I suggest that the real differentiator should be whether the item
is useful, persistent, human-readable metadata or if simply transitory machine-readable
data. The former managed as elements and attributes and the latter managed as
PIs.
I think change tracking is very transitory and of little value
as human-readable metadata (unlike change history or review comments which are
human-readable). Change tracking really has little value outside of the editing
environment. Change bars and other change tracking marks do not usually persist
beyond a single revision.
For what it is worth, I think this is a good use of PIs.
Cheers,
Rob Hanna
From: Ogden, Jeff
[mailto:]
Sent: October 2, 2009 2:23 PM
To: Michael Priestley; Park Seth-R01164
Cc: dita
Subject: RE: [dita] Standard DITA processing instructions?
I think the DITA TC could include discussion of “DITA
standard PIs” in the spec.
I think having something standard for change tracking would be a
good thing, but I wonder if that might not be better done in a way that would
work for DITA and non-DITA document types. And if you buy into that, the
question is what group would be a good one to work on it? I guess it
could be a recommendation from the DITA TC that had wider application than just
DITA, but I can imagine that there may be some other group that might be
appropriate and willing to do that work too.
I think you have to look at using PIs vs. more traditional
element/attribute markup on a case by case basis.
An advantage of PIs is that implementations that don’t
understand or support a particular PI should ignore them. That
isn’t as true for element markup.
In the case of change tracking, I think there could be some real
advantages to using PIs.
-Jeff
From: Michael Priestley
[mailto:]
Sent: Friday, October 02, 2009 2:12 PM
To: Park Seth-R01164
Cc: dita
Subject: Re: [dita] Standard DITA processing instructions?
My usual
thought is that if we want to standardize it, why make it PIs, which by their
very nature cannot be controlled by a schema?
For example,
for change tracking there's already some attributes that might be used - but if
we wanted something new, we could create a domain specialized from <data>
and make it broadly available.
Michael
Priestley, Senior Technical Staff Member (STSM)
Lead IBM DITA Architect
http://dita.xml.org/blog/25
From:
"Park
Seth-R01164" <>
To:
"dita"
<>
Date:
10/02/2009
02:04 PM
Subject:
[dita]
Standard DITA processing instructions?
I hope this is
not taboo to suggest, but...
Is it in the TC
purview to provide a way for DITA application developers to undergo some level
of coordination for processing-instructions for common user events?
For instance,
if there were a common nomenclature for "change tracking" PI
notation, an author, editor, reviewer, and publisher could use the most
appropriate tool for his/her specific function and the change tracking feature
would work across those different purpose-specific tools.
Something like
this would allow DITA files to be exchanged where not only the content is
guaranteed to be interoperable, but also commonly used application
"features".
bracing for the
blow,
-seth
