[osis-core] <emph> and formatting in OSIS

[Chris Little]

Harry Plantinga wrote:
> Chris,
> 
> I think I understand your concern about formatting, rend,
> emph, etc.  Correct me if I'm wrong.  Maybe you're thinking it's
> going to be very hard to USE osis in your software and 
> get attractive results without formatting in the document.
> Maybe you're thinking that parsing a stylesheet, handling XSLT,
> etc. would be a Big Pain to have to do in your software.
> (I may be way off in left field; if so, please forgive and/or
> ignore me.)

My concern hasn't got anything to do with our software.  We do all our 
transforms manually and we'll manage, whatever is decided upon.

My concern is that emphasized text, by which I mean text that is 
rendered differently than surrounding text for no reason other than 
emphasis, is extremely common yet is not dealt with by OSIS.  It is 
common in old texts that will be encoded in OSIS and it shall remain 
common in new texts yet to be written.

The only way to encode emphasized text is with the seg element, which 
was added for all the unforseen cases where a person might want to mark 
small segments of text.  Emphasis is not an unforseen case.  I've been 
asking for a means to denote it since January, and really have yet to 
hear a good reason why <emph> should not be added.

What I'm asking for is not a means to dictate rendering style in the 
document, but a way to indicate that a section of text is different and 
in what way it is different from that surrounding it.  <emph 
type="italic"> shouldn't mandate that the text it contains be rendered 
in italics but that it was encoded with italics in mind or was copied 
from a print publication that rendered it with italics.

Even if we do not add <emph>, we must identify standard type values for 
the seg element to indicate types of emphasis.  I don't see why we 
should be planning for de facto standards to make up for the failures of 
a standard that is not yet completed when we have the ability to prevent 
those faults from being in the final standard.

--Chris