So atm im a -0 on this as i know of active users. So i don't think its a good idea.
But as i personally dont contribute in this area i cant say no to it. Get Outlook for Android On Thu, Mar 7, 2019 at 6:50 PM +0000, "Justin Bertram" <jbert...@apache.org> wrote: I understand that there are users who still use those clients. My concerns are: - no real community around those clients - no active development - no apparent vision for future improvements - small pool of developers who've made commits in the last few years - based on OpenWire If users continue to use these clients they're going to continue to find bugs. However, it appears to me that the community resources to address such bugs have been dwindling for some time. I'm not sure we can draw conclusions that the current low level of development is down to the stability of the code-base or low usage or developer apathy or whatever. However, we know that NMS & CMS are not built around an industry standard protocol which means the potential pool of users, developers, etc. is quite severely limited. I think we can also say that OpenWire doesn't have a particularly bright future. Given all this I think we'd serve users better by pointing them to other options, particularly to AMQP clients where they can join an active community. If they need imperative (i.e. non-reactive) APIs they can advocate for those in those communities. We can leave all the NMS & CMS documentation and downloads available while clearly marking the projects as deprecated/retired. Justin On Wed, Mar 6, 2019 at 7:08 PM wrote: > So these are still being used by users. We have a group in my org using > them. Not everyone in .net and c++ is reactive... some prefer the jms like > apis. > > > > > I think low level of development is down to stability. > > > > > Get Outlook for Android > > > > > > > > On Tue, Mar 5, 2019 at 11:37 AM +0000, "Christopher Shannon" < > christopher.l.shan...@gmail.com> wrote: > > > > > > > > > > > I would be fine making these projects deprecated but I think there was some > pushback from users on this in the past. But I am all for deprecating > projects that are no longer maintained actively and have alternatives such > as AMQP clients. > > On Fri, Mar 1, 2019 at 2:13 PM Justin Bertram wrote: > > > As I've been looking at updating the ActiveMQ website I've wondered about > > the status of both NMS & CMS. Contributors, commits, and releases for > them > > appear to have been low historically and have dwindled recently even > more. > > > > As I understand it, one of the goals of updating the website is to be > more > > up-to-date and clear so that users can make informed decisions on what > > software to use. In that vein, should we change the "status" of these > > projects? > > > > To summarize what I found, there are 9 repositories* total for NMS & CMS. > > In the last 3 years there has been a total of 26 commits by 3 developers > > (with 1 developer doing 23 of those commits). > > > > These days users have client options across a wide variety of platforms & > > languages mainly due to the proliferation of AMQP (and STOMP to a lesser > > degree). Also, the "JMS style" interfaces of NMS & CMS have waned in > > popularity as application development has moved to a more "reactive" > > approach. Does it make sense anymore to maintain our own stable of > > interfaces & clients? Should we mark these as retried or deprecated? > > > > > > Justin > > > > * https://github.com/apache/activemq-cpp > > https://github.com/apache/activemq-nms-msmq > > https://github.com/apache/activemq-nms-zmq > > https://github.com/apache/activemq-nms-ems > > https://github.com/apache/activemq-nms-xms > > https://github.com/apache/activemq-nms-stomp > > https://github.com/apache/activemq-nms-amqp > > https://github.com/apache/activemq-nms-openwire > > https://github.com/apache/activemq-nms-api > > > > > > > >
