[Zope-dev] Defining Zope 3.

Martijn Faassen faassen at startifact.com
Fri Apr 17 05:58:40 EDT 2009


Hey Martin,

Martin Aspeli wrote:
[snip]
> Sigh... this discussion is just really difficult. I don't really 
> understand what the problem is here, or why it's indicative of what's 
> "wrong" with this community, but then I'm pretty lost in concepts and 
> names at this stage.
> 
> I think you're reading way too much into what I wrote though. I just 
> meant "thing using the Zope Toolkit but adding more stuff", e.g. an app 
> server project or a shared management UI project.

The difference is I think was that I'm not talking about just code; a 
KGS is just a list of versions. I'm talking about code and documentation 
and a presentation of this code as a whole that people can learn about 
and play with.

The notion that it's just enough for Zope 3 to be pieces of code is part 
of what led me to Grok.

If the notion of Zope 3 can be limited to just a greater set of packages 
where compatibility is tested, it's not really much of a project to 
speak of. That's fine, we have other projects like Grok that do care, 
but I'll stop worrying about it.

> I'll say again, though: Gary's version of the story ("the Zope 3 
> community has become focused on supporting other app servers and 
> frameworks, and is renaming the software stack that serves that purpose 
> to the Zope Toolkit" reads pretty well to me). Better than the other 
> stories I've seen here, because it doesn't really concern itself with 
> specific packages or features or a delta of those against a hypothetical 
> smaller "toolkit". In other words, I have a pretty good idea of what it 
> means just from reading that sentence, and I can draw some conclusions 
> about what it may mean for my existing Zope3-based projects and what it 
> may mean for other projects (Grok, Zope 2, bfg) that have used Zope 3 
> components.

Sure, it's a reasonable approach. I am just frustrated that the notion 
of a project that is something *more* than the Zope Toolkit seems to be 
so incredibly hard to explain in this context. That in itself speaks for 
Gary's proposal, as it's pretty easy to explain.

The alternative explanation is that people understand what I'm talking 
about just fine. After all I'm talking about a project like Grok (or 
Django or Pylons or TurboGears or Rails or BFG) but one that takes a 
more "traditional" approach to configuring things (ZCML). The project 
described in Philipp's book, for instance.

People may just all not care about it? People only seem to be interested 
in attracting new users to (bits of) this platform the context of Grok 
or BFG or Plone. I find it interesting and somewhat frustrating, but 
that's all history and I don't really need this project to exist anyway. 
I just thought that some people on this list do need such a project.

So I'm ready to just go with Gary's plan and present this as a renaming 
recognizing that Zope 3 has become something very different, as there 
really indeed doesn't seem to be anything else left.

Regards,

Martijn



More information about the Zope-Dev mailing list