From a.rosta@dtn.ntl.com Tue Jul 17 15:26:24 2001 Return-Path: X-Sender: a.rosta@dtn.ntl.com X-Apparently-To: lojban@yahoogroups.com Received: (EGP: mail-7_2_0); 17 Jul 2001 22:26:23 -0000 Received: (qmail 33465 invoked from network); 17 Jul 2001 22:24:54 -0000 Received: from unknown (10.1.10.142) by l9.egroups.com with QMQP; 17 Jul 2001 22:24:54 -0000 Received: from unknown (HELO relay3-gui.server.ntli.net) (194.168.4.200) by mta3 with SMTP; 17 Jul 2001 22:24:54 -0000 Received: from m156-mp1-cvx1b.bir.ntl.com ([62.255.40.156] helo=andrew) by relay3-gui.server.ntli.net with smtp (Exim 3.03 #2) id 15Md1s-0007C4-00 for lojban@yahoogroups.com; Tue, 17 Jul 2001 23:09:21 +0100 To: Subject: RE: [lojban] registry of experimental cmavo Date: Tue, 17 Jul 2001 23:24:02 +0100 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2910.0) Importance: Normal X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2615.200 In-Reply-To: From: "And Rosta" X-Yahoo-Message-Num: 8691 Jorge: > la and cusku di'e > >List of experimentals on my wishlist: > > > >1. In BAhE: next word begins nonstandard construction. > > I don't think it could be in BAhE. You want something that > will stop the parser from parsing what follows, but we don't > want {ba'e} to do that. OK. Maybe it could be in SA? You insert it before the word that begins the nonstandardness, and the parser then ignores the sentence. This would require that "sentence-1 i sa sentence2" deletes sentence2, not sentence1, and that "sentence1 i blahblah sa blahblah sa blahblah likewise does not delete sentence1. > >2. bridi-to-sumti converter > > Would something in selma'o LU do it? It could take more than a > bridi, but I often find that I want to put more than a bridi > inside a du'u. What are examples of things more than a bridi that you want inside a du'u? At any rate, I hadn't forgotten this good suggestion of yours, or its near-equivalent, viz. just using "la'e lu", but I have two reservations about it. First is the near-mandatoriness of LIhU; a pure bridi-to-sumti converter would be terminable with an often omissible KU or KEI. Second, the Refgram says: # The implicit quantifier for all types of quotation is "su'o" # [...], because quotations are analogous to "lo" descriptions; # they refer to things which actually are words or sequences of # words. Now I very much don't want something that denotes a class of texts/utterances. I would want something unquantifiable that denotes a sentence (or other abstract linguistic form). In effect, I want something that is semantically like LI rather than like LO. (This is how I feel about all Lojban's quoting devices, btw.) I could live with a grammar change that allows "li" to have as complement lu/zo and ideally sumti-tails, so as to allow "li ka", "li du'u". > >3. bridi-to-ROI converter, expressing "p in PA possible worlds in > >which q" > > How about just something in ROI, say {xu'e}, then we can say > something like {

roru'e le du'u } (or your new LU instead > of {le du'u}). This way we don't need any change in the grammar. Good suggestion. > Besides, what would your bridi-roi tag as a sumti tcita? How do you say "Every time I say goodbye, I cry a little"? > >4. takes cmevla as complement and yields da-series KOhA, allowing > >bare cmevla to function as da-series KOhA thereafter. > > This can be done already with cmene: > {su'o da goi la alf ro da goi la bet su'o da goi la gam zo'u}. I didn't realize that one could reuse "da" in that way. And presumably "ko'a". So my desideratum is a mere abbreviatory device. > >5. takes cmevla as complement and yields ko'a-series KOhA, allowing > >bare cmevla to function as ko'a-series KOhA thereafter. > > What you really seem to be after is allowing bare cmevla to > function as sumti. I think that would require changes in the > resolution algorithm, and if that were an option I would go > for more fundamental changes than that. Agreed. > >6. In SE: fill places from x2/x3/x4/x5 onwards with zi'o > > I don't think that drawing attention to them should be the way > to deal with those places. You and I have had this debate before, so I won't repeat it again. We agree on the ends but not the means. > >7. forethought sumti-tail connectives [tho experimental usage > >of bridi-tail connectives as sumti-tail connectives would suffice?] > > I can't think of a context where {le ge broda gi brode} would > cause ambiguity. If it doesn't, I don't see why it should not > be allowed. That's what I think. > >8. nonveridicality indicator (a) with grammar of NA, (b) that can > >occur in relative phrases > > I don't know, maybe {je'ucu'i}? So long as it is possible to unambiguously and uncumbersomely indicate its scope. --And.