Am 02.03.2006 um 14:13 schrieb Rocky Burt:
On Tue, 2006-28-02 at 13:21 +0100, Martijn Faassen wrote:
Philipp von Weitershausen wrote:
I would vote for spelling out Zed (which would also be a little
to google but might create trademark problems). The namespace
could either be 'z' or 'zed'.
Then again, I really should take Jim's side and stay out of naming
Let's please not have a naming discussion again. I think renaming
is really bad marketing myself and naming discussions mostly a
As I sit here spending sooo much time reading this thread, I've
decided its time to throw in my own naive point of view as an ex-J2EE
developer and a Zope 2 developer that generally builds applications on
top of Zope2/CMF/Plone.
Thanks Rocky encouraged me to not stay out of this discussion any
My perspective is from someone who has decided early on (more than
two years ago) to start using zope3 technics and approaches in Zope2
land. I now have a bigger project under my belt, which is probably
one of the biggest zope2/five installations, a portal with 300
subsites in one instance. But that's actually not the main point.
What I face is the interaction with former contractors of this
customer, who we should explain the system, the technology and also
sometimes the history of this zope island. And this is quite a
difficult expirience, at least for me.
We started this with the goal to transform this to a pure zope3
application sometime, when zope3 the technology and zope3 the main
developer community is ready. So we had a goal, if we will succeed is
dependant on much more than pure technology. And this is my view,
which are only a list of opinions. I hope I can come to a conclusion
in the end.
- Five is a migration path to "something". For us it was always the
hope that this something is something different than the current
Zope2. So we accepted that there are quirks, in the believe, that
this can be removed if a "cleaner" ground can be used.
- Adopting Five and the z3 stuff as an additional resource for Zope2
and keep going with z2 is dangerous. My head starts swirling if I
imagine about the enhancement of AT with z3 technics, it becomes more
and more complex, more and more abstractions need to be understood,
dependancy-lists become longer and longer. The sheer amount of source
code is huge.
- The naming discussion leads to the direction to have "Zed"
something like an additional library which can be used with Zope2.
This hints the usage as an additional layer. We will see products,
which are not pure z3 accessable via Five, but which are using both
and create a new kind of product style, without the intention to move
this to a pure z3 solution.
- If I understand Jim right, he asks, if the z3 developer should
abandon the development of the z3 application server and concentrate
more on the restructuring of the z2 application server. Given the
potential support which can be leveraged if the big z2-communities
(CPS and Plone seem to be the only ones) can be involved, that sounds
reasonable. I fear that it would be nevertheless quite hard, and that
there is not really a big success at least not such a big success, as
the current clean reimplementation of zope3.
- My obersvation from the various discussions here is, that the
people who are using z3 are using it quite differently, much more
creative. This is part of the success to leave the zope-island. And
this leads in some way to uncertainties for the "traditional" z2
developer. On the one side there is the freedom to replace the main
template engine, but this makes it more difficult for a newcomer to a
- Similar to this theme ,"leaving the island", is the explosion of
components we will see. The Zope community has already identified
this and the great packing efforts are underway. Nevertheless it will
be more difficult to understand all these different components, they
are not as approachable as the current products. There will be
bigger components or more application like packages, but the freedom
for the developer also means that the components will be more varied,
using different concepts, will not fit as tightly together as the
current line of CMF/CPS/Plone products. I have the feeling that this
is sometimes missed in the ongoing discussions. We try to find the
one and only solution, where on the other side we opened up the
framework to use quite different solutions.
Ok, what's my conclusion. I think a repackaging is good and perhaps
also the building of bigger mega-packages. But the z2 community needs
to have a goal to evolve to and this needs to be better, and needs to
have features which encourage them to adpopt the new plattform. And I
hope that this plattform is a better z3 application server, much in
the same spirit as the current z3 framework is a better z2 framework.
And z2 developers should face, that they need to rethink and rewrite
some of their products.
Thanks, if you read such far, I hope this does not sound to pathetic,
if so it isn't meant so.
Zope3-dev mailing list