On Fri, Jan 6, 2012 at 5:57 PM, Greg Reddin <gred...@gmail.com> wrote:
> On Fri, Jan 6, 2012 at 4:39 PM, Alex Harui <aha...@adobe.com> wrote: > > Mentors, what criteria would you use to determine if Blaze should be its > own project? > > I would say it would be its own project if it has its own distinct > community. The two communities can certainly overlap, but if there's a > significant number of folks in one community, but not the other, it > would become its own project. > > It's really no problem to start together, then separate at a later > time. I suggest we cross that bridge when we come to it. > > Greg > I pretty much have two views of this one : - Flex SDK and Blaze DS are completely different models, and came from completely different houses. Server Side / Client Side, it came from a paid product vs. it was always a free product, etc. They should be separate for that reason alone. - BlazeDS completely depends on the Flex SDK at this point. When Adobe spun off BlazeDS into it's own project and let it rot on the vine for so many years, they killed off whatever community existed around it. No updates, and when the community submitted updates / patches / enhancements to Adobe, everything was killed on the spot because it wasn't compatible with their pay-for version (LCDS). I personally got in some very tarse conversations with Adobe's engineers about adding a community version of NIO to BlazeDS and got shot down (and told some other nasty things because it competed with their own product). Additionally, they made it really hard by locking away the source to all the communications SWCs for Blaze to introduce anything above and beyond. A lot of people still use BlazeDS, and a lot of major businesses still use it. Does it have a developer community behind it anymore? Doubt it. At the last 360|Flex, I think there were like 10 people in the room when we did a birds-of-a-feather about it. Are there people who would be able to pick up and develop for it? Maybe a very select few, but I doubt they want to do very much with it -- maybe get some bugs fixed that have existed for the last 3 years. I think before I would vote on it, I would need to see what Adobe is really donating, and how they are doing it. Are we getting the FULL rpc.swc with it? Are we getting some of the features that were developed later (like NIO), or will Apache be sued if we develop those types of features because they exist in Adobe's commercial, heavily patented product? Same goes for data management, the other bright star that I would want to have included one day... -Nick
