[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [bpfk] BPFK work

Jorge Llambías wrote:

On Mon, Oct 11, 2010 at 3:29 PM, Robert LeChevalier <lojbab@lojban.org> wrote:


You are taking the fact that they do not parse as evidence that they are
multiple texts.  But formally, if it doesn't parse, it is simply incorrect,
even if pieces of it are correct separately.


Why do you think all our learning materials are offering incorrect
Lojban as examples?


Because we haven't been thinking about it.


The human inputter has to do a break-up, according to
informal rules.


A computer can also be programmed to do that, possibly using
statistical methods, but that's neither here nor there.
Computers don't follow informal rules, only rigidly programmed formal ones (unless one has programmed some kind of learning-behavior) 

> Yes, the

formal grammar needs to be fed the chunk to be tested for correctness.
In point of fact, we realized after CLL had been published, that many of the examples had never been tested with the parser. And most of those were single sentences.
And since the parser has never been complete and current, even that doesn't necessarily prove anything. 


One such informal rule is that a new speaker is a new text.


Excellent.


But that is an informal rule of pragmatics, and not part of the definition
of Lojban.  The proposals, I think, are proposals to override the informal
considerations of pragmatics.  I don't know if this is a good idea unless we
codify the rules of Lojban pragmatics.  But I'm willing to consider the
possibility.


So after all the brouhaha, we are in agreement after all.


I don't think so.


Do you see
now why I said that the proposed new word could not be in any of the
regular selma'o?


No.


The formal grammar will never get to see the new word,
If we formally codify the informal rules of Lojban pragmatics, then those rules become part of the formal grammar. 


it is only useful to the human (or computer) who will be
deciding what chunk to feed to the parser, but it has to be invisible
to the parser itself.
As far as I am concerned "the parser" refers to the entire "black box" program that takes input and produces a parse. Any "pre-parser" and indeed if relevant, "lexer" are just pieces of what I consider the "Lojban parser". You seem to be trying to treat different parts of a formal system as different and unrelated systems. This makes little sense to me. 

lojbab

--
You received this message because you are subscribed to the Google Groups "BPFK" group.
To post to this group, send email to bpfk-list@googlegroups.com.
To unsubscribe from this group, send email to bpfk-list+unsubscribe@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/bpfk-list?hl=en.