Hi. I personally do not see why these supplemental characters cannot be created, as done for other Latin-1 characters (http://www.unicode.org/charts/PDF/U0080.pdf
I'm a bit confused though, and have a question: how are characters from current Marshallese texts and from new texts with the new characters going to be properly matched? Could four such characters also be supplemented with Phillipe's first suggestion, a new combining cedilla which the characters would decompose to, along with the base character in each case? : > Can't we imagine : > - the encoding of a new non ambiguous cedilla (i.e. COMBINING CEDILLA > ATTACHED BELOW) ? > 2013/6/21 "Jörg Knappen" <jknappen_at_web.de> >> Dominikus Dittes Scherkl schrieb: >> >> >Why not instead encoding a new combining "MARSHALLESE CEDILLA" >> >that ought to be used with g, k, l, m, r and their uppercase counterparts? >> >> This is not a good idea, because the combining "MARSHALLESE CEDILLA" can >> be combined with the letter C, too. Is there anyway that the combining cedilla can be restricted to combining with certain letters? I guess not; sigh. >> This creates all kind of havoc with the Ç (including fake >> internationalised domain names). The remaining letters >> with cedilla need to be precomposed and non-decomposable. Hmm -- I just got finished reading at unicode for the second or third time (I don't know why I keep rereading this but I do) that much of the confusion caused by combining of letters in IDNs is the problem of the higher level domain registries, not of unicode, but yes perhaps a new combining cedilla would create one more problem for the IDNs??(Would it help "phishing" by that much to have one more character to play with?) Nevertheless Asmus says that it is likely that the four precomposed characters probably at some point somehow will be decomposed. From: Asmus Freytag <asmusf_at_ix.netcom.com> Date: Wed, 19 Jun 2013 09:28:33 -0700 > On 6/19/2013 6:36 AM, Michael Everson wrote: >> Only in text which has been decomposed. Not all text gets decomposed. > All text may get decomposed without warning. > As data is shipped around and processed in various parts of a > distributed system, nobody can make any safe assumptions on the > normalization state for their data. They may get composed, decomposed, > or they may miraculously remain in whatever mixed normalization state > they were created in. > The point is, any technical argument or design decision that implies > that one has control of the normalization state is ipso facto suspect. > A./ I don't see how else mapping to old texts can be done anyway if there is nothing for the new characters to decompose to. Will each then decompose to the base character and the current cedilla? Is this is not going to cause the same problems that we have now? Or am I wrong? In any case what are they going to decompose to? To the old base character and nothing? Is that a possibility at all? Or would it be a disaster? (Just a question; since I am not familiar with Marshallese, I can't answer it myself.) > > --Jörg Knappen Best, --C. E. Whitehead [email protected]
