From ybf2u@curry.edschool.virginia.edu Sun Feb 19 13:34:39 1995 Received: from moose.cs.indiana.edu by whale.cs.indiana.edu (5.65c/9.4jsm) id AA01885; Sun, 19 Feb 1995 13:34:39 -0500 Received: from virginia.edu (uvaarpa.Virginia.EDU) by moose.cs.indiana.edu (5.65c/9.4jsm) id AA26847; Sun, 19 Feb 1995 13:34:37 -0500 Received: from curry.edschool.virginia.edu by uvaarpa.virginia.edu id aa00320; 19 Feb 95 13:34 EST Received: (from ybf2u@localhost) by curry.edschool.Virginia.EDU (8.6.8/8.6.6) id NAA85875; Sun, 19 Feb 1995 13:34:34 -0500 From: Yitna Benyam Firdyiwek Message-Id: <199502191834.NAA85875@curry.edschool.Virginia.EDU> To: dan'El yaqob , Yonas Fisseha Date: Sun, 19 Feb 95 13:34:34 EST In-Reply-To: <199502161919.OAA75247@curry.edschool.Virginia.EDU>; from "Daniel Yaqob" at Feb 16, 95 2:19 pm X-Mailer: ELM [version 2.3.1 PL11] Status: R Sorry guys. Mea culpa! I've been negligent in keeping up with my end of the project. It's been difficult--to say the least--to shift form Clarence Thomas, to Perl, to community college ethos, to race-relations in Charlottesville, to Ethiopic punctuation order on a computer matrix for Unicode, without short circuiting once or twice along on the way. But I'm hanging on, mostly due to your persistence... danEl, I'm afraid I don't have introductory text ready :( I don't want to say it's not crucial, but perhaps just not as pressing as the stuff you guys are working on. I'll try to respond to some of those questions as fully as I can now in this brief respite... > glyph similarity. I'll check my fidel poster tonight. I *think* the ordering > was like: > > " " . , ; : ? > > The question mark was english though, The above for Ethiopic doesn't bother > me, but I want to have a concrete reason for it. To order the punctuation danEl, one thing I've learned is to be suspicious of order in linguistic phenomena. the whole thing is really more of a choatic system that is completelyy unpredicatable in the type of order that will spew out from it. We should not be tempted by the hindsight view of results to suppose the order is inherent in the system. In addition, we need to remember that the Ethiopic writing system, unlike English, is in rapid evolution even as we speak. (Not to say english is not evolving in its own way...) Many prescriptions are made by those who would like to influence the outcome of the transition one way or another. We've witnessed this in Ethiopic from Menelik on to the current turmoil of national languages. Having said that, I'd add that, using the English system, I would say the comma and semicolon are more closely related in that they operate on the relationship within and between sentences (pause, and full stop between sentences with related ideas), and therefore can be grouped (in my view) with the period and other sentence-end punctuation. Colons ont he other hand, relate sentences with non-sentence structures and are better off grouped with dashes, parentheses, brackets, etc. I generally agreed with Yonas' descriptio of usage for commas and semicolons and colons. > by SOME concept of flow control is highly logical to me, but may reuslt in > an untraditional (or uncommon) arrangemnt. > > Hmmm, in the bigger picture, when latin punctuatinon is used with Ethiopic, > such as - + / \ ( ) $ % etc, which takes lexicographic precedence? Will > latin punct come before ethiopic punct or after? Will it come before > Ethiopic numbers or after all of fidel? What about the arabic numbers > where do they... Sigh, maybe Unicode addressing defines this for the world. Since all of the latin puctuations and symbols that do not have Ethiopic equivalent ned to be included as part of the Ethiopic set, I would not worry too much about their place in the system. I wouuld, however, groups the symbols and puctuations separately. Also, following the ASCII order wherever there is no other reasonable choice is not a bad idea. > > Q: Will the matrix need a $ sign ? Whenever I see the br mark it is a > dollar sign but always with two || and not | which is often what the latin > font gives. Do we want to but the two || dollar sign in the address table > so people can be assured it is availabe? Yes, that's a good idea. However, I'm not sure the double lines are typically Ethiopian because they are seen many places representing the US dollar. It may be simply in ASCII that the single line was chosen. Again, if there's no reason, go with the ASCII. . . . danEl also asked whether we should work on the second Appendix in the CSES report. I believe we should. Much of what has been done for SERA in terms of developoing the latin "names" for the characters is already in place. So, it shouldn't require any more work. Again, very sorry to have fallen behind. I hope the above redeems me a bit in your eyes. Iknow you'll understand. tired! yTna