[sword-devel] OSIS recommendations to SIL/JAARS

DM Smith dmsmith555 at yahoo.com
Wed Jul 25 04:19:21 MST 2007

On Jul 24, 2007, at 8:48 PM, Kahunapule Michael Johnson wrote:

> You can revert to the rather early HTML form of
> <p> with no </p>, which works fine

There are at least two things wrong with this:
1) It is not xml and won't validate as well-formed xml. Every open  
tag needs a corresponding end tag or needs to be marked. So you need  
either <p></p>  or <p/>.

2) It is semantically invalid. There is no content to the paragraph.  
Remember a processor is free to style a paragraph in any fashion,  
including ignoring containers without content. (Such as HTML does  
with an empty <div/>)

We are used to html processing producing a blank line between  
paragraphs. This is a matter of presentation. It is equally valid for  
a paragraph to start on the next line and be indented. In HTML this  
would be equivalent to <br/>&nbsp;&nbsp;&nbsp;&nbsp; Or merely start  
on the next line. Or as I have seen in one publication, odd number  
paragraphs have margins shifted by a half inch.

Imagine a culture where paragraphs don't start on a line, but are  
preceded by one symbol and ended by another, with inter-paragraph  
markers being elided into a new symbol.

Note: OSIS does not presume any presentational effects with a <p>. It  
is not the equivalent to HTML's <p>, but rather TEIs. In HTML <p> is  
a block element. In TEI it is a content element. The difference is  
that blocks stack and content is merely content.

See: http://www.tei-c.org/release/doc/tei-p5-doc/html/CO.html#COPA

My suggestion to use <lb/> with a type attribute indicating whether  
it is a substitution for a begin paragraph or an end paragraph allows  
for the full semantic of <p> and </p> (stylesheets can give whatever  
meaning they want to either) and gives a nod that most people expect  
paragraphs to be stacked and visually separated. Thus a processor  
that is unaware of the type attribute will probably give an expected  
visual representation.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.crosswire.org/pipermail/sword-devel/attachments/20070725/c1e4a3aa/attachment.html 

More information about the sword-devel mailing list