Yuck. Too confusing.
Only to human powered markup engines :-).
And how do I encode text that is
two levels of smaller and contains a piece that is only
one level smaller? That is: [example slightly reformated by dse]
<smaller> two points smaller
<smaller> fourpoints smaller
<BIGGER> two points smaller </BIGGER>
four points smaller </smaller>
</smaller>
would be:
11 (normal, or some degree of small)
<smaller> 9
<smaller> 7 </smaller>
9
<smaller> 7 </smaller>
9 </smaller>
11
Which works identically for either system, within the context I specified.
There is a subtlety here which involves bigger contrasted to /smaller.
My form enjoys a shallower max nesting depth (3/2). Taken to extremes, your
scheme could end up testing the depth of a short stack for text which yo-yos
while descending in size (6/3 in the following):
11 11
<smaller> 9 | <smaller> 9
<smaller> 7 | <smaller> 7 </smaller>
<BIGGER> 9 | 9
<smaller> 7 | <smaller> 7
<smaller> 5 | <smaller> 5 </smaller>
<BIGGER> 7 </BIGGER> | 7 </smaller>
</smaller> | </smaller>
</smaller>
</BIGGER>
</smaller>
</smaller>
</smaller>
Your form has the advantage of being less context sensitive.
--
dana s emery <de19(_at_)umail(_dot_)umd(_dot_)edu>