[sword-devel] The OSIS Name element and SWORD rendering?

David Haslam dfhdfh at protonmail.com
Tue Jun 27 07:09:13 EDT 2023


For all our software developers to consider...

As far as I know, SWORD doesn't make wide use of bold style outside of titles and headings. cf. We are already familiar with its general use of italics to render text wrapped within the transChange element, and elsewhere.

Those of us who are used to a bicameral writing systems (scripts with upper and lowercase letters) have constant clues while reading the Bible that help us recognize whenever a word is a name. The initial letter of a name is capitalized, eg. David or Troy.

But many languages have non-Roman scripts and lack the distinction between lowercase and uppercase in their typography. This I suggest may make it more difficult for Bible readers to see at a glance if a word is a proper noun, i.e. the name of a person or a place, etc. For new Bible readers (hitherto unfamiliar with the whole of Scripture) that's a further minor yet real obstacle to better understanding.

It occurs to me that SWORD developers could help in this regard, providing the font used for the writing system can distinguish between bold and normal styles.

Were there to be a new filter that SWORD could use to toggle how words wrapped in a name element are rendered, then a front-end could display such names in either bold or normal style. A module could have this new key in the conf file:

GlobalOptionFilter=OSISNames

The front-end would have an option in its UI to control this feature.

In order to test the concept once it had been implemented, we would only need a module that has extensive use of the OSIS name element. And we already do!

For the latest KJVA module v3.1 (currently in Beta), in the DC books only (for now at least) there are 4018 text_items wrapped in such a name element.

NB: I used text_items for a reason, as 21 instances are for a bipolar name such as <name>Judas Maccabeus</name>.

At present, none of these w elements have any attributes, but those distinctions can be developed later.

For the Protocanonical Books in KJVA and for the KJV module itself, we have not yet made use of the name element. This is because OSIS 2.1.1 does not allow a name element within a w element.

Marking names must therefore be done outside w elements. This might require that the "backwards greedy matching" (of the Strong's markup) be rolled back first. Meanwhile, an alternative approach would be to wrap each name element within a seg element, e.g. <w lemma="strong:H05731">in <seg><name type="geographic">Eden</name></seg></w> which does validate.

If you would like to see SWORD enhanced to support this proposed feature, then please encourage the API developers to start to code for it, and encourage the front-end developers to start to include it in the UI. The rendering would be done by the SWORD engine, but the UI would provide the option do switch it on or off as the user may choose.

Best regards,

David

Sent with Proton Mail secure email.


More information about the sword-devel mailing list