From - Tue Dec 17 10:04:22 1996
Reply-To: Clark Nelson <Clark_Nelson@CCM.JF.INTEL.COM>
Date: Tue Dec 17 10:04:22 1996
Sender: Lojban list <LOJBAN@CUVMB.BITNET>
From: Clark Nelson <Clark_Nelson@CCM.JF.INTEL.COM>
Subject:      Re: Final warning about baseline
X-To:         LOJBAN@CUVMB.CC.COLUMBIA.EDU
To: John Cowan <cowan@LOCKE.CCIL.ORG>
X-UIDL: 306160709d8efc8097a35d4f9c760abe
X-Mozilla-Status: 0011
Content-Length:  2971
Message-ID: <rk7KEAVNUYD.A.n8B.m60kLB@chain.digitalkingdom.org>


As it happens, I was already in the process of composing a message to
you about this when you sent out the "final warning."

In the discussions we had accompanying the corrections I sent for the
BNF grammar, we talked about some corrections that should be made to
both grammars.  But none of those changes seem to have been made to
grammar.246 on ftp.access.digex.net.  Were the changes lost?  Abandoned?
Or are they just not yet available for public review?

Here are the issues I remember, in somewhat prioritized order.  I don't
recall you acknowledging all of them, so you may not think they're all
problems. And, at any point, feel free to jump in and say, "It's just
too late to make a change like that."

paragraph_10 is missing an alternative for fragment_20.

tag_491 should be optional before KE in gek_sentence_54.

There is also a long list of places in the grammar where
free_modifier_32 could be allowed, and maybe should be.

These are places where it is pretty clear that it isn't allowed solely
as a result of a mistake in a recent modification:
        after TUhE in statement_C_14 (610 should be 447)
        after ZOhU in prenex_30

These are recent additions to the grammar where it was probably never
considered whether free should be allowed:
        after PEhE in terms_A_81
        after CEhE in terms_B_82
        after VUhO in sumti_90
        after BIhE in MEX_A_311

These are places where free isn't allowed by the current grammar, even
though there are other places in the grammar where free is allowed after
the same selma'o:
        after KE in gek_sentence_54
        after KE in tanru_unit_B_152
        after MOI in tanru_unit_B_152
        after KE in operator_B_372
        after BO in selbri_F_136

And now, for those who really like to reach (:-), here are the places in
the grammar where free modifiers *could* be allowed without syntactic
problems (according to yacc), and there's no clear justification for why
they *aren't* allowed:
        after SOI in discursive_bridi_34
        after NUhI (both occurrences) in term_set_85
        after LA in sumti_G_97
        after LI in sumti_G_97
        after LA in description_110
        after LE in description_110
        after ZIhE in relative_clauses_121
        after GOI in relative_clause_122
        after NOI in relative_clause_122
        after JAI in tanru_unit_B_152
        after ME in tanru_unit_B_152
        after NUhA in tanru_unit_B_152
        after BE in linkargs_160
        after BEI in links_161
        after FUhA in MEX_310
        after PEhO in MEX_B_312
        after MAhO in MEX_operator_374
        after NAhU in MEX_operator_374
        after NIhE in operand_C_385
        after MOhE in operand_C_385
        after JOhI in operand_C_385

If you like, I can send you a modified yacc grammar that incorporates
all these changes.

--
Clark Nelson
clark_nelson@ccm.jf.intel.com