Return-Path: Message-Id: <9110180203.AA00497@relay1.UU.NET> Date: Fri Oct 18 01:27:53 1991 Reply-To: Logical Language Group Sender: Lojban list From: Logical Language Group Subject: parser comments sought X-To: lojban@cuvmb.cc.columbia.edu To: John Cowan , Ken Taylor Status: RO X-From-Space-Date: Fri Oct 18 01:27:53 1991 X-From-Space-Address: cbmvax!uunet!cuvma.bitnet!LOJBAN The following is a chunk of Nick's Polish story, as output by the Beta test Lojban parser. As you can see, it is well nigh unintelligible, at least for checking to make sure that BAI-tagged-sumti are attached as sumti modifiers vs. floating free, etc. Nora wrote a crude formatter, which processes this output into separate sentences with numbers to help you match parens. Nick feels that this offers too small an improvement, and feels that the parser needs some kind of intelligent indented display. 1. How useful do you-who-read this find each of the two outputs, especially for checking such fine points as grouping? 2. Is an intelligent indenter important to determining whether you want to obtain the parser soon (i.e., without such a feature, we can probably release it before the end of the year - with smarts might take several more months). This is partly an economic issue. Our finances have gotten almost as bad as last year, and getting our in-progress software to the point where someone will buy it (and pay real money) is getting to be vital to our survival. Income from our paper products like JL is running less than 1/3 of our production costs these days because so many aren't paying. Tentative parser price is $20 (MS-DOS only, so far), if this affects your opinion. But we also want a product that people will USE, not just buy and put in a disk file that they'll never look at again. lojbab@grebyn.com ____________________ ({ge <[zo ranmi] cu [tcita (di'e VAU)]> gi <[la'o py orzesko py] [finti VAU]>} {ni'o ni'o} {tu'e <[({<[({ <[(so'i roi) (se xlafunca)] [({ ce } ce {lo KU}) VAU]>} {i ca bo} {< le tcadu KU> cu <[cmalu je pindi] VAU>}) i ({ KU} VAU)] KU'O>} { <[(loi ta'urxa'u KU) (le {ri dimna} KU)] VAU>})] [i ku'i] [({la } {goi ko'a GE'U}) ({ <[(va KU) (za {le KU})] VAU>} gi'e { <[(ca {le nunmakcu KU}) ({se si'u} {lei <[se xlafunca] xabju> KU})] VAU>} VAU)]> <[ri ({ badri} VAU)] [i (se mu'i) bo] [ko'a ({ <[(fi {lu <[ki'u ma] VAU> li'u}) ({le tsani KU} e {le dertu KU})] VAU>} gi'e {ki'asku <[(lu {ko <[sidju a'o nai] VAU>} li'u) ({le tsani KU} e {le dertu KU})] VAU>} VAU)]>} i {< lei [nu (smadi { VAU}) KEI] KU> cu } ) i (tu'e {<[({la iupiter} {goi ko'e GE'U}) ({ } VAU)] ({ge <[zo ranmi] cu [tcita (di'e VAU)]> gi <[la'o py orzesko py] [finti 11 11 1 1 2 211 11 1 1 VAU]>} 111 {ni'o ni'o} {tu'e <[({<[({ <[(so'i roi) (se 1 1 1 112222333 3 3 3 334 4 4 xlafunca)] [({ ce } ce {lo KU}) VAU]>} 45 5 55 5 54 4 44 333 {i ca bo} { cu <[cmalu je pindi] VAU>}) 3 3 33 3 33 3 333 i ({ KU} VAU)] 333 3 3 34 4 44 444 4 4 43 KU'O>} { <[(loi ta'urxa'u KU) (le {ri dimna} KU)] VAU>})] 33 33 3 334 4 4 4 4 43 3332 [i ku'i] [({la } {goi ko'a GE'U}) ({ <[(va 2 2 233 3 33 3 33 333 3 334 KU) (za {le KU})] VAU>} gi'e { <[(ca 4 4 4 4 4 443 33 33 3 334 {le nunmakcu KU}) ({se si'u} {lei <[se xlafunca] xabju> KU})] VAU>} 4 44 44 4 4 44 4 4 443 33 VAU)]> 322 <[ri ({ badri} VAU)] 2 2 22 333 3 3 32 [i (se mu'i) bo] [ko'a ({ <[(fi {lu <[ki'u ma] VAU> li'u}) 2 3 3 2 2 333 3 334 4 44 4 4 44 ({le tsani KU} e {le dertu KU})] VAU>} gi'e {ki'asku <[(lu {ko <[sidju 44 4 4 443 33 3 334 4 44 a'o nai] VAU>} li'u) ({le tsani KU} e {le dertu KU})] VAU>} VAU)]>} 4 44 4 44 4 4 443 33 3222 i { VAU}) KEI] KU> cu }) 2222 i (tu'e {<[({la iupiter} {goi ko'e GE'U}) ({ } VAU)] 2 22233 3 3 33 333 3 3 33 32 lojbab