Entirely correct
Get Outlook for Android On Fri, Mar 8, 2019 at 2:02 AM +0000, "Justin Bertram" <jbert...@apache.org> wrote: Ultimately I'm much more concerned about the development resources committed to the code-base than I am the number of end users. I think it's bad for the credibility of the project to promote a piece of software that doesn't have good developer support. Five people have commented on this DISCUSS thread, but none of them have volunteered to support the NMS and/or CMS code-base. I certainly don't blame anybody for that because we're all free to work on what we want, but if we can't identify a handful of developers who are committed to these code-bases then I think we should mark them as deprecated. I don't really see deprecation as changing the current status quo, and I certainly don't think it's taking anything away from existing users. If the code-base is as stable as you suspect then users can just continue on using it. If problems arise anybody is free to fork the code and fix it themselves, etc. This is open-source after all. If the word "deprecated" is off-putting we can use "end-of-life" or "retired" or something else. I think we just need some kind of caveat on main NMS & CMS pages that indicates there are no developers committed so there's no guarantee of bug fixes or future releases so they may be better off with an alternative. The ActiveMQ website has had so much stale information for so long I think it would be very good for the project to be clear about project status. The community went through the same kind of thing just a few years ago with LevelDB. Justin On Thu, Mar 7, 2019 at 7:04 PM wrote: > I think the use of nms api is a little more active than you believe. > > > > > > > Then the open wire one has a download on nuget of the last release alone > 167k times according to stats here > > > > > https://packages.nuget.org/packages/Apache.NMS.ActiveMQ/ > > > > > > > Like wise theres a quite active netstd (not apache) that uses nms api, > last release 7 months ago. > > > > > > https://packages.nuget.org/packages/Apache.NMS.ActiveMQ.NetStd/ > > > > > Get Outlook for Android > > > > > > > > On Fri, Mar 8, 2019 at 12:46 AM +0000, "Justin Bertram" < > jbert...@apache.org> wrote: > > > > > > > > > > > Thanks for clarifying the condition of your NMS AMQP provider contribution. > > To be clear, the NMS website [1] lists 8 providers: > > - *ActiveMQ* - last release in 2016; > https://github.com/apache/activemq-nms-openwire > - *STOMP* - last release in 2013; > https://github.com/apache/activemq-nms-stomp > - *MSMQ* - last (and only) release in 2009; > https://github.com/apache/activemq-nms-msmq > - *EMS* - last (and only) release in 2009; > https://github.com/apache/activemq-nms-ems > - *WCF* - last (and only) release in 2009; can't find any repo > - *AMQP* - never released; https://github.com/apache/activemq-nms-amqp > - *MQTT* - never released; can't find any repo > - *XMS* - never released; https://github.com/apache/activemq-nms-xms > > Then there's a repository at https://github.com/apache/activemq-nms-zmq > for > a ZeroMQ provider that isn't listed on the website at all with no apparent > releases. > > > Justin > > [1] http://activemq.apache.org/nms > > On Thu, Mar 7, 2019 at 3:24 PM Ragnar Paulson > wrote: > > > I've been haphazardly following this discussion. I did some work on the > > NMS API last year, possibly the 20-odd commits you refer to. > > > > It is correct to say that NMS is just a general open interface to any > > transport, openwire is just one. The work I did was for AMQP. I've > > forgotten how many transports are implemented in the NMS project but I > > believe it was somewhere between 3 and 6 (STOMP is there as well). > > > > Also the measure of the importance of the API should be in the > > downloads/use of the API, not in the code modifications. But on the other > > hand, if there are no tickets or bugs being raised it's probable it's not > > being used. > > > > The work I did to create an AMQP transport for Apache.NMS was for > > AMQPnetLite, not for QPID AMQP. The main thing it lacks is nuget > > packaging to be generally available in the .NET world. > > > > That said, I have found my priorities have changed and I'm not the person > > to be the champion here, so I can't argue strongly for maintaining a > > development tree that I won't likely contribute to very much. It would be > > nice to see it carried on, but if the community as a whole does not see > any > > use for this then better to let it atrophy. > > > > Regards, > > Ragnar > > > > > > On Thu, Mar 7, 2019 at 3:58 PM > > wrote: > > > > > I think he was wanting to make a new one that wrapping latest qpid, > > which > > > would be awesome. > > > > > > > > > > > > > > > Get Outlook for Android > > > > > > > > > > > > > > > > > > > > > > > > On Thu, Mar 7, 2019 at 8:48 PM +0000, > > wrote: > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > So there is already an amqp nms implementation > > > > > > > > > > > > > > > Get Outlook for Android > > > > > > > > > > > > > > > > > > > > > > > > On Thu, Mar 7, 2019 at 8:08 PM +0000, "Clebert Suconic" < > > > clebert.suco...@gmail.com> wrote: > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > There was a guy at some point trying to implement AMQP on NMS. what > > > happened to that front? any knows about it? > > > > > > On Thu, Mar 7, 2019 at 3:04 PM wrote: > > > > > > > > So the point of these is they provide a clean api regardless of > > > underlying protocol. > > > > > > > > > > > > > > > > > > > > Its not based on openwire so i disagree on your point there, it is > > > providing a higher level api abstraction. Which open wire is just one > of > > > many protocols implementing the api. > > > > > > > > > > > > > > > > > > > > E.g. amqp switch over from open wire here are clear points for a > > > developer to switch from one to the other, as there is an > implementation > > > for both against nms, its very easy to switch over. > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > -- > > > Clebert Suconic > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > >
