Antique languages in epub

[carmenchu]

Hello!
I hope somebody can help with this: it is there some recommendation/best practice about dealing with words in an antique language in an epub?
And by 'antique language' I don't mean Latin nor Classic Greek, but beauties like Hittite or Phyrgian--so, I suspect that both spelling and pronunciation are dubious, and for specialists...
However, I feel that they ought to be got out of the way of the spell-checker, and if one were to use a text-to-speech
Any help will be appreciated.

[Jellby]

Use the appropriate lang code, spell checking and text-to-speech should obey it, and that could mean ignoring the word.

[carmenchu]

Please, are there language codes for Old Indic, Avestan, Old Church Slavonic, Old Norse, Old English, Old High German, Phyrgian, Hittite, Luwian...? Where does one find them?
I was rather hoping for some comprehensive label on the lines of xml:lang="exclude" / "none" / "PIE", that would deal with all those bits--some are only roots: 'xxx-'

[jhowell]

I am not familiar with those languages but it occurs to me that some may be based on obsolete alphabets. If so there may not be existing fonts to support them or even Unicode code points defined for some of the needed characters.

[carmenchu]

Originally Posted by jhowell

I am not familiar with those languages but it occurs to me that some may be based on obsolete alphabets. If so there may not be existing fonts to support them or even Unicode code points defined for some of the needed characters.

Well, neither am I familiar, and in fact I am cleaning the book (from OCR) as I read. There were some glyphs rendered by gif images (just like I used to render equations before MATHML) but fortunately, I was able to find all of those characters in utf-8--chiefly Latin subset, and some Greek. I suspect that in such cases (most of the languages referred to in the book are preliterate) linguists try to give a 'transliteration',or a 'reconstructed pronunciation'. Most are not worse than *k’ṃtom (the m has a point below) but I feel that there should be some epub mark-up to distinguish such from 'normal language'... So, I am asking--hope that somebody knows ??

[DNSB]

Are we looking at Hittite, Middle Hittite, Neo-Hittite or Old Hittite? The ISO 639-3 language code list has all four of those. Phrygian only has the one entry.

See ISO 639 Code Tables for the complete searchable list.

Wrapping an string in the language code should prevent a spellchecker from attempting to spellcheck it unless it has a matching dictionary. Something like <span xml:lang="xpg">*k’ṃtom</span> for example.

[carmenchu]

My, oh, my! Ask to learn: I never suspected that there would be so many codes for dead languages!
Now, a short and perhaps silly question: would it work just to mark those words (or roots, mostly) as <span xml:lang="pied">*k’ṃtom</span>, i.e., with a non-existing language code, or it would it sound all the bells in epubChecker? Really, the author is hardly ever giving words in a particular language, but rather pointing out 'cognates' or common roots in related languages... It seems rather an overkill to insert more than thirty different <span xml:lang="---">...</span> for mere fragments, mostly not belonging to actual languages, when what is really required is the notice 'don't treat this as a common word'.

Anyway, my thanks for the link to the code tables: good to know that there is such a complete reference for out-of-the-way languages.

[DNSB]

I've never tried a non-existent language code but give it a try and see what happens.

Hey, if it's got room for Klingon, antique languages are nothing.

[Jellby]

There are some special codes for cases where no other existing code is appropriate. Of course, what a particular application will do or not do with such codes is... unknown (I think calibre does not accept 'zxx' as a book language, for example).

[Jellby]

Originally Posted by carmenchu

Please, are there language codes for Old Indic, Avestan, Old Church Slavonic, Old Norse, Old English, Old High German, Phyrgian, Hittite, Luwian...?

You could refer to:
https://www.loc.gov/marc/languages/language_name.html

Old Indic: USE Vedic: Assigned collective code [san]
Avestan: [ave]
Slavonic, Old Churh: USE Church Slavic: [chu]
Old Norse: [non]
Old English: USE English, Old (ca. 450-1100): [ang]
Old High German: USE German, Old High (ca. 750-1050): [goh]
Phrygian: Assigned collective code [ine]
Hittite: [hit]
Luwian: Assigned collective code [ine]

I had some "fun" translating all language names in calibre...

[carmenchu]

Thanks again! I have tried xml:lang="none" and EpubCheck has passed it. Moreover, searching the iso database for that found no match--I tried "gib" for "gibberish", but there is a real language with such code...
Aside from this particular non-fiction book, there are tons of my favourite SF & Fantasy works chock-full of words in the author's tooled lingo. Klingon is nothing to it--think of Edgar Rice Burroughs! I suspect the Tarzan, Barsoom, Venus, ... series haven't got their own iso entries.

[PoP]

The IANA language registry and RFC5646: Tags for Identifying Languages might also be useful resources.

I used these extensively doing this.

A cursory inspection found the Gibanawa, the tlh, and all the Hittite variations of the above posts , yet not the Phyrgian

[Doitsu]

Originally Posted by carmenchu

Thanks again! I have tried xml:lang="none" and EpubCheck has passed it.

AFAIK, EPUBCheck doesn't check the values of xml:lang attributes. For example, xml:lang="nope" isn't flagged either.
You could use "und" or "zxx".

[carmenchu]

Many thanks, everybody! Besides learning a lot about language codes (until now, I had only bothered about the ones that I can read--more or less)--I am definitely clear about how deal with 'not really a language'.

[Tex2002ans]

Originally Posted by Doitsu

You could use "und" or "zxx".

"und" would be correct in this case.

Because from the link you posted:

When the text is non-linguistic

Use the subtag zxx when the text is known to be not in any language.

This would apply for text such as type samples, part numbers, illustrations of binary data, etc. The definition of zxx in the IANA Language Subtag Registry is 'no linguistic content'.

For example:

<p>Here is a list of part numbers: <span lang="zxx">9RUI34 8XOS12 3TYY85</span>.</p>

so zxx is for gibberish + und is for linguistic content, but unknown what language.