Re: New Package Proposal: Enterprise A&A

> Sorry, I don't dispute they aim to solve the same problem (A&A). I don't > even dispute the fact there are structural difference...after all both > were developed without the knowledge of the other. I've only harped on > the two differentiators that if you compare my original post in this > thread to what is shown in the PEAR descriptor for LiveUSer which are a) > web-service based and b) designed from scratch to eventually support the > Liberty project. After hearing more about LiveUser, it seems that b) is > possible though it was never articulated as a goal. Also a) is possible > with LiveUser. Is that accurate? I would say so. > If these aren't merged I don't see a real need for one having a container > in one to support the other. If someone else creates one to suit that > purpose, great, but this should really be a 'this or that' or 'coke vs > pepsi' decision. If they are merged, great. I agree, a merging would be for the best. > Also, some further discussion on permission management needs to be done > since both handle them differently. Which reminds me, I had asked for a > use case or two around permission management that may 'break' either or > both approaches...still waiting on that. > > Again, I'd be open to the discussion. I have to have a system that does > this. What I have built was the result of having built a requirements > document, developing uses cases and establishing a design that would work > (I'd be happy to share these docs). Furthermore this set of documentation > was agreed upon by other project members (remember, this all started > because we had a need for this). If after analyzing Live User to see how > it meets those requirements and granted that any areas of concern can be > addressed, I have no problem merging with Live User. I'm assuming you'd > feel the same in that you'd feel comfortable merging into A&A if it meets > your needs. For my part, i do. > We should first compare the requirements both systems and analyze them > against both system before we assume that the migration will be > A&A->LiveUser as opposed to LiveUser->A&A. I just want to give this the > due diligence it deserves. I can honestly say I'd be willing to move my > code into LiveUser granted that migration path makes the most sense. Yes. We need to spend more time on this. My previous email was written a bit fast. As to me, i can always use a working livuser in production, leave it at that and switch to a promising system anyway. > I need to do the same for LiveUser. However, I think comparing > requirements (if you have them) would be a better starting point. Markus is the man ! Arnaud.