[sword-devel] OSIS Users Manual considered confusing

Kahunapule Michael Johnson kahunapule at mpj.cx
Sat Mar 3 12:38:39 MST 2012

On 03/03/2012 06:47 AM, Matěj Cepl wrote:
>> Nearly half the OSIS editor body is subscribed to this list, so please,
>> if you have an honest question about the standard, humbly ask and know
>> that there might be an answer you hadn't considered.
> I am not sure why I should be humble, when noticing that as a user of this standard I found it very much lacking in what exactly is expected from any standard: setting unequivocal rules about how to create an inter-operable product with the expectation that somebody else using documents produced by my software will be able to use them without a problem.
> Again, no personal offense meant, and I don't doubt specific knowledge of anybody involved in creating OSIS standard.

My complements to you on your restraint, Matěj. I understand your disappointment. I also understand what OSIS actually is, and why it is that way. OSIS is a common format for many visions, meaning that it is really not just one format, but many, and subject to a variety of interpretations. The vision of unambiguous interoperability isn't part of OSIS, unless further clarified in a document like http://crosswire.org/wiki/OSIS_Bibles. Even then there is room for improvement. Hopefully that improvement will
take place as work proceeds.

As far as making OSIS work with verses as containers and sections and paragraphs/poetry elements as milestones, I tried that several years ago. The theory surrounding the ability to transform between dominant structures is sound, but OSIS doesn't really support doing that in all of the conditions I encountered in real Bible translations in that mode. The committee responsible made some assumptions about Bible formatting that were not shared by my good friends, the Bible translators. That means that either
the Bible formatting has to change (not an option for me), or OSIS has to change (not within my power to do, and contrary to its design goal of stability). After that disappointment, I gave up on OSIS for a while. I may try to generate an improper subset of OSIS sufficient to generate a Sword module from, replacing usfm2osis.pl, which doesn't support all of the translations I'm working with. I think it is worth one more try...

I applaud any effort to directly support USFM (or an equivalent XML format like USX) in any component of SWORD, because that is where the majority of minority-language translations are.

Please take a look at http://ubs-icap.org/chm/usfm/2.3/index.html and/or http://ebible.org/usfx/. Another XML equivalent of USFM is USX, generated by Paratext. USX isn't officially documented, yet, but if you can get a copy of Paratext and examine its USX output, you can figure it out. If you can get your module into any of those formats, it can be added to the 220 Scripture projects (and counting) that I have queued up to convert to Sword format.

As for what I think of some of the philosophical changes in OSIS vs. USFM, other than simply changing from \-escaped markup to XML, is probably best expressed at http://ergofabulous.org/luther/ on a philosophical level (not a personal level!) and at http://ebible.org/usfx/Bible-encoding.htm on a technical level. The latter document is a bit dated, but not a lot has changed since then except the emergence of OXES (and attempt to make OSIS more practical) and USX (another XML equivalent of USFM). Not every
organization who claimed to embrace OSIS actually did so in a significant and compliant way, for a variety of reasons. The Sword Project is the major exception.

For my dear brothers who worked on the OSIS standard so long for so many years, I appreciate your work, your sacrifice, and your intentions.


More information about the sword-devel mailing list