Dear All,

I am new to the list and have followed the discussion that Elena's
message inspired last week.  I even responded but the system bounced it
back.  I hope this one gets through.

I have been studying P5 and just completed a TEI workshop.  I have a
small grant to develop a program (involving SVG) enabling a user to
designate, on a MS image, the page's various "revision sites" and then
transcribe the sequencing of revisions in each site.  This will be a
part of a larger critical archive on Melville.  A tall order.

I can relate to Elena's concerns.  I am not particularly clear on the
utility of <subst>.  It can link (I think) a specific del to a specific
add, but the subst might also be linked to another revision somewhere
else, so I am wondering if it wouldn't be "cleaner" just to xml:id all
dels and adds.

Elena's concern about the timing of revisions is well founded.  The best
one might be able to do is establish a relative sequencing, within a
revision site and sometimes with steps in that site to steps in another
revision site.  I plan to designate revisions sites on a given ms page
(as I did in my Typee electronic edition put out by Uof Virginia's
Rotunda imprint).  And I think I will be xml:id'ing each del and add,
and @seq'ing them within the given revision site.  And using links to
given dels and adds in the site to del and adds in other sites (if
appropriate).  

I hope this makes sense.

yrs,
John Bryant

___________
John Bryant, English Department, Hofstra University, Hempstead, NY 11549