Re: best practices for <l> and manuscript transcription

In my opinion, there are two approaches to assinging IDs ( <at> xml:id).

One is to have a scheme for IDs designed to avoid collisions, much like 
the one you suggest.  There's a similar scheme for divs in Humanities 
E-Book XML titles ( 
http://www.humanitiesebook.org/xml/doc/acls-hebook-doc.html#divs ).  The 
trick here is that if you revise your encoded text later and add or 
remove an element, you have to decide whether to change your IDs to 
clean things up or break the system.  So if you think your text will be 
frequently revised, you might instead assign random ID strings to each 
line.  XSLT can do this easily for you. With such IDs, it won't be 
awkward if you add a line later.

As for numbering elements ( <at> n), you should think about what is the most 
stable in your text.  Will pagination likely change in the future as the 
text is presented in different media?  If so, don't restart numbering on 
each page.  Humanities E-Book numbers sequentially through the text ( 
http://www.humanitiesebook.org/xml/doc/acls-hebook-doc.html#para ).  I 
believe the intention is that users could cite a particular paragraph 
number for the book as a whole, not a number relative to a certain 
chapter or page.  Absolute numbering works well with MLA style, which 
basically recommends citing such numbering when available.

--Kevin

On 7/22/64 2:59 PM, Karen Desmond wrote:
> Hello,
> I'm wondering if anyone has thought through some best practices for assigning
> an xml:id to a transcription of a line of text in a manuscript.
> I am using the<l>  tag at the moment, until the new<ge:line>  element is
> approved.
> I am thinking of something like this
> <l n="1" xml:id="Cn54.1f57rl5">  where this means, manuscript sigla Cn 54.1,
> folio 57 recto, line 5.
> My dilemma is, should I put n="5" because although it is the first line of my
> text, it is the fifth line on that particular page. And then do I restart the line
> numbering with each new page of my manuscript?
> I will be using these transcriptions in Juxta, that generates a critical apparatus
> based on the line number, hence my wanting to start at line 1.
> My concern is to provide a clear-cut way for scholars to be able to cite a
> particular line. On my website I will be providing separate diplomatic
> transcriptions of each manuscript source of my text (with editorial emendations
> and corrections provided in notes), and then an automatic collation generated
> by Juxta.
> (their automatic collation is standalone at the moment, although they say it will
> be purely web-based from version 2.0 on)
> Thanks!
> Karen