[Zope3-dev] Re: visions, brands and roadmaps in the sand

[Martin Aspeli]

On Thu, 02 Mar 2006 19:39:54 -0000, Lennart Regebro <regebro at gmail.com>  
wrote:

> Technology-wise:
>
> 1. Zope2 uses more and more of Zope3s technologies. Anything that can
> be merged gets merged. That means security, pagetemplates and the
> publisher, and maybe more. Basically, we want to strip Zope2 as much
> as possible, until Zope2 basically runs "on top" of Zope3. This also
> includes making sure that a Zope3 product can continue to run on
> Zope3, even when Zope2 also runs there. I.e, we get to a situation,
> where you can run Zope3 products unmodified in Zope2.

+1

> 2. Zope3 may also get slightly streamlined, so that Zope3 is what is
> needed to run Zope2 but not more. That means that Zope3 would lose the
> ZMI.

+1 in the context of Jeff Shell's vision about having one version with and  
one without the ZMI (in whatever capacity) as a useful data/settings admin  
interface.

> 3. We start rewriting CMF tools as Zope3 utilities, with the CMF tools
> as a thin BBB layer (that may finally disappear). This becomes the
> basis for the ECM "toolkit" for Zope3.

+1

> 4. We develop TTW development tools for Zope3. I think me and Jim both
> agree we should have these. I think me and Jim has a completely
> different vision of what they should be. :)

+1 (though lower priority for me at present... I'm more concerned about  
breaking CMF TTW template cuatomisation etc. in a Five world)

> This, as far as I can see, is compatible with 99% of the visions we
> have discussed here, and it will keep us busy for a year. :-)

Probably more.

> Marketing-wise:
>
> I can see the point in separating out the non-webby parts of Zope3 to
> gear it towards people who don't want a big applictaion server. And,
> to be honest, if I did anything in python today, I'd want to use teh
> component architecture, even though what I do didn't touch the web.
> I'm not sure we want to name it Z but i'm not completely opposed
> either. But I prefer Zope Component Architecture.

+1, I like the Zope 3 CA idea.

> I am opposed to renaming Zope 3. It seems to me that the renaming of
> Zope3 is suggested for three reasons:
>
> 1. "It makes Plone look bad by not running on the latest version of  
> Zope."
> I honestly couldn't care less. Sorry.

I don't know of anyone in the Plone community who's uttered this  
sentiment. I certainly don't care.

> 2. "Python people doesn't like things called Zope".
> Well, Zope3 was heavily geared towards making Zope pythonic. What has
> the answer from the Python-crowd been? "Eeew, you are using XML!".
> Honestly, I don't think renaming Zope3 will suddenly make the
> python-crowd go "Oh, cool, you use XML!" I could be wrong, but that's
> what I think.

But a Zope 3 CA that's non-web-focused and continues the ZCML slimming  
tradition (if I can make adapters and interfaces without ZCML I think I'll  
be happy - browser pages and views etc. don't matter in this case) and can  
be separately deployed/depended upon makes sense.

> 3. "There is not enough hype!"
> You don't need to rename Zope3 to hype it.
>
> I think renaming will mean we loose a brand, and we make people
> confused, and we need to change a whole lot of code. :) It doesn't
> seem worth it.

I was always behind the idea of naming it Zope 3 <something>, not lose the  
Zope or the 3. The <Something> is to have something to emphasise a brand  
around. I think Paul had a really good post about this, though. I think  
even Zope 3 CA and Zope 3 AS would work fine as names to build hype around.

Martin

-- 
(muted)