Return-Path: Date: Thu, 16 May 91 14:17:14 -0400 From: cbmvax!uunet!husc9.harvard.edu!shoulson (Mark Shoulson) Message-Id: <9105161817.AA01627@husc9.harvard.edu> To: infmx!godzilla!cortesi Cc: lojban-list@snark.thyrsus.com Reply-To: cbmvax!uunet!cunixf.cc.columbia.edu!shoulson In-Reply-To: David Cortesi's message of Wed, 15 May 91 14:39:08 PDT <9105152139.AA03310@godzilla.informix.com> Subject: planned languages archive Status: RO X-From-Space-Date: Thu May 16 16:08:41 1991 X-From-Space-Address: cbmvax!uunet!husc9.harvard.edu!shoulson Return-Receipt-To: cbmvax!pyramid!infmx!godzilla!cortesi@uunet.UU.NET From: infmx!godzilla!cortesi@uunet.UU.NET (David Cortesi) Date: Wed, 15 May 91 14:39:08 PDT X-Mailer: ELM [version 2.3INFORMIX PL45] I have tried using the planned-languages archive server with mixed results. The machine responded with commendable promptness, compared to some others I've used. Unfortunately I found problems with the files I retrieved. I retrieved the lojban brochure in French, just for a kick. Unfortunately what I got is not usable. There is a helpful note at the top of the file listing the PC character codes that were used for accented characters and how to get rid of them with a "tr" command. Unfortunately those are no longer in the file: they appear to have been converted at some point by stripping their high bits. What once was octal 202 now is ^B, etc. I couldn't say if that happened on the way into the archive or on the way out, but that's how it got to me. Also the french brochure has a large number of broken lines. These most likely occur where an accented character (probably octal 212, 'e) when stripped of its high bit, turned into a linefeed or cr. Youch. Looks like it got mangled on the way into the archive, as our original seems to have the same difficulty. Sorry we didn't notice it. I'll try to fix it. With any luck, it shouldn't take much more than a little replace-string, plus something for the \0212 characters. I also retrieved the only gismu list shown in the index. This turns out to be the '88 baseline version which we all know & love. I had been hoping to find the latest revision, the one that contains 100-byte English renderings. Will that be sent to the p-l server soon? That's what lojbab sent us. I hear they're working on getting other stuff over (including, presumably, a later Gismu list). As I understand it, part of the problem is that LLG does most of its work on another computer system, and conversion takes time, which is a valuable commodity. It's not high on the priority list. Finally, I had hoped to see the cmavo file sent out with the recent JL, but it wasn't visible in the index file. I *really* want to get that so I can print it sorted by morpheme, not by cmavo (then maybe it will be of some use...) We haven't got a cmavo list, only the gismu list. Perhaps that will be remedied. Thanks for your interest. ~mark o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o N2KOT Mark E. Shoulson: shoulson@cunixf.cc.columbia.edu