Christophe Combelles wrote:
Stephan Richter a écrit :
Not responding to anyone in particular:
I see there are plenty people with opinions. I would love not to do the Zope 3 releases anymore! I am tired of the endless discussions.
Think I am frustrated? Absolutely!
All the suggestions made here require more work, more manpower. But there is nobody doing the work. In fact, I am not even officially the release manager anymore. Remember, other people took over that job, because they wanted to do a release every 6 months? I said back then: Forget it. Nobody believed me and now it has been almost a year since the 3.3.1 release. The only reason I am doing the releases is to tell the world that we are still out there, improving the framework.
I believe the reason of your frustration comes from the fact there are probably not so many people who fully understand the whole release process, the kgs, the buildout, how things are scattered into all these eggs, and all the technology surrounding the transition that zope is going through. I'm not here for a long time and I had to spend hours reading docs and searching every single bit of information in the svn.
I think you're right on the money. I really have very little idea of how Zope 3 is supposed to be used right now, or what "Zope 3" really is (and no-one fully agrees, as evidenced by other posts in this thread). Having to piece together that information from the mailing list is pretty dire. The website work that the Foundation has commissioned/supported is part of this. It's about mapping out the parts of the Zope universe - Zope 2, CMF, Zope 3 libraries, Grok... People like Philipp and Martijn are drafting the texts and pictures that attempt to clarify all this. When those are public and coherent, we'll see how people react and revise accordingly. That'll help, and will probably be a catalyst for the community to actually get this story straight. Right now it's not. Me? I'm just trying to prod people and manage expectations. Personally, I am fully depenent on the people doing Zope 2 release management and Grok work to make sense of these libraries and package it up in some hopefully-useful way for me to use. I don't think "Zope 3" is a starting point the way Zope 2 is/was - I think it was trying to be that for a while, but that people changed their minds. Now, Grok is the starting point. Maybe Zope 2 w/ Plone form another starting point for a certain audience. Once people know about Zope, they may want to use some of its packages (I used zope.component with Pylons, for example), but that's only if they really want to and already know and love Zope. And yeah - the release management story is a big part of this. As Stephan pointed out, it's completely irresponsible to stop doing (regular!) releases the established way until the new way is clearly defined and documented and a until migration path exists. Martin -- Author of `Professional Plone Development`, a book for developers who want to work with Plone. See http://martinaspeli.net/plone-book