From sabren@xxxxxxxxxxxxx.xxxx Thu Oct 28 14:02:17 1999 X-Digest-Num: 269 Message-ID: <44114.269.1469.959273825@eGroups.com> Date: Thu, 28 Oct 1999 17:02:17 -0400 (EDT) From: "Michal Wallace (sabren)" We also have a lojban webring that I have been pressuring people to join! > (Robin and John: add those HTML fragments to your pages!) I host its main > page at http://www.decadezero.org/lojban_webring.html but my sites are > suffering growing pains this week. Yeah, I keep seeing the URL, but I haven't gotten it to load. > In my opinion, most lujvo should be field-specific. So, a given lujvo might > have one meaning with respect to automobiles, and another with respect to > spaceflight. This is my own idea of the best solution to your above dilemma. > > You see, a lujvo selects one of the many, many interpretations of a tanru. Hmm... But from a computer perspective, that would make lojban harder to use... :) I wonder, though, how to avoid this.. I mean, I don't know every word in the english language. How can you prevent two people from coming up with the same lujvo for two slightly different tanru? > I have initiated the formation of a computer-related glossary, but it's got > a ways to go. Robin, Twery, Nicholas - your input here would be greatly > appreciated. And yours too, Michal! Out of curiosity, how are you tracking this? (Bob, too).. Perhaps the dictionar(y|ies) could be database-driven and put on the web? > I've seen parsers, but they all seem to barf incomprehensible diagrams > instead of sentences, even crappy sentences like babelfish. I think I might be able to manage a crappy sentence with a lot of parentheses on my own.. And probably do a lot better after studying YACC. > I considered doing a Java servlet version of Logflash. I haven't seen the > DOS version - the Mac version isn't all that bad. If you spearhead a > project, I will certainly consider joining. I'm interested. I run a self improvement website (http://www.manifestation.com/) which I'm currently revamping into a subscription service.. One of my plans has been a generic online flashcard/learning system that allows for active learning like the logFlash program, and "passive" learning inspired by various superlearning techniques. You can see the current version at http://www.manifestation.com/neurotoys/learnfast/ ... I've been told it doesn't work in netscape, so I've started reworking it in flash. > You know, a CD rom sounds like a dandy idea. Yeah.. I've got a little spanish CD rom.. there's almost no english at all, except the manual. You just see little drawings and the text, and can hear the actors.. It also has a feature that listens to you and scores your pronunciation/accent... Except for the last bit, simple stuff.. Yet it's really effective. > Although, if we had a decent Lojban --> English parser, which output on the > level of babelfish, I think that would be even dandier. However, I > personally know almost nothing about all that parsing stuff. > > I figure a nearly direct translation from the parser, with all the weird > crap removed, and with maybe 20 - 50 rules for making the logician's > English sound more like Standard English might be enough. This is what I > REALLY want to see done! part of the reason I suggested a perl version was because the dictionary looks like: "x1 eats x2 ..." and it would be fairly easy, once you know which sumti are x1 and x2, to do a search and replace (at least with perl's regular expressions).. > http://extropy.com/ completely offtopic: I saw the foresight exchange a few months ago, and thought it was a really cool idea.. :) Cheers, - Michal ------------------------------------------------------------------------- http://www.manifestation.com/ http://www.linkwatcher.com/metalog/ -------------------------------------------------------------------------