On Sun, Jan 17, 2010 at 9:43 AM, Chris McDonough <chr...@plope.com> wrote:
> Hanno Schlichting wrote:
>> On Sat, Jan 16, 2010 at 11:57 PM, ken manheimer <ken.manhei...@gmail.com>
>>> i'm getting the impression that you have a talent for understatement. :-)
>> Yeah. People keep telling me that, no idea why :)
>> through a happy accident, a search through my inbox for "repoze" turned
>>> shane hathaway's february 2009 zope.pipeline proposal. it's very
>>> illuminating. it definitely helps me understand more about what all the
>>> fuss is, and more of what's going on.
>> In the end the handling of a request/response and the notion of
>> publishing is on of the core parts of our web frameworks. For a long
>> time we had relative stability in that area, with ZPublisher and
>> zope.publisher being the only two options. Today there's repoze.zope2,
>> repoze.bfg, bobo, the zope.pipeline proposal all trying to improve in
>> that area. But these are technology innovations only usable in their
>> respective frameworks. Unless you want to work on improving the
>> publishing story yourself, you can largely ignore these debates and
>> stick with whatever each tool gives you. Nobody can tell if we are
>> going to diverge further or if ideas will converge again at some
> For the record, Shane indicated to me earlier this year that he doesn't
> intend to see zope.pipeline to its logical conclusion; instead, when it
> makes sense, he'll pull features out of Zope and into middleware when those
> components can be used independent of the rest of the Zope stack.
interesting. the perspective in his proposal on the zope plumbing is still
mighty informative, at least for someone as marginally informed as i am, and
it is good to know i shouldn't expect zope.pipeline, per se.
> tres has often referred to the ironic chinese curse, "may you live in
>> times" - these are interesting times, indeed, for a web application
> Yes, they are. Zope has switched from being one or two major projects
> into a whole landscape of different options going in all kinds of
> different directions at lots of different levels. You have to evaluate
> these tools independently and see how each one does for the job you
> have in mind.
> I think there are only three true Zopey options at the moment for
> application integrators: Zope2/Plone, Grok, or BFG. All the other stuff
> (like Zope3, repoze.zope2, etc.) are more or less either just libraries or
> promising thought experiments, and integrators should ignore them.
that's helpful. as i mentioned in my previous posting, the accommodation of
bfg on google's app
especially interesting as a prospective resource for as a "glue" for
Repoze-dev mailing list