> Yes, we are using a very flexible search engine that can process
> any kind of elements.
Except, apparently, empty ones :-)
Seriously, it sounds cool -- what are you using?
> Maybe I've been influenced by the Italian meaning of the term
> "sequenza"... from a narratological point of view is a
> homogeneous series of narrative elements. Is it possible to use
> the word "sequence" in English with this meaning?
Now that I've read your description of what's *in* the <sequence>s,
it seems much more sensible -- you seem to be using "sequence" as a
generic description of what's being contained by the element, not as
a name to indicate it is in parallel with another similar element. In
which case "sequence" might not be my first choice, but at least it
makes sense. Of course, you should feel free to use the Italian
BTW, I included all of the normal div* attribute classes in the
declaration I posted. If you are confident you won't be using any of
the attributes in a particular attribute class, you may wish to
remove it from the declaration.
a.divn: type=, org=, sample=, part=, met=, real=,
a.global: well, alright, you shouldn't remove this one
 Only useful if you have several different classifications of this
same sort of beast -- sounds unlikely given your description.
 Definitionally org="uniform", so not needed.
 My recollection is that you've already chosen to use next= and
prev=, so may not be needed.
 Only if you're using the verse base tagset. These attributes are
available just so you can specify defaults for all the poems or
aria in a given div*, lg*, or now <sequence>. Unless there is
some metrical consistency or pattern to when chunks are performed
in parallel, it may not make sense to use them here.