All, Glad to see we got this ball rolling. I wish I could have participated. I plan to participate in the next round of discussions. Some follow up thoughts.
1) On the OT, I think we should put up a wiki page and start listing some options and pro's and con's. 2) On the product vs platform. I think we need to build a solid platform first, on top of which we can build a solid product. 3) We are struggling with some hard questions. There are 3 or 4 major things on the table right now. A. Core OT Algorithms B. Core Data Model C. Client Server Decoupling D. Client Server Protocol Changing most of these will potentially be disruptive. We need to figure out the implications and cost benefit of these things. To move forward we may need to rearchtiect some things. We also at the same time need to drive towards a release from an apache point of view to grow within this system. There are parallel paths here. 4) I think we need to really put a roadmap up on our web site that lists all of the issues we think we need to tackle in some priority order. Good work guys. ~Michael On 7/31/13 4:42 PM, "John Blossom - Shore Communications Inc." <jblos...@shore.com> wrote: >Joseph, > >I agree that the fundamental issues that you raise are the key of how to >get development moving. "Mobile strategy" is there at this point simply to >say that whatever solutions that we have, they need to be able to work >well >in a mobile environment. My own opinion is that Apache Wave needs to be a >platform from which multiple products can be built, with default apps that >demonstrate its power and act as benchmarks for service and component >integration. So I am all for more of a deep dive into OT, to getting some >consensus on other core architecture issues and to getting a platform >released that can reflect those decisions. It's up to the community, >ultimately - I want the most motivated people to step forward and to drive >these things as soon as possible. I see my job mostly as trying to help >drive talent and respect our way through improved communications and >helping to formulate the overall requirements as production and proof of >concept code begins to move forward. > > >On Wed, Jul 31, 2013 at 4:11 PM, Joseph Gentle <jose...@gmail.com> wrote: > >> I still want a technical deep-dive on OT at some point. >> >> Issues: >> - Are we building a platform or a product? (In my opinion our platform >> isn't good enough, and thats where I want to focus our efforts for >> now) >> - What OT algorithms should we use? >> - If we use Torben's OT system, how expensive is it actually for daily >> use? Is there a way to make it fast and still sign operations? Is >> there a way to use OTR? >> >> This stuff is more important than figuring out our mobile strategy for >>now. >> >> -J >> >> >> On Wed, Jul 31, 2013 at 10:15 AM, John Blossom <jblos...@gmail.com> >>wrote: >> > Thanks to Alex North, Ali Lown, Bruno Gonzalez, Jérémy Naegel, and >>Joseph >> > Gentle for their active participation in our first Apache Wave team >> > hangout. We also saw Christian Grobmeier trying to connect briefly, >>and >> > some others are listed as having went, but that was the core group >> online. >> > You can view the event at: http://www.youtube.com/watch?v=QNF_-ogcoIA >> > >> > We had a pretty good discussion, though it would have benefited from >> others >> > more focused on OT, no doubt. I think that we've demonstrated how to >>use >> > the technology to communicate in the community and to further >>leverage it >> > to make the world more available of Apache Wave development and where >>it >> > stands. I don't what to be the permanent host of these hangouts, or >>any >> > person-to-person communications, but I am glad to help facilitate >>them as >> > needed. It's a communication tool, and any way that we can use it as a >> > community in ways that reinforce its goals and values is good. >> > >> > The main question coming out of this hangout from my perspective is >>this: >> > how to we choose a path for development that will result in a powerful >> > solution for mobile collaboration that can support multiple apps in a >> > common data model? I don't know that a Hangout is the place to solve >> that, >> > this is probable the best place to start. In the meantime, I will be >>glad >> > to help create a new "office hours" event on Hangout and have people >> decide >> > what they want to discuss. Thanks to everyone who participated, the >> > logistics seem to have worked out pretty well. >> > >> > John Blossom >>
