From rlpowell@digitalkingdom.org Thu May 13 16:27:59 2004 Received: with ECARTIS (v1.0.0; list lojban-list); Thu, 13 May 2004 16:27:59 -0700 (PDT) Received: from rlpowell by chain.digitalkingdom.org with local (Exim 4.32) id 1BOPcF-0000QU-2k for lojban-list@lojban.org; Thu, 13 May 2004 16:27:51 -0700 Date: Thu, 13 May 2004 16:27:51 -0700 To: lojban-list@lojban.org Subject: [lojban] Re: erasure words Message-ID: <20040513232751.GM4461@digitalkingdom.org> Mail-Followup-To: lojban-list@lojban.org References: <20040513213804.GG16333@fysh.org> <20040513214744.GA4461@digitalkingdom.org> <20040513222637.GI16333@fysh.org> <20040513224341.GF4461@digitalkingdom.org> <20040513231908.GL16333@fysh.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20040513231908.GL16333@fysh.org> User-Agent: Mutt/1.5.5.1+cvs20040105i From: Robin Lee Powell X-archive-position: 7823 X-ecartis-version: Ecartis v1.0.0 Sender: lojban-list-bounce@lojban.org Errors-to: lojban-list-bounce@lojban.org X-original-sender: rlpowell@digitalkingdom.org Precedence: bulk Reply-to: lojban-list@lojban.org X-list: lojban-list On Fri, May 14, 2004 at 12:19:08AM +0100, Zefram wrote: > Robin Lee Powell wrote: > >I completely understand that. There is no valid way to handle > >something when all I know is that it erases *something*. > > Yes. That's half of my point. The other half is that there *are* > valid ways to handle something when all you know is that it *doesn't* > erase anything. I'm not sure quite how much one can do in this case; > it's not a lot, but you can at least, in theory, allow the erroneous > word to be erased. > > In the example "le broda cei'au si brode", if one knows for certain > that "cei'au" is not an eraser then one can confidently preprocess to > "le broda brode". Just for the record, my parser already handles that (because SI is one of the cases where any morphologically valid Lojban word is handled, along with SA, SU, ZO, ZOI delimiters, ZEI and BU. > > Despite all of this discussion, the language is *not* actually > >open for change. > > My understanding is that the language may be added to in the future, > albeit at a slow pace, and that this is what the extended cmavo space > is for. Am I right here? Yes. I just wanted to avoid the impression of a free-for-all, which I can imagine a newbie getting from these discussions. > For clarity: I mentally divide language discussions here into a > handful of categories: > > * working out what the current language definition really means > (e.g., the "zo fu bu" debate) Not in that case; we know exactly what it means: the official grammar chokes on it, therefore it's invalid. I was asking what made sense for it to mean, given the word definitions and not worrying so much about the official grammar. > * artifacts of the current implementations that we might want to > define differently (e.g., "zei" at the beginning of text) Right. > * possible future additions to the language (any proposed new cmavo) Yep. > * tricky parts of the language that we could have done differently > ("zoi si") Yep. > The borders between these categories are fuzzy in some cases, and some > conversations have aspects of more than one category. My > understanding is that only the last of these four categories is really > problematic, because it is only in that category that the language > definition is being contradicted so as to invalidate existing usages. > Anything in the third category can't become official in the near > future, but doesn't actually cause a problem. Is this a good view of > things? It's very good, yes. Go you. > I'm personally very clear on the distinction between what a particular > parser implements and the official language. I've had to play a bit of catch-up with you; you have more sophistication in these areas than I was expecting. My apologies for selling you short in advance. > Something I might have missed so far: is there a taboo here against > discussing possible language changes due to fear of rehashing the > (extensive) language design work? That question is, itself, taboo, because of the huge arguments it starts. :-) The language is strongly divided between people who consider the current language definition sacrosanct and those who think that a clean definition is more important than sticking with the status quo. I fall in the middle, but I'm closer to being a conservative in most cases. > Perhaps I have been insufficiently conservative, for which I > apologise. You can be as conservative or not as you like. We still listen to xorxes, after all. :-) (xorxes is very active, and a known non-conservative). > >1. I'm already doing it, so I'm not sure what the point is, although > >I would *love* someone to look over my PEG grammar. > > I'd like to. I've never seen PEG before. Allow me to apologize in advance: while the grammar does the right thing, it is messier than it needs to be in a few places, and is not well commented. -Robin -- http://www.digitalkingdom.org/~rlpowell/ *** I'm a *male* Robin. "Many philosophical problems are caused by such things as the simple inability to shut up." -- David Stove, liberally paraphrased. http://www.lojban.org/ *** loi pimlu na srana .i ti rokci morsi