> From: Rodney Schneider <[EMAIL PROTECTED]> >> >> Hmmm. Problem is that I think the only people that have successfully done >> this have used been using fulcrum. Perhaps the answer is that t2.2 doesn't >> support this, but if you need to do it then use the unreleased fulcrum. > > While I am happy that the goal is to release Turbine 2.2 ASAP, I really think > this issue needs to be resolved in such a way that people aren't forced to > use an unreleased version of the software. That is exactly why there are so > many Turbine users/developers out there using their own custom version of > Turbine 2.1. I guess I am worried that the same thing will be repeated with > Turbine 2.2...
I have been giving this quite some thought lately and yesterday I recalled a comment made by Mark Fleury at a JBoss presentation here in Sydney a few months ago. Mark was commenting on some overzealous pushing for a JBoss 3.0 final release. His comment was something like: "With open source you can make your own release - just take a cut from cvs head then use it and fix it as necessary." So why am I pushing for a release? - In increasing order of importance to me: 1. It has been a long time since 2.1 final was released. 2. We are in a cycle of betas that does not seem to have any end in sight. 3. There have been significant functional improvements since turbine 2.1 and I would like to gain access to these for my production applications. 4. My preference is to build my own cut of turbine from a true release rather than one I fabricate myself as this will keep me closer to the action on turbine-dev and enable me to be more relevant on turbine-user. Why will I build my own release after 2.2 final is released? - So that I can selectively apply patches that are relevant to my production system. I can't run a production system from a cvs head build as this may result in an unstable application. Also, I may need to introduce some patches that are only relevant to my application. So why bring up Mark's comment? - Because it now seems clear that the extend-user requirement is for the short term going to require that we use the unreleased fulcrum. I think it has been made reasonably clear that turbine 2.2 won't use fulcrum. Even though my application requires me to extend turbine-user, my opinion on this is: well okay, lets work on getting 2.2 out and I will manage a stable fulcrum build to meet my requirements. After 2.2 is out we can decide on what are the most critical missing features and target these for 2.3 or whatever. > >>> everyone currently using turbine 2.1 should start moving to 2.2 NOW! and >>> report all problems, bugs, ... and send patches!! ;-) >> >> Agreed, but those that need to extend turbine-user will have to work with >> fulcrum (where we may incorporate Henning's DBSecurityService proposal). > > I guess there could be a Turbine 2.21 release (or something like that) that > allows people to extend turbine-user in a standard way? How about: "The standard way to extend turbine_user in turbine 2.2 is to use the unreleased fulcrum DBSecurityService (perhaps replaced by Henning's proposal) in place of the turbine 2.2 integrated DBSecurityService (to do this you...)." Along with a recommendation that for a production environment you should keep a copy of the source for the build you are using so that you can fully support it until the longer term solution is resolved (which is planned for immediately after the 2.2 release). I am trying to be pragmatic about this - my goal is definitely to have a release that includes the ability to extend turbine-user, but if an interim step is required that does not fully support this then it is still a step in the right direction. From your comments I think we are basically in agreement, but I thought what Mark said adds some good perspective on how to deal with this type of situation. Cheers, Scott -- Scott Eade Backstage Technologies Pty. Ltd. http://www.backstagetech.com.au -- To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]> For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>
