=A0 of radians or degrees or whatever (i.e. {lo julra&#39;o li so no}), in<=
br>
=A0 which case that&#39;s how far the turn went; or an object or place, in<=
br>
=A0 which case the turn was the minimum distance required to end with<br>
=A0 the x1 facing that object or place. =A0That last is not OK if the x1<br=
>
=A0 has no natural front.<br>
<br>
=A0 Unless clearly specified otherwise, the x3 and the x4 from the point of=
<br>
=A0 view of the circle of movement perpendicular to the x2 vector, as<br>
=A0 everything was positioned before the movement started. =A0This is<br>
=A0 almost always the same as saying that the x3 and x4 are from the<br>
=A0 point of view of the x1 before the movement started.<br>
<span class=3D"HOEnZb"><font color=3D"#888888"><br>
--<br>
<a href=3D"http://singinst.org/" target=3D"_blank">http://singinst.org/</a>=
 : =A0Our last, best hope for a fantastic future.<br>
.i ko na cpedu lo nu stidi vau loi jbopre .i danfu lu na go&#39;i li&#39;u =
.e<br>
lu go&#39;i li&#39;u .i ji&#39;a go&#39;i lu na&#39;e go&#39;i li&#39;u .e =
lu go&#39;i na&#39;i li&#39;u .e<br>
lu no&#39;e go&#39;i li&#39;u .e lu to&#39;e go&#39;i li&#39;u .e lu lo mam=
ta be do cu sofybakni li&#39;u<br>
<br>
--<br>
You received this message because you are subscribed to the Google Groups &=
quot;lojban&quot; group.<br>
To post to this group, send email to <a href=3D"mailto:lojban@googlegroups.=
com">lojban@googlegroups.com</a>.<br>
To unsubscribe from this group, send email to <a href=3D"mailto:lojban%2Bun=
subscribe@googlegroups.com">lojban+unsubscribe@googlegroups.com</a>.<br>
For more options, visit this group at <a href=3D"http://groups.google.com/g=
roup/lojban?hl=3Den" target=3D"_blank">http://groups.google.com/group/lojba=
n?hl=3Den</a>.<br>
<br>
</font></span></blockquote></div><br><br clear=3D"all"><br>-- <br>mu&#39;o =
mi&#39;e .aionys.<br><br>.i.e&#39;ucai ko cmima lo pilno be denpa bu .i doi=
.luk. mi patfu do zo&#39;o<br>(Come to the Dot Side! Luke, I am your father=
. :D )<br>
<br>

<p></p>

-- <br />
You received this message because you are subscribed to the Google Groups "=
lojban" group.<br />
To post to this group, send email to lojban@googlegroups.com.<br />
To unsubscribe from this group, send email to lojban+unsubscribe@googlegrou=
ps.com.<br />

For more options, visit this group at http://groups.google.com/group/lojban=
?hl=3Den.<br />


--e0cb4efe35362feb4704c831858f--