{cpedu} places vs. {minde, fraxu} etc.

20 views
Skip to first unread message

la gleki

unread,
Nov 4, 2012, 8:28:59 AM11/4/12
to loj...@googlegroups.com
cpedu - x1 requests/asks/petitions/solicits for x2 of/from x3 in manner/form x4.
minde - x1 issues commands/orders to x2 for result x3 (event/state) to happen; x3 is commanded to occur.

Although two gismu are obviously semantically close to each other they differ in the order of arguments.
Is it just a historical quirk?

I guess this is the "error" of {cpedu} cuz other similar gismu follow the structure of {minde}.

fraxu - x1 forgives x2 for event/state/activity x3.

Michael Turniansky

unread,
Nov 12, 2012, 3:34:54 PM11/12/12
to loj...@googlegroups.com
If you think of x2 as an object (which it can be), you will see that cpedu is actually in the lebna/cpacu/(te) dunda/benji/(te)mrilu, etc.... family.  The object changing hands being in the x2 spot, and the x3 spot being the other person involved in the transaction.  It's actually pikci which is the minde family.

       --gejyspa



--
You received this message because you are subscribed to the Google Groups "lojban" group.
To view this discussion on the web visit https://groups.google.com/d/msg/lojban/-/GN8TBZ74oIAJ.
To post to this group, send email to loj...@googlegroups.com.
To unsubscribe from this group, send email to lojban+un...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/lojban?hl=en.

Michael Turniansky

unread,
Nov 12, 2012, 3:35:16 PM11/12/12
to loj...@googlegroups.com
(vecnu, also)

la gleki

unread,
Dec 17, 2012, 9:28:43 AM12/17/12
to loj...@googlegroups.com


On Tuesday, November 13, 2012 12:34:56 AM UTC+4, gejyspa wrote:
If you think of x2 as an object (which it can be), you will see that cpedu is actually in the lebna/cpacu/(te) dunda/benji/(te)mrilu, etc.... family.  The object changing hands being in the x2 spot, and the x3 spot being the other person involved in the transaction.  It's actually pikci which is the minde family.

Yes, that's what bothers me.
Why do we have
{se ckire} being the addressee but
{te ciksi} being the addressee.

It's very hard to remember, non-intuitive etc.

Jonathan Jones

unread,
Dec 17, 2012, 10:22:07 AM12/17/12
to loj...@googlegroups.com
On Mon, Dec 17, 2012 at 7:28 AM, la gleki <gleki.is...@gmail.com> wrote:
On Tuesday, November 13, 2012 12:34:56 AM UTC+4, gejyspa wrote:
If you think of x2 as an object (which it can be), you will see that cpedu is actually in the lebna/cpacu/(te) dunda/benji/(te)mrilu, etc.... family.  The object changing hands being in the x2 spot, and the x3 spot being the other person involved in the transaction.  It's actually pikci which is the minde family.

Yes, that's what bothers me.
Why do we have
{se ckire} being the addressee but
{te ciksi} being the addressee.

It's very hard to remember, non-intuitive etc.

You'd have to ask the people who came up with the definitions in the first place, but I doubt they could answer you, either. i doubt any intention was payed to regularity among "families" of gismu, especially considering how many gismu have irregular places within their family. (I don't recall the exact one, but I believe the first irregular gismu I came across was one of the metric measurements, mass, maybe?)
 

       --gejyspa



On Sun, Nov 4, 2012 at 8:28 AM, la gleki <gleki.is...@gmail.com> wrote:
cpedu - x1 requests/asks/petitions/solicits for x2 of/from x3 in manner/form x4.
minde - x1 issues commands/orders to x2 for result x3 (event/state) to happen; x3 is commanded to occur.

Although two gismu are obviously semantically close to each other they differ in the order of arguments.
Is it just a historical quirk?

I guess this is the "error" of {cpedu} cuz other similar gismu follow the structure of {minde}.

fraxu - x1 forgives x2 for event/state/activity x3.

--
You received this message because you are subscribed to the Google Groups "lojban" group.
To view this discussion on the web visit https://groups.google.com/d/msg/lojban/-/GN8TBZ74oIAJ.
To post to this group, send email to loj...@googlegroups.com.
To unsubscribe from this group, send email to lojban+un...@googlegroups.com.

For more options, visit this group at http://groups.google.com/group/lojban?hl=en.

--
You received this message because you are subscribed to the Google Groups "lojban" group.
To view this discussion on the web visit https://groups.google.com/d/msg/lojban/-/xfuzoHfZXzgJ.

To post to this group, send email to loj...@googlegroups.com.
To unsubscribe from this group, send email to lojban+un...@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 )

Jonathan Jones

unread,
Dec 17, 2012, 10:23:45 AM12/17/12
to loj...@googlegroups.com
On Mon, Dec 17, 2012 at 8:22 AM, Jonathan Jones <eye...@gmail.com> wrote:
On Mon, Dec 17, 2012 at 7:28 AM, la gleki <gleki.is...@gmail.com> wrote:
On Tuesday, November 13, 2012 12:34:56 AM UTC+4, gejyspa wrote:
If you think of x2 as an object (which it can be), you will see that cpedu is actually in the lebna/cpacu/(te) dunda/benji/(te)mrilu, etc.... family.  The object changing hands being in the x2 spot, and the x3 spot being the other person involved in the transaction.  It's actually pikci which is the minde family.

Yes, that's what bothers me.
Why do we have
{se ckire} being the addressee but
{te ciksi} being the addressee.

It's very hard to remember, non-intuitive etc.

You'd have to ask the people who came up with the definitions in the first place, but I doubt they could answer you, either. i doubt any intention was payed to regularity among "families" of gismu, especially considering how many gismu have irregular places within their family. (I don't recall the exact one, but I believe the first irregular gismu I came across was one of the metric measurements, mass, maybe?)

It is my hope that at some very near future point this irregularity issue if addressed and fixed. Since it's always with the least used places, I don't believe making the irregular gismu regular would have much of an impact on historical works.

Jonathan Jones

unread,
Dec 17, 2012, 10:36:45 AM12/17/12
to loj...@googlegroups.com
On Mon, Dec 17, 2012 at 8:22 AM, Jonathan Jones <eye...@gmail.com> wrote:
<snip>
I don't recall the exact one, but I believe the first irregular gismu I came across was one of the metric measurements, mass, maybe?
<snip>

It was easy to find it: mitre has four places versus 3with all the other metric gismu, with the third being direction of measurement rather than standard like the others, and the standard being x4.

The non-metric length minli, is also irregular, in that it has its x3 and x4 in reverse order from the other non-metric, although excepting that it has the exact same place structure.

Interestingly, the metric and non-metric lengths are also irregular to each other, since the metric has a direction of measurement, and the non-metric does not.

la gleki

unread,
Dec 17, 2012, 10:56:59 AM12/17/12
to loj...@googlegroups.com


On Monday, December 17, 2012 7:23:45 PM UTC+4, aionys wrote:
On Mon, Dec 17, 2012 at 8:22 AM, Jonathan Jones <eye...@gmail.com> wrote:
On Mon, Dec 17, 2012 at 7:28 AM, la gleki <gleki.is...@gmail.com> wrote:
On Tuesday, November 13, 2012 12:34:56 AM UTC+4, gejyspa wrote:
If you think of x2 as an object (which it can be), you will see that cpedu is actually in the lebna/cpacu/(te) dunda/benji/(te)mrilu, etc.... family.  The object changing hands being in the x2 spot, and the x3 spot being the other person involved in the transaction.  It's actually pikci which is the minde family.

Yes, that's what bothers me.
Why do we have
{se ckire} being the addressee but
{te ciksi} being the addressee.

It's very hard to remember, non-intuitive etc.

You'd have to ask the people who came up with the definitions in the first place, but I doubt they could answer you, either. i doubt any intention was payed to regularity among "families" of gismu, especially considering how many gismu have irregular places within their family. (I don't recall the exact one, but I believe the first irregular gismu I came across was one of the metric measurements, mass, maybe?)

It is my hope that at some very near future point this irregularity issue if addressed and fixed. Since it's always with the least used places, I don't believe making the irregular gismu regular would have much of an impact on historical works.

It won't be that easy.
{mi cpedu do lo nu dunda lo jdini} is na smudra although will for sure be perfectly understood by everyone. You can't confuse abstractions with objects.
Just as {mi pikci do lo nu dunda lo jdini} is smudra.
However, such playing with places breaks the meaning of {se cpedu}. We'll have to avoid sumti-raising all the time in order to avoid ambiguity in it's worst sense of polysemy (you will never know whether your lojbo friend has a habit of changing place structure or not).

The only reasonable two alternative solutions would be 
1. Invent new gismu with the new place structure and slowly forget old ones (lojbanic corpus of texts can be automatically converted)
2. Split individual places into separate gismu (like "under conditions" is to be replaced by {va'o}) and invent new gismu for those fragments of older gismu. Or just ignore superfluous places whenever possible (if "by standard" is the last place, ignore it. If not invent a new gismu to replace the old one).

Well, enough herecy for today. Where is an appropriate quote from our shepherd?

2010.II.25 <rlpowell> Broca: To expand on my earlier comment about conservatism and you and stuff, I want Lojbanists to be active in moving the language towards awesome and away from things they hate.  I don't want to throw out the language we have, by any means (if I wanted that, I'd just go make a new one).  I want excellent Lojbanists like you to never again have to say "I hate that aspect of the language; it's dumb.  But I support it anyways."  I think ...


By the way, who is Broca?

la gleki

unread,
Dec 17, 2012, 11:00:18 AM12/17/12
to loj...@googlegroups.com


On Monday, December 17, 2012 7:36:45 PM UTC+4, aionys wrote:
On Mon, Dec 17, 2012 at 8:22 AM, Jonathan Jones <eye...@gmail.com> wrote:
<snip>
I don't recall the exact one, but I believe the first irregular gismu I came across was one of the metric measurements, mass, maybe?
<snip>

It was easy to find it: mitre has four places versus 3with all the other metric gismu, with the third being direction of measurement rather than standard like the others, and the standard being x4.

The non-metric length minli, is also irregular, in that it has its x3 and x4 in reverse order from the other non-metric, although excepting that it has the exact same place structure.

Interestingly, the metric and non-metric lengths are also irregular to each other, since the metric has a direction of measurement, and the non-metric does not.

Any examples of how {te mitre} is used? For me it's just {farna} inside gismu definition. Am I wrong?

Jonathan Jones

unread,
Dec 17, 2012, 12:32:55 PM12/17/12
to loj...@googlegroups.com
On Mon, Dec 17, 2012 at 9:00 AM, la gleki <gleki.is...@gmail.com> wrote:
On Monday, December 17, 2012 7:36:45 PM UTC+4, aionys wrote:
On Mon, Dec 17, 2012 at 8:22 AM, Jonathan Jones <eye...@gmail.com> wrote:
<snip>
I don't recall the exact one, but I believe the first irregular gismu I came across was one of the metric measurements, mass, maybe?
<snip>

It was easy to find it: mitre has four places versus 3with all the other metric gismu, with the third being direction of measurement rather than standard like the others, and the standard being x4.

The non-metric length minli, is also irregular, in that it has its x3 and x4 in reverse order from the other non-metric, although excepting that it has the exact same place structure.

Interestingly, the metric and non-metric lengths are also irregular to each other, since the metric has a direction of measurement, and the non-metric does not.

Any examples of how {te mitre} is used? For me it's just {farna} inside gismu definition. Am I wrong?

No idea. I've never used that place at all: {lo dinju cu galtu mitre li vono}
 

--
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 view this discussion on the web visit https://groups.google.com/d/msg/lojban/-/R_XjVyGn4soJ.

To post to this group, send email to loj...@googlegroups.com.
To unsubscribe from this group, send email to lojban+un...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/lojban?hl=en.

Jonathan Jones

unread,
Dec 17, 2012, 12:33:44 PM12/17/12
to loj...@googlegroups.com
On Mon, Dec 17, 2012 at 8:56 AM, la gleki <gleki.is...@gmail.com> wrote:
On Monday, December 17, 2012 7:23:45 PM UTC+4, aionys wrote:
On Mon, Dec 17, 2012 at 8:22 AM, Jonathan Jones <eye...@gmail.com> wrote:

On Mon, Dec 17, 2012 at 7:28 AM, la gleki <gleki.is...@gmail.com> wrote:
On Tuesday, November 13, 2012 12:34:56 AM UTC+4, gejyspa wrote:
If you think of x2 as an object (which it can be), you will see that cpedu is actually in the lebna/cpacu/(te) dunda/benji/(te)mrilu, etc.... family.  The object changing hands being in the x2 spot, and the x3 spot being the other person involved in the transaction.  It's actually pikci which is the minde family.

Yes, that's what bothers me.
Why do we have
{se ckire} being the addressee but
{te ciksi} being the addressee.

It's very hard to remember, non-intuitive etc.

You'd have to ask the people who came up with the definitions in the first place, but I doubt they could answer you, either. i doubt any intention was payed to regularity among "families" of gismu, especially considering how many gismu have irregular places within their family. (I don't recall the exact one, but I believe the first irregular gismu I came across was one of the metric measurements, mass, maybe?)

It is my hope that at some very near future point this irregularity issue if addressed and fixed. Since it's always with the least used places, I don't believe making the irregular gismu regular would have much of an impact on historical works.

It won't be that easy.
{mi cpedu do lo nu dunda lo jdini} is na smudra although will for sure be perfectly understood by everyone. You can't confuse abstractions with objects.
Just as {mi pikci do lo nu dunda lo jdini} is smudra.
However, such playing with places breaks the meaning of {se cpedu}. We'll have to avoid sumti-raising all the time in order to avoid ambiguity in it's worst sense of polysemy (you will never know whether your lojbo friend has a habit of changing place structure or not).

Not all of the irregular gismu have that same problem. Obviously some are more difficult to fix than others.
 
To view this discussion on the web visit https://groups.google.com/d/msg/lojban/-/I8JL2O5S0vkJ.

To post to this group, send email to loj...@googlegroups.com.
To unsubscribe from this group, send email to lojban+un...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/lojban?hl=en.
Reply all
Reply to author
Forward
0 new messages