Re: [lojban] xoi and new soi as bridi relative clause

2015-07-21 18:12 GMT+03:00 guskant <gusni.kantu@gmail.com>:

Le mardi 21 juillet 2015 14:28:18 UTC, la gleki a écrit :

2015-07-21 17:18 GMT+03:00 guskant <gusni...@gmail.com>:
It's not simplicity of grammar but similarity of form of description sumti and inserted bridi. If {sei}-clause could enclose a sentence, the grammar would be 23 bytes smaller. If we add {soi}-clause to "free", the structure and the usage will be very similar to {sei}-clause. If we had the structure of "SOI sentence SEhU" from the beginning of Lojban, I would have never used "SEI terms selbri SEhU".

Similarly, you can eliminate LE and move it to LOhOI saving even more rules.

Thank you for the advice. It seems apparently a nice idea, but I see more importance in LE-clause than SEI-clause, so I would rather keep it: A LE-clause automatically picks up the first argument from predication, while LOhOI can change the argument to pick up by {ke'a} afterwords. They have more different aspect compared with the similarity between SEI and new SOI.

By the way, I have no intention to force any change to ilmentufa or camxes, so don't be so much worry. You can continue being illogical and inconsistent as you like.

So simply put we have:

1. bridi = terms? CU_elidible selbri terms?

2. sei-clause = terms? CU_elidible selbri

3. sumti-tail = selbri

Of course, we ignore side constructs like quantifiers and relative clauses.

So are you asserting that it's better to align 1. and 2. thus losing backward compatibility with previous grammar of 2.?

Are you asserting that it's the best way to minify the grammar?

If you answer yes to both questions then I disagree since it's similarly possible to align 2. and 3. and then you wont lost any compatibility.

For you I aligned 2. and 3. in altatufa parser: http://mw.lojban.org/extensions/ilmentufa/altatufa-stodi.html

Here, we get

{sei gau mo}, {sei gau mi CU mo}, {sei mo}

and

{lo gau mo}, {lo gau mi CU mo}, {lo mo}

both sumti tail and sei-clause use the same bridi

We can also say that x1 of sumti tail has the default value of {zo'e} while {sei}-clause can have the default value e.g. {mi} so that

{sei gleki} = {sei mi gleki}

x1 then can be overriden explicitly

{sei do gleki}

and

{lo fa do gleki cu sipna} ~= {lo gleki noi du do cu sipna}

Of course, we will still be "illogical and inconsistent" since {lo do gleki} would mean not what one could expect after learning {sei do gleki} and 1. is not aligned but you yourself crossed out this reasoning by saying "I see more importance in LE-clause than SEI-clause, so I would rather keep it"

It's a road to nowhere since in the end one can come and say "There is still the lack of logic here, I assert that usage is insignificant so let's change this rule".

--
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.