From LOJBAN@CUVMB.BITNET Sat Mar 6 22:56:45 2010 Reply-To: "Mark E. Shoulson" Sender: Lojban list Date: Wed Jan 29 17:07:23 1997 From: "Mark E. Shoulson" Subject: selma'o X-To: lojban@cuvmb.cc.columbia.edu To: John Cowan X-UIDL: 6ce2d6dfa6bc98016279f5a1b8ef9102 Status: U X-Mozilla-Status: 0001 Content-Length: 1468 X-From-Space-Date: Wed Jan 29 17:07:23 1997 X-From-Space-Address: - Message-ID: -----BEGIN PGP SIGNED MESSAGE----- Something in one of lojbab's forwardings to conlang set me thinking. How many selma'o do we really use regularly? Obviously not all of them; MEX is hardly ever used; when was the last time you said ke'e, etc etc. Several possible projects can stem from this... 1. try, a priori, to decide on a relatively small set of selma'o and see if you can work comfortably within it. 2. try, a priori, to minimize the set of selma'o, as small as you can without getting "too" cramped (subjective judgement). 3. analyze existing texts and see the relative proportions of selma'o in use (e.g. replace all words with their selma'o and count up totals). Anyone find these fun? Let's see, off the top of my head, what selma'o would I really need? BRODA (of course. Gismu and counting lujvo), LE, LA, CMENE, NU, PU, A, UI, NA, NAI, BAI, COI/DOI, PA, ZOI, I, JA, KOhA, SE, FA, GOI, NOI are among the biggies... sometimes KU, KEI, even KUhO, BE, ... ROI, GOhA,... I'm pretty sure I missed a couple of really important ones, especially really common ones that have very few cmavo (like I nearly left out NA and NAI...) Thoughts? ~mark -----BEGIN PGP SIGNATURE----- Version: 2.6.2 Comment: Processed by Mailcrypt 3.4, an Emacs/PGP interface iQB1AwUBMu+8HMppGeTJXWZ9AQHYuQL/Zq+V3dCyVaLlsUzCOxuFP5U2SOJzrNSk EuUGbh1Rs81RebtdTsF2ik3NzT6d7AHPnwNt9R1Kqo8RUX//pSILAmXgHUcaKEhS 8YOL5oixhMXVgFb7X+xM7WF2n5AbtXoi =ztd3 -----END PGP SIGNATURE-----