On Tue, Jun 3, 2014 at 5:55 PM, Claudio Beccari wrote: > On 03/06/2014 13:21, Mojca Miklavec wrote: > >> We need something from the standard, see >> >> http://www.iana.org/assignments/language-subtag-registry/language-subtag-registry >> and I don't find anything suitable on the list. > > I examined the file at the link indicated above; I could not figure out where > that code is needed in the hyphenation files; I suppose it is not needed > there.
The code is not really *needed* for hyphenation, but we would like to use a consistent naming scheme for the patterns. In short: we can in principle pick any name, but I consider it a bad practice to rename the files later. To TeX users it shouldn't make a difference. If nobody is willing to deal with the standard, we don't really have to do that at all. The only visible difference is whether we put "classiclatin" or something else into language.dat/language.def. > Actually the file contains an entry: > > %% > Type: language > Subtag: la > Description: Latin > Added: 2005-10-16 > Suppress-Script: Latn > > which is good for Latin, and porbabily is associated to the existing > loadhyph-la.tex file and its friends. Yes. That subtag "la" gives the name to "hyph-la.tex", "loadhyph-la.tex" etc. > Now classic Latin is not a different language; it uses the same alphabet and > I could not invent a new entry to that: the ISO regulatins don't distinguish > modern Latin from medieval latin and from classical Latin. Yes, I'm aware of that. > The Wiki ISO 639 entries for Latin are: > > Language Language Native name 639 > family name 1 2T 2B 3 6 notes > > Indo-European Latin latine, lingua latina la lat lat lat lats ancient We don't need a new ISO 639 entry (and cannot register one either). Just a subtag to specify the language variant. Similar to "el-monoton"/"el-polyton" and "de-1901"/"de-1996". > I know that it is necessary to name another language for a different > hyphenation pattern set > the same as it is done with Greek (even if it is not actually used, or it is > used in a funny way -- the (babel-)greek.ldf does not ever use the ancient > greek pattern set nor the modern monotonic greek pattern set, it uses only > the polytonic greek pattern set). > > Polyglossia, on the opposite does not have any problem in using three greek > pattern sets: apparently they refer to languages "monotonic", "polytonic", > and "ancient". In the language-subtag-registry for greek I find only: You need to talk to Arthur for support of a new language variant for Latin in Polyglossia. And probably prepare something for babel. > %% > Type: language > Subtag: el > Description: Modern Greek (1453-) > Added: 2005-10-16 > Suppress-Script: Grek > > %% > Type: language > Subtag: grc > Description: Ancient Greek (to 1453) > Added: 2005-10-16 > > %% > Type: language > Subtag: grk > Description: Greek languages > Added: 2009-07-29 > Scope: collection > > %% > Type: variant > Subtag: monoton > Description: Monotonic Greek > Added: 2006-12-11 > Prefix: el > > %% > Type: variant > Subtag: polyton > Description: Polytonic Greek > Added: 2006-12-11 > Prefix: el > > plus several "script" entries. > > In similarity with Greek a new entry could be added as such: > > %% > Type: variant > Subtag: classic > Description: Classic Latin > Added: 2014-06-03 > Prefix: la > > I think this is the least invasive addition to the subtag list. Yes, something like that would make sense. But: does one need more than one variant? It would be weird to request, say, "la-classic" now and "la-archaic" + "la-modern" + "la-neo" a few years later when you realize that even different patterns are needed for those. What set of subtags would make sense? Would you be willing to draft a proposal and submit a request for inclusion? See http://www.langtag.net/register-new-subtag.html There is one potential problem with the above mentioned subtag though. According to RFC 5646: Requests to add a 'Prefix' field to a variant subtag that imply a different semantic meaning SHOULD be rejected. For example, a request to add the prefix "de" to the subtag '1994' so that the tag "de-1994" represented some German dialect or orthographic form would be rejected. The '1994' subtag represents a particular Slovenian orthography, and the additional registration would change or blur the semantic meaning assigned to the subtag. A separate subtag SHOULD be proposed instead. This means that registering "la-classic" would prohibit anyone else from registering "<anotherlanguagetag>-classic". There is a chance that no other language would ever need that tag anyway, but it's something I wasn't really aware of until now. I'm also a bit confused by what Wikipedia says: The word "Latin" is now taken by default as meaning "Classical Latin", so that, for example, modern Latin text books describe classical Latin. [1] Classicists use the term "Neo-Latin" to describe the use of the Latin language for any purpose, scientific or literary, after the Renaissance. [2] [1] http://en.wikipedia.org/wiki/Classical_Latin [2] http://en.wikipedia.org/wiki/New_Latin So to me it's not even clear which of the both patterns should be called "latin". But just to make sure that I understand it properly: are two pattern sets needed because exactly the same words would hyphenate differently in classical and modern Latin? Or is it just that the vocabularies of both are so different that it's very difficult or impossible to cover both variants of the language? (I'm asking because I would like to know if there is a trick to cover both variants with the same set of patterns or if that's theoretically impossible because the rules are too different.) Thank you, Mojca
