On 27/12/12 07:17, Saeed Sarrafzadeh wrote:
> ... If it's decided to adapt the Lite to the new version of the
> guideline and to restrict those elements to the interventions made by
> persons other than encoder and editor, then at least the <supplied>
> element should be added to the Lite.
I considered this possibility, but decided against it, on the grounds
that adding yet another element to Lite was not a good idea, especially
since for the majority of lite users this distinction is not likely to
matter at all. Lite isn't meant to solve everyone's problems! Instead,
I have added a note to the text pointing out that there are other ways
of doing things if you use additional elements from the Guidelines,
similar to the existing remarks about abbreviations.
> Also I faced with a more general case in this regard. It's the problem
> with understanding the @resp use cases: when we are faced with an
> existing document in a hard format (printed or manuscript) we faced
> with some existing mark-up in the text such as highlitings, indexes,
> ... and one may want to encode such features. In the course of marking
> such features in the new xml-form of the text, he/she may make some
> decisions on what that markup means in each case (for example it may
> be an emphasis, a foreign word, a term , ...). In cases where the
> encoder faced with an addition or deletion or so, the @hand can be
> used to distinguish between the responsibility for the feature itself,
> and the responsibility for the encoding using that element. The other
> features like highliting, ... how can be handled in this regard? Who
> should be registered as the responsible person in such cases? The
> encoder (who decided for the mark-up)?
Yes, in general. The @hand attribute indicates the agency which the
encoder believes is responsible for writing something. The @resp
attribute indicates the agency which the encoder believes is responsible
for a textual modification captured in the markup. All markup in a
document indicates the belief of the person who marked it up --
sometimes those beliefs concern other agencies whose actions are
(believed by the encoder to be) discernible in the text.
>> An <interp> can be pointed to from any element, not just a seg. A <span>
>> for example could be associated with one.
> Of course, the <interp> can be point to from any element, but that element can't cross the hierarchy of the document as noted as the first restriction in the alternative practice of using the <seg> with a @type. This restriction seems to be only resolved using the <span> and <anchor> elements (<span> for the interpretation and <anchor> for marking the locations in the main text). Please let me know if I'm wrong.
No, you're not wrong, and I'm grateful to you for calling attention to
this misleading passage. I've now revised it, hopefully for the better,
and also reinstated the missing @corresp attribute. This bug-fixed
version of Lite will be available at the next release of P5, scheduled
for mid January next year.