Thanks Adam for clarifying the questions. Also I agree with Upayavira, the primary discussion it might be more about "ideas" and the community's "engagement" with them. After that, tech aspects would come.
So, in this regard I would like to share some thoughts about SwellRT as a product... a) Where SwellRT fit in the market? Competitors? SwellRT current vision is closer to products like Firebase, Meteor and Realm. They are new breed of frameworks/platforms to write apps. They provide as key feature, real-time data storage with simple document-based data models. Their aim is to simplify and speed up web/app development. And of course, they allow to build real-time collaboration features easily. Of course, these projects are matured, but they still have pros and cons. What it seems clear to me is the trend: to develop heavier apps/webapps (because nowadays devices have a lot of computing power and it means just coding for one system) and lighter backends providing common "services" (notifications, storage, auth...). b) What Wave/SwellRT's tech could offer in this market as innovation? Wave/SwellRT could compete with features like: - Open Source and JVM world: I guess there is still a part of the world happy to see a Java friendly framework, despite it works for Web (but hopefully for android/iOS). - Simpler API: with sugar syntax, for example, in SwellRT we are working in a JS syntax just based in mutable objects. Also with API concepts easy to understand: objects and participants. - Full-featured collaborative writing: Wave was designed for text editing, whereas these new frameworks are focused in JSON. For example, annotations is a cool feature not easy to provide I guess. Also the Wave's text editor is very good yet. - Federation: it is the hardest selling point for developers in general because it doesn't provide benefits in the short term. However, it is the entrance to innovative things like cross-app interoperability, organic scalability... 2016-10-05 23:47 GMT+02:00 Upayavira <u...@odoko.co.uk>: > I want to see a proposal regarding importing SwellRT that gives me > confidence that bringing SwellRT into Wave will actually lead to an > active project. > > A way this could be achieved *before* bringing SwellRT would be if > everyone who is interested in contributing headed over to SwellRT, and > started contributing over there. Then, we'd be bringing both code and > community into Apache, which would give me far more confidence than just > importing code but with no confidence that anyone is actually going to > do anything with it. > > Upayavira > > On Wed, 5 Oct 2016, at 10:03 PM, Adam John wrote: > > Pablo, a lot of great information in this slide deck. I hope others have > > a > > chance to review as well. Outstanding work. > > > > Price, very thoughtful responses. I agree with the overall conclusion - > > SwellRT should be brought into Wave. > > > > I like the idea of moving the SwellRT fork in to replace the current > > branch > > of Wave development because it moves the project reasonably forward and > > makes sense overall. It does not seem anything current would be lost in > > that move. It seems like we have everything to gain. However, there > > might > > be work in progress that is affected. > > > > It would be great if contributors on the project took a look and shared > > some thoughts. > > > > Q3) For current contributors; are you in favor of bringing the fork home? > > > > - > > Great attendance at our last meeting, and familiar ground was covered. > > (agenda > > and notes > > <https://docs.google.com/document/d/11j_WQGYAtDlN8Wqx8jJglPpw6tJznvMGf > dLOvQu96i0/edit>) > > We're largely covering the next steps in recent emails. > > > > If the group agrees, that we should bring SwellRT into Apache Wave, then > > there needs to be a proposal drafted. > > > > Q4) Does anyone have interest, experience or desire to help with this > > task? We do not expect to start until after the next meeting. > > > > - > > Perhaps 2-3 weeks is time enough to consider the questions posed? > > I'd like to plan our next steps; > > I suggest *10/26 as the next discussion* - based on consensus in the list > > of course. > > > > The goal of the next meeting will be to provide a chance to address any > > questions regarding bringing the projects together. Perhaps this could > > be > > a technically deeper discussion. > > > > Q5) Does anyone have interest in a standing co-work session? Especially > > important would be current contributors. I think this could be a good > > way > > for some on the list that have stalled or reached impasse to begin to > > make > > progress in helping out. > > > > Thanks, everyone for your work and efforts. I believe that if each of us > > does just a little bit over the next few weeks we will continue to see > > the > > progress we need in this project. > > > > Adam John > > (914) 623-8433 > > Google+ <http://google.com/+AdamJohn1> | LinkedIn > > <http://mradamjohn.com/> > > > > On Wed, Oct 5, 2016 at 12:34 PM, Pablo Ojanguren <pablo...@gmail.com> > > wrote: > > > > > Thanks for your answer Price, > > > > > > I guess we should not delay this discussion... > > > > > > I'd happy to run another call if you think it can move things forward. > > > > > > > > > > > > 2016-10-01 18:40 GMT+02:00 Price Clark <gpwcl...@gmail.com>: > > > > > > > Pablo, thanks for the presentation. > > > > > > > > While my qualifications to answer this are 0 getting to listen to > > > > Upayavira talk this week (the last Apache mentor if I'm not mistaken) > > > make > > > > me feel the answers to 1 and 2 are easy to answer. > > > > > > > > 1.) Upayavira communicated very fervently that there just isn't > enough > > > > oomph in wave's development. Every year around the time that the > > > retirement > > > > conversation is brought up, activity similar to this starts brewing > and > > > > then it all dies down in a few months. From this perspective "Does > > > SwellRT > > > > tackle current Wave problems?" The answer is unequivocally yes, > SwellRT > > > is > > > > a more actively maintained fork of Wave and given the slowing/slowed > pace > > > > of Wave *a merge with SwellRT is likely the only way to save this > > > project*. > > > > > > > > 2.) I would also like to bring up another point Upayavira made, > > > > "Communities are built around good ideas and bad code." Running with > > > that I > > > > thing that good ideas attract tinkerers and 'people with ideas' that > > > could > > > > eventually become 'contributors with ideas'. In some senses SwellRT > > > > splinters Apache Wave in a way that developers on this mailing list > have > > > > been alluding to for a while. The client side code is not well > understood > > > > and is definitely in the way of the server. SwellRT has a more > general > > > goal > > > > of supplying a server that is capable of powering a front-end like > the > > > > original vision of google wave. This means that merging with SwellRT > > > would > > > > force a separation of the client and server and allow for people with > > > > interests in either a front or back end to work in tandem. This seems > > > like > > > > an ingenious way to attract more people; anyone with an interest in > the > > > > backend technology OR a way to use said technology in an application > > > could > > > > be a potential contributor. Unless I'm mistaken it seems like SwellRT > > > > offers a set of features that could be classified as a superset of > Wave's > > > > features. So, it seems like most or all of SwellRT would be at home > in > > > > Wave. Pablo also reasonably stated that he'd prefer to work in one > > > project. > > > > > > > > As for me, as soon as a direction is clear I would love to talk to > > > > someone actively maintaining/writing code so I can help them > contribute > > > to > > > > whichever code survives in whatever way possible. > > > > > > > >
