Remove "complete specifications"

[solpahi]

Some gismu definitions explicitly ask for so-called "complete sets" or "complete masses" in some sumti places. Get rid of them and allow incomplete specifications everywhere.

See also: http://www.lojban.org/tiki/BPFK+gismu+Section%3A+Complete+vs.+Incomplete+Specifications

[durka]

I don't really understand this issue. But I do have grep!

Here is what I found (29 gismu total):

$ pcregrep -rM 'complete(ly)?[\s\r\n]+(set|mass|specif)' gismu
gismu/b/bende.yaml:    - (x1 is a mass; x2 is a set completely specified); Also orchestra (= {zgibe'e},
gismu/b/bridi.yaml:      is a set completely specified); See also {sumti}, {fancu}.'
gismu/c/cipra.yaml:    - Also examination, proxy measure, validation; (a set in x3 must be completely
      specified).  See also {ciksi}, {troci}, {jarco}, {pajni}, {saske}.
gismu/c/ciste.yaml:      also elements (set completely specified); x4 systemic functions/properties.
gismu/c/cnano.yaml:    - Also mean, normal, usual; (x3 specifies the complete set).  See also {tcaci},
gismu/c/cuxna.yaml:      x3 (complete set).
gismu/f/fadni.yaml:      is complete set).  See also {cafne}, {rirci}, {kampu}, {lakne}, {tcaci}, {cnano}.'
gismu/f/fenso.yaml:    - (x2 if a set must be a complete specification); See also {cilta}, {jivbu}, {jorne},
gismu/j/jbini.yaml:    - x2 (a complete set, generally ordered) defines the bounds/limits/range for x1.  See
gismu/k/kampu.yaml:      of set x2 (complete set).
gismu/k/kancu.yaml:    - (x2 is complete set); See also {kanji}, {satci}, {merli}.
gismu/k/krili.yaml:    - 'x2: composition including x2, which need not be complete specification.  See
gismu/k/kruji.yaml:    - '[x2: composition including x2, which need not be complete specification]];
gismu/l/liste.yaml:    - Also roll, log.  (x2 is completely specified); (cf. {porsi}, {girzu}, {cmima}
gismu/p/pesxu.yaml:    - 'x2: composition including x2, which need not be complete specification.  See
gismu/p/porsi.yaml:    - Also (adjective:) x1 is serial.  (sets are completely specified); See also cmavo
gismu/r/ralju.yaml:      prime, (adverb:) chiefly, principally, mainly; (x2 is complete specification
gismu/r/rirci.yaml:    - (x3 is complete specification of set); (cf. {cizra}, {fadni}, {cafne}, {kampu},
gismu/r/rokci.yaml:    - 'x2: composition including x2, which need not be complete specification. See
gismu/s/sakta.yaml:      x3, which need not be complete specification.  See also {silna}, {titla}.'
gismu/s/sisku.yaml:    place structure: x1 seeks/searches/looks for property x2 among set x3 (complete
      specification of set).
gismu/s/slami.yaml:    - 'x2: composition including x2, which need not be complete specification.  See
gismu/s/slilu.yaml:      x3 (complete specification).
gismu/s/staku.yaml:      which need not be complete specification.  See also {kliti}.'
gismu/s/steci.yaml:      associated]; (x3 is completely specified set)]; See also {srana}, se {ponse},
gismu/s/stura.yaml:    - (x2, if a set, is completely specified); See also {ganzu}, {morna}, {ciste},
gismu/t/turni.yaml:    - x2 need not be complete specification of set of governed; reign/rule (= {noltru});

Did I miss any? Are we proposing to change all these definitions?

[solpahi]

la durka cu cusku di'e

I don't really understand this issue. But I do have grep!

Here is what I found (29 gismu total):
Did I miss any? Are we proposing to change all these definitions?

All of those will automatically change when set-places turn into plurals
(which is not really a change to the language, but to the documentation
of it. Nobody uses sets anymore)

[xorxes]

On Thu, Jun 5, 2014 at 12:20 PM, selpahi [email protected] wrote:

Some gismu definitions explicitly ask for so-called "complete sets" or
"complete masses" in some sumti places. Get rid of them and allow
incomplete specifications everywhere.

I'm not at all sure about that, not for all of them. "gunma" should still
mean "x1 consists of x2", not "x1 includes x2", which is "se cmima", there
is an important distinction between those two. Also conversely between
cmima "x1 is a member of x2" and "se gunma", "x1 constitutes x2". Also the
x1 of the "PA mei" predicates must be complete specifications, or else
these predicates lose all of their meaning. There may be specific cases
that merit revision, but in many cases I think the "complete specification"
requirement is correct.

[solpahi]

la .xorxes. cu cusku di'e

On Thu, Jun 5, 2014 at 12:20 PM, selpahi [email protected] wrote:

Some gismu definitions explicitly ask for so-called "complete sets" or
"complete masses" in some sumti places. Get rid of them and allow
incomplete specifications everywhere.

I'm not at all sure about that, not for all of them. "gunma" should still
mean "x1 consists of x2", not "x1 includes x2", which is "se cmima", there
is an important distinction between those two.

gunma's definition doesn't actually mention the word "complete", so it
doesn't fall under this change proposal :P

But yes, this kind of distinction is important, I agree.

Also conversely between
cmima "x1 is a member of x2" and "se gunma", "x1 constitutes x2". Also the
x1 of the "PA mei" predicates must be complete specifications, or else
these predicates lose all of their meaning.

There may be specific cases
that merit revision, but in many cases I think the "complete specification"
requirement is correct.

That's possible. I think that even if we find that they are indeed
justified, it's still useful to know that. This is a proposal from the
BPFK pages, so no matter the outcome, it's good to look into it, I think.

mi'e la selpa'i

[xorxes]

On Thu, Jun 5, 2014 at 7:06 PM, selpahi [email protected] wrote:

I think that even if we find that they are indeed
justified, it's still useful to know that. This is a proposal from the
BPFK pages, so no matter the outcome, it's good to look into it, I think.

I agree.

[lagleki]

pe'i lo re moi te sumti be zo gunma cu skicu lo ro se cmima

2014-06-05 19:20 GMT+04:00 selpahi [email protected]:

Some gismu definitions explicitly ask for so-called "complete sets" or
"complete masses" in some sumti places. Get rid of them and allow
incomplete specifications everywhere.

See also:

http://www.lojban.org/tiki/BPFK+gismu+Section%3A+Complete+vs.+Incomplete+Specifications

You can merge this Pull Request by running

git pull https://github.com/selpahi/gimste patch-17

Or view, comment on, or merge it at:

#31
Commit Summary

• Remove "complete specifications"

File Changes

• A global/remove "complete specifications"
  https://github.com/balningau/gimste/pull/31/files#diff-0 (3)

Patch Links:

• https://github.com/balningau/gimste/pull/31.patch
• https://github.com/balningau/gimste/pull/31.diff

—
Reply to this email directly or view it on GitHub
#31.