I include everything in "ODF?" that is in the ODF 1.0-1.2 specifications. I include the provisions of for <text:tracked-changes> and the special ones under the spreadsheet content.xml (I can't look at the spec right now because I am doing server maintenance).
It might be reasonable to maintain the segregation or do a better unification. I think that applies in all of the cases in which ODF? occurs. Even though the details I personally focus on are related to <text:tracked-changes>, that is because anything we do is a show-stopper if we don't get this case right. That may be true for the spreadsheet row-column manipulations too, but I don't feel comfortable with them, at this point, to offer any appraisal. I'm trusting you to drag our attention to those.
Also, I can see a case where a spreadsheet cell, the content of which is a <text:p> or other text-content pattern, is subject to text-level change-tracking along with spreadsheet row-column manipulations, so I don't see these things staying completely separate.
- Dennis
-----Original Message-----
From: Andreas J. Guelzow [mailto:]
Sent: Sunday, April 24, 2011 15:52
To:
Cc:
Subject: Re: [office-collab] Let's Keep the Cases Straight
On Sun, 2011-04-24 at 15:34 -0600, Dennis E. Hamilton wrote:
> I notice that there is a tendency to view whatever already happens using the ODF 1.x change tracking as part of ECT and that GCT escapes it.
>
> I think we need to look at the choices in front of us where
>
> a. ODF? is what there is in major circulation now, more-or-less related to what is in the ODF 1.0-ODF 1.2 specifications (the ? related to the uncertainty around bugs and limitations in implementations and the specification too.
>
The way read ODF1.0 to 1.2, there are really 2 change tracking mechanisms defined. One for text documents and one for spreadsheet documents. Which one are you referring to?
> b. GCT is the general proposal that uses the DeltaXML approach as
> a basis
>
> c. ECT is the approach to extending change tracking using ODF? as a starting point.
>
> That means we have these to wonder about:
>
> 0. ODF? as it is or maybe tidied up a tiny bit (I number this 0
> because I don't think we can get away with this)
>
> 1. GCT
>
> 2. ODF? + GCT (unless we believe we'll ever get away with abandoning
> ODF? without a way forward, even if it means a protracted deprecation
> period, sort of like OOXML part 4.)
>
> 3. ODF? + ECT (the only place where ECT applies)
Since ODF? really consists of 2 parts, one could argue that there are many more possibilities.
Andreas
