On Mon, Mar 5, 2012 at 7:49 PM, Philippe Verdy <verd...@wanadoo.fr> wrote: > Le 5 mars 2012 18:33, Denis Jacquerye <moy...@gmail.com> a écrit : >> [1] pp.19-24 >> http://www.archive.org/stream/atlaslinguistnot00gilluoft#page/18/mode/2up > > I note an interesting character in your page : the « open g » used to > denote the « g dur français » show in the middle of the page on the > right. And I'm not sure how it should be encoded (as a registered > variant, or using a new character). It does not seem to match the > variant encoded for IPA (single storey only, instead of the allowed > double storey for the standard "g"). > > Another character is the special "z" with an extra stroke attached on > the left of the standard descender of the italic letter and used to > denote the « z doux anglais ». I'm not sure this is a cedilla or > something else. > > Did I miss something ?
These are just glyphic variants. I was a bit confused at some point about a couple of them, but seeing more texts using this alphabet, it is clear they are just font dependant. The g is an italic script g with an open bowl but still just script g, the z is just italic, the s is just an italic script-like s. In some other books these have more regular shapes. They do not occur in minimal pairs even in the Atlas linguistique, i.e only one g, z or s is used. > > There are many diacritics used in that document (and probably in the > companion series of 25-pages booklets referenced at end of the text, > if there are available in that Library where you found this book), > intended to represent the phonetics with a set of symbols distinct > from IPA, not the orthography. Most of them are clearly identifiable > in Unicode, even if they may coccur in stacked pairs. There are several diacritics I could find in books on archive.org, but the number is not alarming. It would be interesting to actually check the maps of the Atlas or find other books using this phonetic alphabet. There's also a number of subscript letters with diacritics: ã ç̂ ȼ (or ꞓ) ė (or é) ī (or ĩ) õ (or ō) ò œ̃ œ̆. There might be more. Would these also be candidates for encoding? -- Denis Moyogo Jacquerye
