New soi's scope

[And Rosta]

I've just noticed that new soi "always takes top scope". Why is that? Not knowing the discussion that led to that choice, I incline to the view that in a reformed Lojban there would be a new syntax that syntacticizes the left-to-right scope rule (so that clause structure is binary branching, as in Xorban), in which case you'd expect soi to scope over what it precedes. Most if not all tags (e.g. tense) could be seen as abbreviations of soi phrases, and hence you'd expect soi phrases to have the scopal properties of tags.

--And.

[la durka]

The reason for making it that way (the term "scope leaper" has been used) was precisely to provide a way to escape the normal left-to-right scoping rule, to make certain things easier to express without rearranging the entire sentence structure. A similar proposed bridi modifier (xoi as currently defined in jbovlaste) is right-scoping like tags.

la selpa'i has some discussion in the second half of this post.

mu'o mi'e la durka

[TR NS]

On Sunday, September 28, 2014 11:17:43 PM UTC-4, la durka wrote:

The reason for making it that way (the term "scope leaper" has been used) was precisely to provide a way to escape the normal left-to-right scoping rule, to make certain things easier to express without rearranging the entire sentence structure. A similar proposed bridi modifier (xoi as currently defined in jbovlaste) is right-scoping like tags.

la selpa'i has some discussion in the second half of this post.

Thank you, so much. THAT is an explanation. I finally understand {voi}.

So, where as {poi} is an "identifier", narrowing down the possible identity of the referent, {voi} is a "qualifier". It doesn't restrict the identity, but selects the quality of the referent that makes it pertinent. Another translations besides simply "as" might be "in so far as".

[Alex Burka]

Well, careful... that post is proposing a redefinition of voi. (I don't understand its current meaning either.)

--
You received this message because you are subscribed to a topic in the Google Groups "lojban" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/lojban/y0lJZ_kOI2I/unsubscribe.
To unsubscribe from this group and all its topics, send an email to lojban+un...@googlegroups.com.
To post to this group, send email to loj...@googlegroups.com.
Visit this group at http://groups.google.com/group/lojban.
For more options, visit https://groups.google.com/d/optout.

[And Rosta]

That is a different proposal than the one for using /voi/ for {poi'i}, which is the one you hadn't understood.

It struck me as quite reasonable and understandable that you, newish to these topics, didn't understand that proposal, but not that you blamed the proposer for that.

--And.

[Gleki Arxokuna]

Most pages on the wiki are like that. Few people can understand them.

E.g. as of subjunctives at that time I had to copy all of the related pages into one Word document to read them and understand them which took several days.

They need simpler definitions and more easy real life examples.

Luckily, we have a new wiki on mw.lojban.org that is free from cherishing and preserving those old discussions as they are. We can edit them to make them more human friendly.

As of now I'm doing mostly reformatting work (since they are also not readable due to weird formatting).

The previous discussions will be saved and frozen forever in the tiki (the older wiki) so everyone can go there at any time if they are doing historical studies.

--
You received this message because you are subscribed to the Google Groups "lojban" group.
To unsubscribe from this group and stop receiving emails from it, send an email to lojban+un...@googlegroups.com.

[And Rosta]

On 29 Sep 2014 04:17, "la durka" <dur...@gmail.com> wrote:
>
> The reason for making it that way (the term "scope leaper" has been used) was precisely to provide a way to escape the normal left-to-right scoping rule, to make certain things easier to express without rearranging the entire sentence structure. A similar proposed bridi modifier (xoi as currently defined in jbovlaste) is right-scoping like tags.
>
> la selpa'i has some discussion in the second half of this post.

Ah, I see. Surely the basic form should be xoi, in selmaho XOI. If scope-leaping is desired -- and it should be admissible only if the scope-leaping rules are robustly defined, which is a tall order -- then surely it should provide variants of all scope-sensitive constituents, without arbitrarily privileging bridi relatives.

(Whether xoi must be in XOI rather than SEI is unclear, partly because the right criteria for selmahofellowhood are unclear (if selmaho means actual grammatical word-class rather than the sense it has in Lojban's current pseudogrammar).)

--And.

>
> mu'o mi'e la durka
>
> El domingo, 28 de septiembre de 2014 20:37:55 UTC-4, And Rosta escribió:
>>
>> I've just noticed that new soi "always takes top scope". Why is that? Not knowing the discussion that led to that choice, I incline to the view that in a reformed Lojban there would be a new syntax that syntacticizes the left-to-right scope rule (so that clause structure is binary branching, as in Xorban), in which case you'd expect soi to scope over what it precedes. Most if not all tags (e.g. tense) could be seen as abbreviations of soi phrases, and hence you'd expect soi phrases to have the scopal properties of tags.
>>
>> --And.
>

[TR NS]

On Monday, September 29, 2014 4:24:47 AM UTC-4, And Rosta wrote:

On 29 Sep 2014 04:45, "TR NS" <tran...@gmail.com> wrote:
>
> On Sunday, September 28, 2014 11:17:43 PM UTC-4, la durka wrote:
>> la selpa'i has some discussion in the second half of this post.
>
>
> Thank you, so much. THAT is an explanation. I finally understand {voi}.
>
> So, where as {poi} is an "identifier", narrowing down the possible identity of the referent, {voi} is a "qualifier". It doesn't restrict the identity, but selects the quality of the referent that makes it pertinent. Another translations besides simply "as" might be "in so far as".

That is a different proposal than the one for using /voi/ for {poi'i}, which is the one you hadn't understood.

Wait, that's a different "new voi" proposal? And both proposed by la selpa'i?

 

It struck me as quite reasonable and understandable that you, newish to these topics, didn't understand that proposal, but not that you blamed the proposer for that.

Am I reading that right? Instead of explaining, you are taking the time to rub my face in it? 

[And Rosta]

On 29 Sep 2014 09:59, "TR NS" <tran...@gmail.com> wrote:
>
> On Monday, September 29, 2014 4:24:47 AM UTC-4, And Rosta wrote:
>>
>>
>> On 29 Sep 2014 04:45, "TR NS" <tran...@gmail.com> wrote:
>> >
>> > On Sunday, September 28, 2014 11:17:43 PM UTC-4, la durka wrote:
>> >> la selpa'i has some discussion in the second half of this post.
>> >
>> >
>> > Thank you, so much. THAT is an explanation. I finally understand {voi}.
>> >
>> > So, where as {poi} is an "identifier", narrowing down the possible identity of the referent, {voi} is a "qualifier". It doesn't restrict the identity, but selects the quality of the referent that makes it pertinent. Another translations besides simply "as" might be "in so far as".
>>
>> That is a different proposal than the one for using /voi/ for {poi'i}, which is the one you hadn't understood.
>
>
> Wait, that's a different "new voi" proposal? And both proposed by la selpa'i?

That's my impression. But the "as"-like voi is mooted just in a blog post, not in any list of formal proposals.

>>
>> It struck me as quite reasonable and understandable that you, newish to these topics, didn't understand that proposal, but not that you blamed the proposer for that.
>
> Am I reading that right? Instead of explaining, you are taking the time to rub my face in it? 

No, just to point out that the ethos and etiquette of the internet is that one takes it upon oneself to educate oneself as far as one can, and seeing that, others will be willing to give one a leg up to the areas one's independent understanding cannot reach. One cannot reasonably expect others to have prepackaged everything for one in easily digestible form.

--And.

[TR NS]

On Monday, September 29, 2014 5:39:54 AM UTC-4, And Rosta wrote:

On 29 Sep 2014 09:59, "TR NS" <tran...@gmail.com> wrote:
>
> On Monday, September 29, 2014 4:24:47 AM UTC-4, And Rosta wrote:
>>
>>
>> On 29 Sep 2014 04:45, "TR NS" <tran...@gmail.com> wrote:
>> >
>>
>> That is a different proposal than the one for using /voi/ for {poi'i}, which is the one you hadn't understood.
>
>
> Wait, that's a different "new voi" proposal? And both proposed by la selpa'i?

That's my impression. But the "as"-like voi is mooted just in a blog post, not in any list of formal proposals.

"impression"? So you don't even know for sure? Can't you tell be reading the proposal?

 

>>
>> It struck me as quite reasonable and understandable that you, newish to these topics, didn't understand that proposal, but not that you blamed the proposer for that.
>
> Am I reading that right? Instead of explaining, you are taking the time to rub my face in it? 

No, just to point out that the ethos and etiquette of the internet is that one takes it upon oneself to educate oneself as far as one can, and seeing that, others will be willing to give one a leg up to the areas one's independent understanding cannot reach. One cannot reasonably expect others to have prepackaged everything for one in easily digestible form.

Then you shouldn't reasonably expect anyone to learn the language either. I don't believe the ethos and etiquette of the internet is to answer an earnest request for explanation with, that's for us to know and you to figure out. I also don't think it is much to ask that proposals be explained with at least a smidgen of plain language. As a programmer I know the importance of writing good comments. I'd sink like a rock if I explained to my coworkers, you "cannot reasonably expect me to have prepackaged everything for you in easily digestible form." 

[And Rosta]

On 29 Sep 2014 12:23, "TR NS" <tran...@gmail.com> wrote:
>
>
>
> On Monday, September 29, 2014 5:39:54 AM UTC-4, And Rosta wrote:
>>
>>
>> On 29 Sep 2014 09:59, "TR NS" <tran...@gmail.com> wrote:
>> >
>> > On Monday, September 29, 2014 4:24:47 AM UTC-4, And Rosta wrote:
>> >>
>> >>
>> >> On 29 Sep 2014 04:45, "TR NS" <tran...@gmail.com> wrote:
>> >> >
>> >>
>> >> That is a different proposal than the one for using /voi/ for {poi'i}, which is the one you hadn't understood.
>> >
>> >
>> > Wait, that's a different "new voi" proposal? And both proposed by la selpa'i?
>>
>> That's my impression. But the "as"-like voi is mooted just in a blog post, not in any list of formal proposals.
>
>
> "impression"? So you don't even know for sure? Can't you tell be reading the proposal?

I don't trust my judgement. There are aspects of Lojban I was never expert on, and in everything I am best rusty. I had largely tuned out for a dozen years; only recently has my interest been rekindled by a new wave of lojbanists, who know their shit.

>  
>>
>> >>
>> >> It struck me as quite reasonable and understandable that you, newish to these topics, didn't understand that proposal, but not that you blamed the proposer for that.
>> >
>> > Am I reading that right? Instead of explaining, you are taking the time to rub my face in it? 
>>
>> No, just to point out that the ethos and etiquette of the internet is that one takes it upon oneself to educate oneself as far as one can, and seeing that, others will be willing to give one a leg up to the areas one's independent understanding cannot reach. One cannot reasonably expect others to have prepackaged everything for one in easily digestible form.
>
>
> Then you shouldn't reasonably expect anyone to learn the language either.

I don't. I wouldn't recommend Lojban to anybody, unless they'd just get a kick out of learning it for the hell of it. Not everybody shares my view on that, but nobody expects new learners to follow and understand highly technical discussion.

I don't believe the ethos and etiquette of the internet is to answer an earnest request for explanation with, that's for us to know and you to figure out. I also don't think it is much to ask that proposals be explained with at least a smidgen of plain language. As a programmer I know the importance of writing good comments. I'd sink like a rock if I explained to my coworkers, you "cannot reasonably expect me to have prepackaged everything for you in easily digestible form." 

As an academic I know that at a stage when you're thrashing ideas around exploratorily it would be exhausting and unreasonable to expect interlocutors to make that discussion intelligible to those who aren't already able to understand it. Once you get to the publication stage, you set yourselves much higher standards of perspicuity, but you're still entitled to assume a certain level of prior knowledge from your readers. Not every academic publication is a beginners' textbook.

--And.

[Gleki Arxokuna]

2014-09-29 16:18 GMT+04:00 And Rosta <and....@gmail.com>:

As an academic I know that at a stage when you're thrashing ideas around exploratorily it would be exhausting and unreasonable to expect interlocutors to make that discussion intelligible to those who aren't already able to understand it. Once you get to the publication stage, you set yourselves much higher standards of perspicuity, but you're still entitled to assume a certain level of prior knowledge from your readers. Not every academic publication is a beginners' textbook.

This only resulted in others wasting their hours in trying to understand what was meant in those discussions. One could instead write an explanation. This almost never happened.

Academic publications imply institutions where those subjects are taught.

In Lojban no textbook can help you [by far].

[Alex Burka]

Yeah, the rules for scope-leaping could be more well-defined, though the concept seems pretty good to me. As it is, it seems if you have two or more {soi} clauses, they have equal scope at the top of the sentence, but I don't know if "equal scope" can be formalized.

How would you see a generalized scope-leaping mechanism working? Would it be some kind of UI to explicitly mark scope? One could argue that scope-leaping should be discouraged as confusing, so maybe {soi} should be the only way to do it.

Is there a succinct description of Xorban's binary branching scope somewhere? I'm not 100% sure what you mean by that.

With respect to selma'o, according to the LMW page, new-{soi} is in a new selma'o SOI (though still terminated by {se'u}), where the only grammatical difference from SEI is that SOI takes a whole sub-bridi rather than a degenerate one. And the selma'o XOI, as described for {xoi} in jbovlaste, seems to have exactly the same grammar. So I suspect the versions of SOI and XOI that we are currently discussing are actually the same selma'o (members of the same selma'o can clearly have different scope semantics; cf. KOhA).

mu'o mi'e la durka ku do'u

You received this message because you are subscribed to a topic in the Google Groups "lojban" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/lojban/y0lJZ_kOI2I/unsubscribe.
To unsubscribe from this group and all its topics, send an email to lojban+un...@googlegroups.com.

[And Rosta]

Alex Burka, On 29/09/2014 17:25:

> Yeah, the rules for scope-leaping could be more well-defined, though
> the concept seems pretty good to me. As it is, it seems if you have
> two or more {soi} clauses, they have equal scope at the top of the
> sentence, but I don't know if "equal scope" can be formalized.

They could be treated as logically conjoined by AND.

> How would you see a generalized scope-leaping mechanism working?

You could have a scope-leaping-marking UI, and then whatever the rule was going to be for soi can be ascribed to this UI instead. I shudder to think of how to formulate the rules to stop them leaking, and then all the sentences that would make no or uncertain sense containing this UI...

> One could argue that scope-leaping should be discouraged as
> confusing, so maybe {soi} should be the only way to do it.

It seems arbitrary to privilege soi. Better to start from a system with simple scope rules and then delicately investigate whether you can come up with workable devices to allow surface word order to follow discoursal semantics or whatever. (My own loglang happens to have utterly free word order, yet is utterly monoparsing.)

> Is there a succinct description of Xorban's binary branching scope
> somewhere? I'm not 100% sure what you mean by that.

The niceties of the analysis of the Xorban remain to be agreed on, but it goes like this: the Xorbo equivalent of Lojbo "lo broda lo brode cu brodi" is "la bbba le ccce dddake" [this from untrustworthy memory], which I'd see as having the following structure, where dependents follow and are indented from heads:

la
bbba
le
ccce
dddake

So each time you introduce a 'sumti' you have the gadri- or quantifier- analogue as head and it has two complements, one for the description/restriction and one for the rest of the proposition.

Lojban's left-to-right scope rule, if we treat Lojban as a variety of human language, must involve a syntax that works along these sorts of lines, so that a bridi is made up of nested binary branching phrases consisting of <sumti/tag, rest of bridi>, with the first element as head and the interpretation that the head scopes over the complement. I haven't worked out the details of that, but the essential insight is that "Scopes-Over" must be a fundamental syntactic relation, and you work backwards from that principle to work out what the syntactic structure must be.

--And.