1 2 Previous Next 23 Replies Latest reply on Dec 30, 2004 3:51 AM by Heiko Rupp

    Milestone 2

    Cody Lerum Newbie

      Just looking for an update. I know you guys are swamped, but what is the current rough guess for M2?

      1 Month?
      3 Months?
      6 Months?

      Thanks!

        • 1. 3846768
          Scott Stark Master

          Here's my negative opinions (Andrew, apologies in advance, but it's a genuine set of opinions):

          - This project has no planned architecture (let alone design). It desperately needs one.
          - This project does not have genuine selling point other than that it is "the same old mail services only this time on JBoss" (yes, I have read "the case for mail services")
          - This project is too unfocused in deployment of effort and feature delivery (for instance, the integration with Nukes bit - that's way too early to be a target)
          - If I see another custom, application-specific user/group management implementation I am seriously going to scream!
          - It's dangerous (even suicidal) to tie an enterprise application to tiger just yet. For most big enterprises 1.3.1 is still a "standard" JVM.
          - No lessons seem to have been learned from the experience of Apache James (the big draw for me in James was the ability to write Mailets, but it failed evaluation for much the same reasons as mentioned above and that have been previously posted in this thread).

          So what to do?

          Scott says "step up to the wire". The question is - what's the reward for what would be a major time investment. That should be made clear first.

          In the meantime, my initial feedback on the "overall vision" side would be:

          - Decide first and foremost WHAT DOES THIS PROJECT GIVE AN ENTERPRISE OVER AND ABOVE SENDMAIL etc or EXCHANGE? Find the single line answer to that question that is sufficient to convince anyone whether commercial or technical.
          - Focus first on user/group management and figure out how to reduce dependencies to the point where it is NOT destined to be an inextricable module of the mail services.
          - Deliver the IMAP so that a webmail interface could be built.
          - Steal the mailets idea from james - yes, even the source as a start point.
          - Look forward a bit and realise that you must ditch ejb2 entity beans and build on ORM (hibernate) - later provide an ejb3 implementation for java 5 when people actually want it
          - Drop any other integration effort for the time being

          Well, that's my 20 cents.

          /k1

          • 2. Re: Milestone 2
            iwadasn Newbie

            Try doing some work on it then. A mail service support IMAP is not a priority task.

            • 3. 3850206
              Kabir Khan Master


              Here's my official whine....

              Please finish Milestone 2 promptly.

              At work I'm trying to use jboss mail services with itracker to make an email-enabled issue tracker, it would help greatly if your mail services worked well. If they did, then we could bundle jboss mail services along with itracker to assist others who have similar problems.

              • 4. Re: How do I get started
                Steve Davis Newbie

                I came across a neat slide presentation on the sourceforge jboss-user mailing lis, which btw is also a good place to hunt got info.

                I am trying to get the slide presentation included as a pdf doc into the Docs section of the sourceforge page. The original is in a ppt format. It also has an example project with source code.

                Here are the links,

                soruceforge: http://sourceforge.net/projects/jboss/

                jboss-user list archive: http://sourceforge.net/mailarchive/forum.php?forum_id=2266

                presentation: http://www.thirdeyeconsulting.com/indyjug/jboss/

                I also attach the pdf version of the presentation (at I will try...). The heasdings are a bit screwed, but it is readable. This presentation was compiled by Chris Bonham.

                PS: realised the attachment should not go over 1M, so I am posting it in 2 halves.

                • 5. Re: Milestone 2
                  Andrew Oliver Master

                  Right now resources are tight because we don't get the kind of customer feedback we need. Shout for it. Demand it. Say "We want it now!" (this is a good start!)

                  • 6. Re: Milestone 2
                    Cody Lerum Newbie

                    OK-

                    See I was being a coward and taking a wait and see stance. Since I can't help with the coding I figured I we wouldn't try and rush you guys.

                    Everyone who is monitoring this project, and are eagerly awaiting a M2,M3, etc release.

                    Lets generate some feedback, and get this thing moving forward.

                    If you guys need specific feedback on the project let us know, and would be glad to help

                    Thanks!

                    -C

                    • 7. Re: Milestone 2
                      Andrew Oliver Master

                      More than anything we need "I'm a JBoss customer or potential customer with use case XYZ we need JBossMail for TUV". The scare us with threats to use big pieces of crap like Domino (okay I'm in hyperbole) haha. Basically for the project to become a priority, like with any company, we have to demonstrate more demand (or better yet revenue). Unfortunately to demonstrate more demand we need it to be mature....which requires the project to be more of a priority....etc. Secondly, we need more developers of course. Tell us what features from M2 you most want, etc.

                      We need pain in the ass users :-) (Though not ones who ask "what is email?" ;-) )

                      • 8. Re: Milestone 2
                        Cody Lerum Newbie

                        Well currently my company (ISP) is using a CommuniGate Pro as it's email server. About 3500 accts. It works fine, but I'm intrigued as to how this system could scale cheaply..clustering and all.

                        I have installed the M1 release, and I'm close to making the next release my personal server. Friends and Family accts.

                        Priorities for me would be.

                        Multiple domain support
                        Clamav, spamassassin support.
                        Web configuration/administration
                        Webmail
                        Exchange Server replacement.

                        I don't know where these fit into your plans, but I think a basic functioning smtp/pop3 server that is easy to configure would get the install base growing rapidly. Then come back to the more advanced features.

                        • 9. Re: Milestone 2
                          Joachim Van der Auwera Expert

                          How about IMAP support. Very hard to find an OSS IMAP mail server.

                          Joachim

                          • 10. Re: Milestone 2
                            Oleg Gusakov Newbie

                            Another raving customer/user

                            We have been using JBoss for some 3 years, this helped to gradually adopt a lot of OpenSource products [and buy some support as well :)]

                            Please move on with JBoss mail. Our company is on the search for a replacement for Notes mail. We tried Apache James, JBoss Mail M1 and the later seems to be viable alternative, except for the fact that:
                            - it does not have IMAP
                            - it's been in M1 mode for quite some time - this one kills all my efforts to present this to a wider public.

                            Where else can we whine to be heard?

                            • 11. Re: Milestone 2
                              Paul York Newbie

                               

                              "scott.stark@jboss.org" wrote:
                              Try doing some work on it then. A mail service support IMAP is not a priority task.

                              Scott, first off Andrew asked for the "whining" requests from "pain in the ass users." Your seemingly snippy reply to a legitimate request from a real or potential customer seems counter-productive to your goals of attracting both customers and (perhaps more importantly at this stage of the project) developers.

                              It seems to me that there is a significant chicken-and-egg problem here. I have to imagine that there is real demand out there for an open-source alternative to Exchange. A real alternative, including IMAP support (how can this not be a priority??), but also calendaring, Outlook (i.e., MAPI) support, mobile device support, etc. If a true F/OSS alternative to exchange existed, especially one that had a strong support infrastructure and a viable company behind it (like JBoss, of course), then I can't imagine that a significant corporate user base would not develop--one that would be more than willing to pay for support and training. The M1 release, while a great start, cannot possibly go into any semblance of production in a real organization. Thus, there won't be any real customers to create the feedback and demands that seems to be required to stimulate support for continued development of the project from within JBoss. Thus, the project may well be dead, which would be a true shame.

                              I had the pleasure of speaking pretty extensively with Marc yesterday. One of the things I walked away with from this conversation was that for the Professional Open Source model to work, especially if you are targeting enterprise customers with infrastructure services, you have to have relatively mature products before you can expect any real return on investment (to be "zero cash-flow" or "cash-flow positive" or whatever term he used). Mail Services is simply not there yet. This is a classic R&D investment conundrum. Are you willing to invest the time and capital necessary to get the product to a point that it can get past the "try it out" phase to the "roll it out" phase?

                              It seems the big problem here is lack of developer support, especially from the outside community. James seems to have much the same problem. Perhaps the project isn't sexy enough or developers don't see personal (or intrinsic) value in contributing to this particular project. It would seem necessary to do one or both of the following: a) invest in the necessary "professional developer" time to get this thing ready for prime-time, or b) put forth a significant marketing effort to attract/incent outside (amateur) developers to participate. Lacking either of these, it seems unlikely that this project can succeed. Andrew indicates that (a) is not happening at present. Assuming this is the case and things won't change internally at JBoss (I would argue perhaps that they should, but won't here), it seems like some effort towards (b) would be a relatively easy and cost-effective solution.

                              I might suggest a couple of things. First, you might consider posting a request for developers on the home page--a "hot project" box or something to entice potential contributors towards this project. Figure out some way to make this project seem sexy and worth a hackers investment. Second, I would suggest taking the time to put together some form of architectural overview at a low enough level that it is fairly simple for a good developer to get started quickly (reducing barriers to entry in economic terms). For my part, I downloaded the source and have been trying to figure out how things are glued together. I've had little luck, thus far. I have no doubt that given enough time, I could figure out how it all works, but that would require a major investment of my time. For a relatively minor investment of Andrew's time, I (and other potential contributors) could get quickly up to speed and contribute meaningfully to the effort.

                              Hope this helps, or at least sparks some more debate on the subject.

                              Paul York, Ph.D. Student
                              Management Information Systems
                              University of Georgia


                              • 12. Re: Milestone 2
                                Paul York Newbie

                                Just one more quick note on IMAP support. Implementing IMAP is probably the most logical way of getting a usable WebMail interface up and running quickly. Most WebMail interfaces I've played with (most natably SquirrelMail and Horde, but I'm sure there are similar Java/JSP based alternatives) simply use IMAP to connect directly to the mail store. Easy and clean. I would assume that eventually you'd want a more full featured and integrated WebMail interface to support more of the advanced features of Mail Services, but simply recommending/modifying one of these IMAP-based projects certainly seems the quickest and easiest path to a good WebMail interface.

                                Paul York, Ph.D. Student
                                Management Information Systems
                                University of Georgia

                                • 13. Re: Milestone 2
                                  Scott Stark Master

                                  And your discussion with Marc should have touched on the JBoss Labs effort which is a proving ground for open source projects. The Mail effort needs to be moved into the JBoss Labs stage 1 effort where it will need to prove that a developer community can be built up to the point where JBoss will provide seed funding to take it to the next level. Professional Open Source is not throwing darts at potential projects and sinking money into them in the hopes that they succeed. Its taking projects that have a proven viability ala the voluteer approach to the next level.

                                  Someone needs to step up to the plate and get this project going.

                                  • 14. Re: Milestone 2
                                    Paul York Newbie

                                    Scott, yes our conversation did include the proposed "incubator" model, which I think is a probably (and reservedly) a good idea. I also agree that Mail Services should perhaps be moved there. But I guess there is a bit of a strategic question here. Does JBoss identify a strategic set of software that it wants to provide to meet customer demand and work to generate developer traction (demand-side economics), or does it look for projects that already have developer traction and try and generate customer demand (supply-side economics). Marc didn't discuss this directly, but I walked away with a bit of a mixed strategic message, here (perhaps you respond to customer demand for existing projects but rely on developers to dictate future projects????). I can see merits on both sides, though I personally would argue that the demand-side option might be more of a positive differentiator for JBoss. Every other open source organization seems to be supply driven. JBoss seems uniquely positioned to provide a demand driven approach, and further seems to have the most to gain from such an approach. Certainly not "throwing darts," but trying to identify the most salient complementary products to your core and investing strategically to get traction.

                                    Having a generic incubator would seem to be a bit counter to this. The "sexier" projects would get the most developer support at the expense of the potentially more lucrative and strategically beneficial projects. Again, though, it shouldn't take much to steer developers towards the more strategic projects, even if you don't directly support it monetarily. Front page marketing, contests, spotlighting, etc. are all cheap and easy ways of incenting developers towards working more concertedly for the betterment of JBoss, rather than JBoss being entirely at the whim of the developers.

                                    Just some thoughts...

                                    Paul York, Ph.D. Student
                                    Management Information Systems
                                    University of Georgia

                                    1 2 Previous Next