Re: [lojban] {la .alis.} book

2015-05-29 1:05 GMT+03:00 Michael Everson <everson@evertype.com>:

On 27 May 2015, at 18:49, And Rosta <and.rosta@gmail.com> wrote:

>> A pause of indeterminate length is a phoneme?
>
> A pause of indeterminate length is phonetic realization of a phoneme.

As you wish. I love oddball linguistics. :-)

> Lojban's design is not at all optimized to make life easy for its learners or speakers. Its users could adopt conventions that eschew many elements of its design, in order to make life easier for its users, but that requires a consensus in usage and entails a de facto repudiation of those many elements eschewed.

Logical referents but semilogical structure? Okay.

In what way is it semi-logical? What is logical for you?

Lojban is logical mostly in a sense that it is easily parsable.

{sei} and bridi tails have different grammar. The selbri itself terminates {sei}-clause, thus the end of {sei}-clause is 100%-predictable. When the selbri is encountered it's the signal that {sei}-clause just ended.

>>> A rough analogy from English is the omissibility of _that_ from the start of complement clauses (e.g. "He knows (that) she is"): what lunatic would 'regularize' an English text by restoring every omitted _that_? And would you make overt *every* covert terminator?
>>
>> It seems to me that a [conjunction] is different from a state process.
>
> I don't understand what a 'state process' is. Like Lojban's terminators, clausal ('subordinating conjunction', 'complementizer') _that_ is usually but not always omissible without affecting sentence meaning.

Well, there’s a particle to start a paragraph (terminated by the next particle of the same kind), and one to start a sentence (terminated by the next particle of the same kind), and one to begin and end a citation (both particles used), and one to begin and end a metalingusitic interpolation (but the terminator is optional because you can use non-structural information to know when it terminates).

> (With regard to the _that_ that introduces relative clauses, the case for analysing bare relatives as containing a null _that_ is more complicated and hence perhaps weaker than the case for analysing bare complement clauses as containing a null _that_.)

I don’t see that as analogous, really.

>> If lu/li’u is regular, why shouldn’t sei/se’u be?
>
> sei/se'u is regular; it just is regular according to rules less simple than you would prefer.

Well, it does make it impossible to apply punctuation automatically, since the only way that one could do so is to have a script that would parse a variable-lengths string following “sei” to find the word of the selbri class. Not that the application of punctuation is a goal of Lojban.

I see, so you want to put European-like punctuation at the end of {sei}-clauses. This is certainly possible. Are you planning to publish any other books apart from Alice? If yes, then it's a matter of several patches to existing parsers to output such punctuation. Instead of " SEhU-clause?" we'll just have " (SEhU-clause / { ' – ' } ) ".

Michael Everson * http://www.evertype.com/

--
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+unsubscribe@googlegroups.com.
To post to this group, send email to lojban@googlegroups.com.
Visit this group at http://groups.google.com/group/lojban.
For more options, visit https://groups.google.com/d/optout.