[Zope-dev] Re: Zope 3.4.0 candidate 1 Released

[Tres Seaver]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Chris McDonough wrote:
> Stephan Richter wrote:
>> On Friday 01 February 2008, Chris McDonough wrote:
>>> If there will continue to be a release schedule for "Zope 3, the appserver"
>>> It would reduce confusion to new users greatly to give the appserver
>>> release a name other than Zope.
>> Well, we had to do the classic Zope 3 release at least one more time. Because 
>> the official story is still: Download the Zope 3.3 tar ball and start using 
>> it. We have to use at least one release to tell people that we are going to 
>> change the process and allow them still both methods.
> 
> Of course.
> 
>> I also think that we have no solid story and/or documentation to promote the 
>> new approach. My hope is that the story and documentation will develop during 
>> the next release cycles.
>>
>> All I am doing is doing something about a pretty pathetic situation. I took 
>> the least oath of resistance.
> 
> Heh.  You're doing yeoman's work.
> 
>> And I am particularly tired of name change suggestions! For many reasons.
> 
> I figured it wouldn't be a popular suggestion.  But I do believe it is the right 
> thing.  It would have been the right thing from the start, but there is still 
> time to repair things.
> 
> I typed four more paragraphs full of markety stuff here but deleted them.  It's 
> not useful.  If no one else thinks it's a good idea, I'm not going to push either.

I would favor the following for a roadmap going forward:

 - No more tarball releases, period.  Nobody should expect to get
   another one, or even anything other than a "critical security fix"
   3.4.1 tarball.  The path for maintenance going forward is going
   to be to release individual eggs with bugfixes, new features, etc.

 - Somebody *might* release a meta-egg which would serve the same
   purpose as the current Zope3 tarball release:  it would pull in all
   the other eggs from the KGS needed to get a "ZMI" up and running.
   That egg should *not* be called "Zope3":  it might be called
   "z3c.zmi", or some such.  There might even be multiple such packages
   (e.g., one which configures one or more of the example application).

 - We should fix up our "smoke test" story so that we can do large-scale
   integration tests of something resembling the current tarball
   release:  this is probably just a buildout, which pulls in all the
   eggs in the KGS, runs all their unit and functional tests in the
   integrated environment, and perhaps runs some additional functional
   / system tests.  Note that I am not proposing to release this beast:
   it exists primarily to enable testing.

 - Outside applications such as SchoolTool, which currently depend
   on a released 'zope3", should begin to move their dependencies to
   the "meta-egg"-based scheme outlined above:  in fact, they are
   probably good candidates for defining such a meta-egg.

 - Deployments which need non-egg-based packaging will need to figure
   out how to use the dependency information in the target meta-egg to
   stitch together their own packages.


Tres.
- --
===================================================================
Tres Seaver          +1 540-429-0999          tseaver at palladion.com
Palladion Software   "Excellence by Design"    http://palladion.com
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFHo0xg+gerLs4ltQ4RAmsvAJ9PLQMuz+vQLQRlP07PicWaBlUggwCdFoeB
pxcgKOG45yl9DFeokdpPk7c=
=Lfhq
-----END PGP SIGNATURE-----