[osis-editors] suggested corrections

[Patrick Durusau]

Michael,

Michael Paul Johnson wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> 
> 
> While attempting to encode the Holy Bible in Tok Pisin (Melanesian 
> Pidgin, a language spoken widely in Papua New Guinea) in OSIS, I 
> discovered two more serious limitations in the OSIS specification & 
> documentation.
> 
> The first difficulty came in attempting to encode the required 
> xml:lang attribute on the osisText element. There is no 2-letter code 
> for this language. Two-letter codes for languages are inadequate for 
> Bible translation use, because there are over 6,000 living languages 
> spoken today on this planet. Two-letter codes are limited to 676 
> combinations. Three letters could work, albeit a bit cryptically, 
> using SIL Ethnologue codes. Tok Pisin does have a perfectly good 
> three-letter Ethnologue code (PDG), which I tried to insert, and that 
> caused a validation failure. Suggestion: make this attribute OPTIONAL, 
> but require the header/work/language element, which does allow 
> language type "SIL".
> 
Actually, it should validate if you use "x-" to preface the value you 
are using for xml:lang. I don't know why the W3C choose such a small 
list of language codes but they do allow IANA language codes and 
anything that is preceded by an "x-".

> The second difficulty came in encoding verse identifiers for verse 
> bridges. Because of word order constraints and translational 
> considerations, it is often better to translate more than one verse as 
> a unit. A few less literal translations and paraphrases even in 
> English use verse bridges, too. In these cases, the verses are marked 
> with a range. Allowing an osisID range in the same way as an osisRef 
> range would solve this problem. OSIS is not usable for a broad range 
> of Bible texts without a good way to encode verse bridges. I recommend 
> using the same syntax for verse bridge OSIS IDs as for OSIS 
> references, but document that the meaning of this syntax for an osisID 
> is for a verse bridge, and should not be used to identify passages 
> that really are translated verse by verse.
>

The value of an osisID can be a list of osisIDs. That is you can say: <p 
osisID="John.1.1 John.1.2 John.1.3"> Note the whitespace separation of 
the IDs.

Unlike the range solution you propose, I don't know of a requirement 
that the refereces be contiguous, although they are most likely to be so.


> One minor point I noticed while attempting to convert the World 
> English Bible and Hebrew Names Version to OSIS 2.0 was that footnote 
> start tags had no equivalent in OSIS. These are the tags that make it 
> easy to convert the text that a note refers to into a hyperlink (as 
> you will see in, for example, http://eBible.org/web/John.htm, which 
> was converted from web.gbf in http://eBible.org/web/webgbf.zip). For 
> now, I threw in some milestone elements with x-startNoteAnchor 
> attributes at those points. The equivalent element in XSEM is the 
> "anchor" element.
> 

Not sure I understand but OSIS has a note element, as well as the HTMP 
<a> element. Oh, you mean how does the anchor get there to point to the 
note? If so, that is a matter of processing, not encoding.

For example, I have a note in the text and I want to print the text to 
paper (yikes!). What gets rendered is a number in the text, usually 
superscripted and a matching number at the bottom of the page or end of 
the chapter. Neither of those is actually marked in the text.

The reason for that is the question you raise, now I want to do an HTML 
document and I want a hyperlink to the note. In addition to doing the 
superscripted number, perhaps even highlighting the word where it is 
found, I automatically generate a hyperlink as well that points to the 
note.

Think of the numbers you see on footnotes as being manual pointers and 
you will see what I am talking about.

Hope you are having a great day!

Patrick


> Thank you for your work towards making OSIS a usable standard. It 
> isn't too far off...
> 
> 
> 
> 
> 
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.2.3 (MingW32)
> Comment: http://eBible.org/mpj/gpg.htm
> 
> iD8DBQFAB9beRI/gxxfXR7sRAh9GAJ44hovUvVFwlyrvXQiKZM0Xfa25CACg4U5l
> 39Wj8iknOIJUCJfNciVGnS4=
> =Bgro
> -----END PGP SIGNATURE-----
> 
> 


-- 
Patrick Durusau
Director of Research and Development
Society of Biblical Literature
Patrick.Durusau@sbl-site.org
Chair, V1 - Text Processing: Office and Publishing Systems Interface
Co-Editor, ISO 13250, Topic Maps -- Reference Model

Topic Maps: Human, not artificial, intelligence at work!