Adding Svante Schubert to the thread, from the ODF Toolkit project. He also chairs the subcommittee at OASIS that has been looking at OT for change tracking in ODF.
On Sun, Jun 16, 2013 at 6:15 PM, Michael MacFadden <michael.macfad...@gmail.com> wrote: > > > On 6/16/13 2:51 PM, "Michael MacFadden" <michael.macfad...@gmail.com> > wrote: > >>Rob, >> >>I would be interested in continuing this conversation. I have been >>working with the top minds in OT for the past few years. I am excited to >>hear the OO is interested in an OT supported mechanism. How far along are >>you in the process? >> It is very early and mainly happening in the standards committee at OASIS. The ultimate aim is to have something that could work across applications, not just between two OpenOffice instances. So this requires a sensitivity to the document model abstraction, to work at the ODF level, not just with an application's internal view of a document. OpenOffice committers are involved in the standardization side of this, as well as LibreOffice and Calligra and Gnumeric, as well as Microsoft. Initially it is about defining the document model, in a way that makes sense to the user. Since tracked changes are visible to the user, to approve or reject, we need it at a granularity that makes sense to them. Then based on those primitives, and the associated actions, we can develop an XML-based notation for expressing the state transformations. That gets us to the static/stored form of traditional change tracking. Not in plan officially is the next step, which would be the protocols for exchanging such information in real-time. But it is a possibility (even a likelihood) that is informing our design decisions. We're mindful that the real-time collaborative editing is the logical next step and we're trying to lay the right foundations for that at the format level. One sub-goal, for enabling the real-time side of this, would be to standardize the protocols at some level, so clients from different vendors could do this kind of collaboration in a heterogeneous kind of way. Is there anything in Wave that would be a good basis for a standard? Of course a perfectly valid approach would be to prototype first and then standardize. Regards, -Rob >>~Michael >> >>On Jun 16, 2013, at 11:00 AM, Rob Weir <robw...@apache.org> wrote: >> >>> I'm not subscribed to this list, but Christian Grobmeier pointed me to >>> John's post about how OT and Wave could be relevant to OpenOffice. >>> >>> I wanted to mention that the idea is being discussed, but at the >>> standards level. The default document format for OpenOffice is Open >>> Document Format (ODF), which is standardized at OASIS and ISO. (I >>> chair the committee at OASIS). We're currently working on ODF 1.3 and >>> as part of that we're adding a new change tracking mechanism based on >>> OT. This is the traditional asynchronous change tracking that office >>> suites have had for years, but modeled on OT terms. >>> >>> And, although not specified at this point, we're also aware that OT >>> enables more interesting modes of collaboration, including >>> synchronous/real-time, co-editing, etc. That's the main reason the OT >>> approach is attractive, is that we can have a single model that will >>> work for change tracking as well as co-editing. >>> >>> Once we get the standard side of this elaborated in more details, then >>> the next step will be to get it implemented in Apache OpenOffice as >>> well as the Apache ODF Toolit (incubating). But the pace of >>> standardization is slow, and I wouldn't expect this before 2014. >>> >>> Regards, >>> >>> -Rob > > -- Opinions expressed in this communication reflect the author's individual personal view, not necessarily that of an amorphous collective. The above statements do not reflect an official position of any organization, corporation, religion (organized or disorganized) or national football association. The contents of said note are not guaranteed to have been spell checked, grammar checked or reviewed for metrical infelicities. The contents of this post may not be suitable for those whose native language is not logic. Caution should be exercised when operating heavy machinery when reading this note, or even when not reading it. Seriously, heavy machinery is dangerous. Be careful.
