Hey all- We are seeing the same situation that JB and Christopher bring up. Users are confused, and we see a significant number of users picking ActiveMQ 5 over Artemis for new projects.
The idea of Artemis as a TLP seems to make a lot of sense at this point. -Matt Pavlovich > On Mar 19, 2021, at 10:37 AM, Christopher Shannon > <[email protected]> wrote: > > JB, > > Right, I am a user of both products also. I am still a heavy 5.x/Classic > user but I also am still trying to get my project/team onto Artemis as well > but migration is always slow. I'm also working quite heavily with Kafka now > as well which has taken up a lot of my time and prevented me from having as > much interaction with AMQ recently. Having a separate TLP does not stop > anyone from contributing and using both projects but it does make things a > lot easier I think in terms of naming, versioning, etc. I think it also > just adds a lot more clarity for the users. Also, even as its own TLP, > there is no reason why Artemis can't still have a goal to be feature > compliant with ActiveMQ 5.x. > > A pretty good example about how things are confusing right now is Artemis > has been advertised as the next generation but there has been a lot of > discussion recently about 5.x release cycles, upgrading the datastore, JMS > 2.0 upgrades, etc. Basically it looks like it is still being actively > development (which it is) so a user might be pretty confused about which > broker to pick and what is going on long term. > > Anyways, I can only speak for myself so maybe most people still think the > two projects should stay together under 1 umbrella. If I am in the minority > and most people want to keep everything together that is fine. > > On Fri, Mar 19, 2021 at 11:15 AM Jean-Baptiste Onofre <[email protected]> > wrote: > >> Hi Chris, >> >> I’m sharing the same feeling and analyze. >> >> Even if I plan to work more on Artemis (first PR will happen soon ;)), the >> current situation is what you describe: we have two communities. >> >> It’s not a bad thing IMHO, it happens in all projects, with time and >> resulting of some decisions taken. >> >> Regards >> JB >> >>> Le 19 mars 2021 à 16:02, Christopher Shannon < >> [email protected]> a écrit : >>> >>> The real issue here is quite obvious and demonstrated over the last >> several >>> years and that is the goal isn't really agreed upon by everyone. Some >>> people think the goal should be to make Artemis the next generation and >>> others don't agree with that. >>> >>> Personally, while I still think having one broker to rally upon and make >>> the best broker possible (Artemis) should be the goal, the realist in me >>> does not believe this will ever happen without significant push back from >>> the members of the community that do not want this to happen. >>> >>> Over the past couple of years my opinion has swayed now to simply making >>> Artemis its own TLP. Even after 6 years we have 2 separate communities >>> entirely. We basically have 2 independent projects with almost no overlap >>> between developers and users so why not just make Artemis its own TLP? I >>> really see no benefit of operating under the same umbrella anymore (I >> guess >>> maybe if you want to keep the branding) >>> >>> On Fri, Mar 19, 2021 at 10:57 AM Jean-Baptiste Onofre <[email protected] >> <mailto:[email protected]>> >>> wrote: >>> >>>> Hi, >>>> >>>> I understand the point. My main concern was in term of versioning (I >> don’t >>>> want to "block" ActiveMQ to do a 6, 7, 8 or whatever release). >>>> >>>> Sorry, but classic is something I don’t fully understand (maybe my >> French >>>> culture ;)). I don’t see what’s "classic" (or maybe in term of >> "previous" >>>> or "older") is. Maybe "vintage" (like classical music compare to house >>>> music) ;) ? >>>> >>>> And, anyway, nobody is using "classic": for the users I know, they use >>>> ActiveMQ and Artemis (not ActiveMQ Classic or ActiveMQ Artemis). >>>> >>>> So, if you agree to have: >>>> >>>> http://activemq.apache.org/artemis <http://activemq.apache.org/artemis> >> <http://activemq.apache.org/artemis <http://activemq.apache.org/artemis>> >>>> http://activemq.apache.org/activemq < >> http://activemq.apache.org/activemq> <http://activemq.apache.org/activemq >> <http://activemq.apache.org/activemq>> >>>> >>>> I think it’s even better than introducing a new name, I agree. >>>> >>>> Then, I’m changing the proposal/question: agree to use activemq and >>>> Artemis on website ? >>>> >>>> Regards >>>> JB >>>> >>>>> Le 19 mars 2021 à 15:48, Bruce Snyder <[email protected] <mailto: >> [email protected]>> a écrit : >>>>> >>>>> Gary stated this well. I agree completely with all of his sentiments >>>> here. >>>>> >>>>> I don't see the point to introducing yet another name as this will >> muddy >>>>> these waters even further, not clarify them. Artemis was meant to be a >>>> code >>>>> name until it matched ActiveMQ enough to be a drop-in replacement. I >>>> don't >>>>> believe that this goal has been achieved yet, has it? Is this still an >>>>> active goal? >>>>> >>>>> Classic is an appropriate and deliberate name that was being discussed >> as >>>>> far back as just prior to the HornetQ donation. If we start officially >>>>> referring to it as ActiveMQ Classic, then we need to explain the intent >>>>> behind this name via the website. >>>>> >>>>> Agreed, the Classic stream needs a major version bump before those >>>>> incompatible changes are introduced. Do this and move forward with >> those >>>>> incompatible changes. >>>>> >>>>> Bruce >>>>> >>>>> On Fri, Mar 19, 2021 at 8:29 AM Gary Tully <[email protected] >> <mailto:[email protected]>> wrote: >>>>> >>>>>> Hi JB, >>>>>> I think "classic" is a good name precisely because of its meaning, it >>>>>> reflects its value and is a good way to differentiate on the website. >>>>>> >>>>>> But I don't think the classic stream should be limited in versioning. >>>>>> If for good reason (a new incompatible openwire version/storage >>>>>> incompatible change/large config update) it needs a major version >>>>>> increment, then go for it. >>>>>> >>>>>> Artemis was always intended as a code name, a generic title, a >>>>>> temporary moniker, till it could take on the activemq mantle, but it >>>>>> does not have to be 6, it can be 10 or 20, or it can be ActiveMQ >>>>>> Artemis. >>>>>> >>>>>> I don't see any point in introducing another "brand" name, the >>>>>> versioning will be sufficient if we want to consolidate on the >>>>>> activemq name in the future, and the Artemis sub brand will be >>>>>> sufficient if we don't. >>>>>> >>>>>> to speak to Lucas, the plan for Artemis is to be be a better ActiveMQ >>>>>> >>>>>> kind regards, >>>>>> gary. >>>>>> >>>>>> On Fri, 19 Mar 2021 at 04:49, Jean-Baptiste Onofre <[email protected] >> <mailto:[email protected]>> >>>>>> wrote: >>>>>>> >>>>>>> Hi, >>>>>>> >>>>>>> Like any Apache (and OpenSource) project, ActiveMQ "umbrella" project >>>> is >>>>>> living and roadmap evolves. >>>>>>> >>>>>>> Justin is right with the project history, but I think the initial >>>> target >>>>>> to "replace" ActiveMQ with Artemis evolves, due to the users. >>>>>>> >>>>>>> Personally, I’m saying still lot of ActiveMQ users, not planning to >>>>>> change to Artemis, and even brand new installation starts with >> ActiveMQ >>>>>> (not Artemis). >>>>>>> Furthermore, in term of features, there are some gaps between >> ActiveMQ >>>>>> and Artemis IMHO. >>>>>>> >>>>>>> I think the mistake was to create a separated repo for Artemis: if >>>>>> Artemis was ActiveMQ "master" branch at the time of the donation, >> then, >>>> the >>>>>> update would be straight forward. >>>>>>> >>>>>>> So, clearly, IMHO, we have two completed separated projects between >>>>>> ActiveMQ and Artemis, because the communities (both users and >>>> contributors) >>>>>> are not the same. >>>>>>> >>>>>>> A possible path to that Artemis become Apache TLP (and so ActiveMQ >>>>>> Artemis name), and ActiveMQ "classic" stays what he’s: Apache >> ActiveMQ. >>>>>>> >>>>>>> If the PMC don’t want to "move" as a TLP, then we should at least >> give >>>>>> space for the two subprojects in the ActiveMQ umbrella and clearly >>>> identify >>>>>> who is what. >>>>>>> >>>>>>> Regards >>>>>>> JB >>>>>>> >>>>>>>> Le 19 mars 2021 à 05:13, Tetreault, Lucas >> <[email protected] <mailto:[email protected]> >>>>> >>>>>> a écrit : >>>>>>>> >>>>>>>> Hey Justin, >>>>>>>> >>>>>>>> Thanks for the additional context. As a newcomer, it seems to me >> like >>>>>> both ActiveMQ "Classic" and Artemis are alive and well. 6 years and 2 >>>> major >>>>>> versions in to development, is the goal and focus for Artemis still on >>>>>> feature parity and becoming ActiveMQ 6? >>>>>>>> >>>>>>>> Thanks, >>>>>>>> Lucas >>>>>>>> >>>>>>>> >>>>>>>> On 2021-03-18, 8:38 PM, "Justin Bertram" <[email protected] >> <mailto:[email protected]>> >>>> wrote: >>>>>>>> >>>>>>>> CAUTION: This email originated from outside of the organization. Do >>>>>> not click links or open attachments unless you can confirm the sender >>>> and >>>>>> know the content is safe. >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> Lucas, >>>>>>>> >>>>>>>> I'm not sure you've been around the ActiveMQ community for very >>>>>> long or >>>>>>>> maybe you have been and forgot some of the history. In any case, >>>>>> I'll >>>>>>>> summarize briefly. >>>>>>>> >>>>>>>> Over a decade ago Hiram Chirino, one of the original ActiveMQ >>>>>> developers >>>>>>>> and chair of the ActiveMQ PMC at the time, created a new broker >>>>>> under the >>>>>>>> ActiveMQ banner named Apollo. It was designed on a non-blocking >>>>>>>> architecture for much better performance than the existing ActiveMQ >>>>>>>> architecture [1]. After ActiveMQ Apollo 1.0 was released the stated >>>>>> goal of >>>>>>>> this project was for it to eventually be integrated with the >>>>>> mainline >>>>>>>> ActiveMQ code-base and serve as its replacement [2]. This fact was >>>>>>>> advertised on the ActiveMQ website although there are no longer any >>>>>>>> references to that since the website was redesigned & updated a >>>>>> year or so >>>>>>>> ago. For whatever reason Apollo never acquired the critical mass >>>>>> necessary >>>>>>>> to replace mainline ActiveMQ. >>>>>>>> >>>>>>>> Then about 6 years ago the HornetQ code-base was donated to the >>>>>> ActiveMQ >>>>>>>> community and that donation was accepted with the goal of creating >>>>>> the next >>>>>>>> generation ActiveMQ broker that would eventually become version 6. >>>>>> Since >>>>>>>> that time work has steadily progressed on the Artemis code-base to >>>>>> bring >>>>>>>> sufficient feature parity with mainline ActiveMQ to allow users to >>>>>>>> transition. Again, this has been communicated via the website and >>>>>> other >>>>>>>> support channels for the last several years. >>>>>>>> >>>>>>>> For what it's worth, I hope that clarifies the current state of >>>>>> affairs. >>>>>>>> >>>>>>>> >>>>>>>> Justin >>>>>>>> >>>>>>>> [1] >>>>>>>> >>>>>> >>>> >> https://hiramchirino.com/blog/2011/01/17/activemq-apollo-looking-impressive/ >> < >> https://hiramchirino.com/blog/2011/01/17/activemq-apollo-looking-impressive/ >>> >>>>>>>> [2] >>>>>> https://hiramchirino.com/blog/2012/02/03/apache-apollo-1-0-released/ >> <https://hiramchirino.com/blog/2012/02/03/apache-apollo-1-0-released/> >>>>>>>> >>>>>>>> On Thu, Mar 18, 2021 at 5:02 PM Tetreault, Lucas >>>>>>>> <[email protected] <mailto:[email protected]>> >> wrote: >>>>>>>> >>>>>>>>> Hi all, >>>>>>>>> >>>>>>>>> It seems to me like the core problem here is that there are two >>>>>> distinct >>>>>>>>> projects operating under one brand and it creates confusion. I >> agree >>>>>> with >>>>>>>>> JB that the "ActiveMQ Classic" and "ActiveMQ 5" branding are not >>>>>> ideal. >>>>>>>>> However, I think that renaming it to ActiveMQ Leto will further >>>>>> dilute the >>>>>>>>> ActiveMQ brand and create more confusion for users. Why not just >>>>>> "ActiveMQ" >>>>>>>>> and "Artemis"? >>>>>>>>> >>>>>>>>> Thanks, >>>>>>>>> Lucas >>>>>>>>> >>>>>>>>> On 2021-03-18, 6:54 AM, "Jean-Baptiste Onofre" <[email protected] >> <mailto:[email protected]>> >>>>>> wrote: >>>>>>>>> >>>>>>>>> CAUTION: This email originated from outside of the organization. >> Do >>>>>>>>> not click links or open attachments unless you can confirm the >> sender >>>>>> and >>>>>>>>> know the content is safe. >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> Hi Justin, >>>>>>>>> >>>>>>>>> I would rather to ActiveMQ Leto 5.17.0 (and then Leto 6.0 at some >>>>>>>>> point). >>>>>>>>> >>>>>>>>> Regards >>>>>>>>> JB >>>>>>>>> >>>>>>>>>> Le 18 mars 2021 à 14:51, Justin Bertram <[email protected] >> <mailto:[email protected]>> a >>>>>>>>> écrit : >>>>>>>>>> >>>>>>>>>> I'm not clear on the versioning you're proposing. Are you saying >>>>>>>>> that the >>>>>>>>>> first release of this subproject would be ActiveMQ Leto 1.0? >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> Justin >>>>>>>>>> >>>>>>>>>> On Thu, Mar 18, 2021 at 7:46 AM Jean-Baptiste Onofre < >>>>>>>>> [email protected] <mailto:[email protected]>> >>>>>>>>>> wrote: >>>>>>>>>> >>>>>>>>>>> Hi Robbie, >>>>>>>>>>> >>>>>>>>>>> My point is about "classic". I understand the meaning but I think >>>>>>>>> it’s not >>>>>>>>>>> a good "tagging". >>>>>>>>>>> >>>>>>>>>>> I don’t want to focus on ActiveMQ 5.x, because it prevents us to >>>> use >>>>>>>>>>> another versioning. >>>>>>>>>>> Why not ActiveMQ 6.0 that would be a new major ActiveMQ release. >>>>>>>>>>> >>>>>>>>>>> To summarize: >>>>>>>>>>> 1. ActiveMQ 5.x is too restrictive for versioning >>>>>>>>>>> 2. Classic is not a good "naming/tagging". >>>>>>>>>>> >>>>>>>>>>> That’s why I’m proposing a new identified name. It means we would >>>>>>>>> have: >>>>>>>>>>> >>>>>>>>>>> - Apache ActiveMQ Artemis >>>>>>>>>>> - Apache ActiveMQ Leto >>>>>>>>>>> >>>>>>>>>>> IMHO, it’s two subprojects under the same "umbrella" (like we >> have >>>>>>>>> Camel >>>>>>>>>>> K, Camel Spring Boot, Camel Karaf, or Karaf runtime, Karaf >>>>>>>>> Decanter, Karaf >>>>>>>>>>> Cave, etc). >>>>>>>>>>> Each subproject deserves a clear naming. >>>>>>>>>>> >>>>>>>>>>> About the website, you got my point: I would like to get all wiki >>>>>>>>> based >>>>>>>>>>> resources, update and clean it to push on a dedicated sub context >>>>>>>>> of the >>>>>>>>>>> website: >>>>>>>>>>> >>>>>>>>>>> http://activemq.apache.org/artemis < >>>>>>>>> http://activemq.apache.org/artemis> >>>>>>>>>>> http://activemq.apache.org/leto <http://activemq.apache.org/leto >>> >>>>>>>>>>> >>>>>>>>>>> Each with its own announcement, download, documentation >> resources. >>>>>>>>>>> >>>>>>>>>>> Regards >>>>>>>>>>> JB >>>>>>>>>>> >>>>>>>>>>>> Le 18 mars 2021 à 12:19, Robbie Gemmell < >> [email protected] >>>>> >>>>>>>>> a >>>>>>>>>>> écrit : >>>>>>>>>>>> >>>>>>>>>>>> The 'classic' terminology on the homepage is used more as a >>>>>>>>>>>> description rather than a name to me, speaking to its quality >> and >>>>>>>>>>>> vintage and in some small straightforward way. ActiveMQ 5 is >> still >>>>>>>>> the >>>>>>>>>>>> way the broker is referenced on the site as a whole so far as I >>>>>>>>> see, >>>>>>>>>>>> rather than ActiveMQ Classic. Essentially everywhere besides the >>>>>>>>>>>> subdir name being 'classic' in the URL for grouping some of the >>>>>>>>> newest >>>>>>>>>>>> component pages. I dont think 'Leto' is particularly more useful >>>>>>>>> than >>>>>>>>>>>> 'classic' as a description, and especially not an improvement >> for >>>>>>>>> the >>>>>>>>>>>> subdir in the URL at this point. It would be quite the opposite >>>>>>>>> for me >>>>>>>>>>>> personally, I think it would be a bad idea. >>>>>>>>>>>> >>>>>>>>>>>> Changing the subdir on the site from 'classic' to something else >>>>>>>>>>>> simplistic and direct such as 5 or 5x or 5.x? Sure, I can see >>>> that. >>>>>>>>>>>> Dropping the "Classic" description suffix from the central box >> on >>>>>>>>> the >>>>>>>>>>>> homepage, leaving only the ActiveMQ 5 titling? By all means. >> Leto? >>>>>>>>> I >>>>>>>>>>>> dont really see that being an improvement at this point at all. >>>>>>>>>>>> >>>>>>>>>>>> On your other proposal of cleaning up mess, presumably that >> means >>>>>>>>> the >>>>>>>>>>>> mass of old 5.x wiki-derived pages on the site in the root (done >>>> to >>>>>>>>>>>> preserve URLs during the site changeover I believe, over >>>>>>>>>>>> individual-page redirects) that are rarely ever touched, and >>>> moving >>>>>>>>>>>> such content into the subdir? Sounds great. >>>>>>>>>>>> >>>>>>>>>>>> Robbie >>>>>>>>>>>> >>>>>>>>>>>> On Thu, 18 Mar 2021 at 08:30, Jean-Baptiste Onofre < >>>>>>>>> [email protected]> >>>>>>>>>>> wrote: >>>>>>>>>>>>> >>>>>>>>>>>>> Hi guys, >>>>>>>>>>>>> >>>>>>>>>>>>> I would like to bring on the table the naming of Apache >> ActiveMQ. >>>>>>>>>>>>> >>>>>>>>>>>>> I think "Classic" is not a good name, and it doesn’t mean >>>>>>>>> anything. I >>>>>>>>>>> think it would make more sense to have a generic name. >>>>>>>>>>>>> >>>>>>>>>>>>> As we have Apache ActiveMQ Artemis, I would like to propose >>>> Apache >>>>>>>>>>> ActiveMQ Leto. >>>>>>>>>>>>> >>>>>>>>>>>>> From a cultural standpoint ;), Artemis is the Greek goddess of >>>> the >>>>>>>>>>> hunt, the wilderness, wild animals, the Moon, and chastity. >> Artemis >>>>>>>>> is the >>>>>>>>>>> daughter of Zeus and Leto, and the twin sister of Apollo. >>>>>>>>>>>>> As "ActiveMQ Classic" is "older" than Artemis, I propose to >>>>>>>>> rename as >>>>>>>>>>> Apache ActiveMQ Leto. >>>>>>>>>>>>> >>>>>>>>>>>>> This name change won’t impact the code repository, it’s more >> for >>>>>>>>> the >>>>>>>>>>> website. >>>>>>>>>>>>> >>>>>>>>>>>>> Related to that proposal, I would like to propose also to >> create >>>> a >>>>>>>>>>> dedicated space for Leto: http://activemq.apache.org/leto < >>>>>>>>>>> http://activemq.apache.org/leto> with a complete cleanup of the >>>>>>>>> mess we >>>>>>>>>>> have today (documentation, download page, announcements, etc). >>>>>>>>>>>>> >>>>>>>>>>>>> Thoughts ? >>>>>>>>>>>>> >>>>>>>>>>>>> Regards >>>>>>>>>>>>> JB >>>>>>>>>>> >>>>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>> >>>>>>> >>>>>> >>>>> >>>>> >>>>> -- >>>>> perl -e 'print >>>>> unpack("u30","D0G)U8V4\@4VYY9&5R\"F)R=6-E+G-N>61E<D\!G;6%I;\"YC;VT*" >> );' >>>>> http://bsnyder.org/ <http://bruceblog.org/> >> >>
