activemq-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Bruce Snyder <bruce.sny...@gmail.com>
Subject Re: [VOTE] ActiveMQ Artemis becomes ActiveMQ 6
Date Wed, 06 Dec 2017 19:49:55 GMT
I don't disagree with the technical reasons, I call this compatibility
because it's about defining a very clear migration path and smoothing the
effort involved from a user point of when migrating from ActiveMQ 5.x ->
Artemis. In fact, I am going to step forward and start to define a roadmap
for this effort and encourage everyone to participate.

I disagree with discussing or even considering anything the vendors want to
do. Even when I worked for LogicBlaze and then IONA, I disagreed with
trying to drive our company agenda via the Apache ActiveMQ project. But
given that employees of different companies participate in Apache ActiveMQ,
I can't control them and I won't let their company agendas control the
Apache ActiveMQ project.

I had no idea that Clebert was going to call for a vote until I saw it in
flight early this morning. So, I thought, 'well, what the fuck, I'll vote'.
Although Clebert works for Red Hat, he does not care what they do or why
they do it. His full-time job is working on Artemis and he is passionate
about it. Unfortunately, his passion and eagerness get the best of him
sometimes causing him to take action as best he knows. Also consider that
Clebert is not very knowledgeable about the ASF and ways of the ActiveMQ
project.

With my thought to step forward and create the roadmap, I intend to work
with anyone who wants to participate to identify the tasks to be addressed
to have Artemis match some level of parity with ActiveMQ with the goal of
eventually having Artemis become ActiveMQ 6.x. I hold no preconceived
notions about this effort or the time it will take, it could take another
year or two years, I have no idea.

Bruce

On Wed, Dec 6, 2017 at 11:19 AM, jgenender <jgenender@apache.org> wrote:

> There is a vote that is more and more looking like an underlying agenda as
> you can start to see a dividing line separated mostly by companies.  Sorry,
> just calling a spade a spade.  Its definitely bringing back the
> knock-down-drag-out threads from a couple of years ago.  That's a shame and
> I really hope that is not the direction this is going.
>
> This is a PR vote that also becomes technical.  It's PR because some folks
> are saying that Artemis AKA HornetQ must become AMQ6 now.  Its technical
> because making it AMQ6 makes assumptions that it will take over from AMQ5.
> I realize that people say "Nobody is saying AMQ 5 (classic - whatever) is
> being deprecated", but guess what?  Making Artemis the new AMQ6 means its
> the next in line and should have a degree of compatibility with the old.
> Remember AMQ3->4->5.  Its an assumption that has been made on numerical
> versions for a majority of software and this one in particular - that's
> technical. It does make illusions of deprecation.  So this is both PR *and*
> technical.
>
> My concerns here are that some people are fine with Artemis ultimately
> becoming AMQ6 given that Artemis gains a solid user base and has reasonable
> compatibility with AMQ5.  NPEs utilizing Openwire seems like some basic
> stuff to fix for minimal compatibility with those who are running AMQ5,
> which fortunately or unfortunately, happens to be the vast majority of our
> community.
>
> The requests by the -1s seem to ask a simple thing that is very reasonable.
> Lets get that adoption rate up and get the compatibility more in line so
> that people DO have a path to upgrade to the next version.  Why is that
> viewed as so unreasonable?
>
> I also want to make a statement that Arthur made earlier about naming and
> vendors and versions.  This is a problem because there is an agenda and
> there is cross marketing going on.  Look at this blog:
>
> https://blog.akquinet.de/2017/02/22/activemq-confusion-and-
> what-comes-with-your-jboss-eap-wildfly/
>
> Even looking at Red Hat's very own GA repo:
>
> https://maven.repository.redhat.com/ga/org/apache/
> activemq/artemis-server/2.0.0.amq-700013-redhat-1/
>
> Is it 2 or is it 7?  Is it JBoss or ActiveMQ?  This repo has a numbering
> with our name on it, but is it even our code?  So when I hear people in the
> community ask "Should we go to ActiveMQ 7", I reply "Huh?  There is no
> ActiveMQ 7".  After a while it turns out they are referring to JBoss AMQ 7.
>
> So sure, we can claim all day that Apache ActiveMQ has nothing to do with
> vendors, but lets be honest.  This has everything to do with vendors and
> this vote and the lines it is drawing proves it.  At the end of the day and
> in theory, you are right... vendors SHOULD have nothing to do with this.
> But the cross pollination of employees and committers unfortunately clouds
> this immensely.
>
> So I ask this.  It seems very reasonable to say that all nay-sayers here
> are
> ok with with Artemis eventually becoming ActiveMQ 6.  They are asking a
> relative simple request: Can we please increase the community adoption and
> get it more compatible before renaming it?  That makes this non-technical
> once that's complete and all parties are fine with Artemis == ActiveMQ 6.
> IIRC, this was exactly what we all agreed upon when bringing in HornetQ and
> why it was named Artemis to begin with.
>
> Why the rush? Fix the reasonable concerns, do what we agreed upon when
> bringing HornetQ into our community, and you can have your cake and eat it
> too.
>
>
>
> --
> Sent from: http://activemq.2283324.n4.nabble.com/ActiveMQ-Dev-
> f2368404.html
>



-- 
perl -e 'print
unpack("u30","D0G)U8V4\@4VYY9&5R\"F)R=6-E+G-N>61E<D\!G;6%I;\"YC;VT*" );'

ActiveMQ in Action: http://bit.ly/2je6cQ
Blog: http://bsnyder.org/ <http://bruceblog.org/>
Twitter: http://twitter.com/brucesnyder

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message