[Zope3-dev] Re: The vision thing

[Jeff Shell]

On 3/5/06, Max M <maxm at mxm.dk> wrote:
> Geoff Davis wrote:
> > Jeff Shell has posted some thought-provoking pieces on his blog that are
> > relevant to Jim's recent attempt to better articulate a vision for Zope:
> >
> > http://griddlenoise.blogspot.com/2006/03/zope-crisis-of-faith-coming-this-march.html
> >
> > http://griddlenoise.blogspot.com/2006/03/crisis-of-faith-messengers-have-been.html
>
>
> Griddle *Noise* and thats exactly what it is... noise.

Well, I do love me the noise. [ Check me out at No Fun Fest 2006,
Brooklyn! http://www.nofunfest.com/ Mar 17 2006 | http://euc.cx/aodl/
]

I say repeatedly that I'm not a good person. I'm cranky. I don't
apologize for it. I see a Zope 3 deliverable in front of me that has
half-assed versions of Zope 2 items (the page templates, etc). Zope 2
is far more mature here. I'd love to have a vision that would move
these half-assed items out of the Zope 3 application server I have
installed here, and fit them into a new vision that does match the
maturity of Zope 2.

And then, when those things are gone, I'll have a simpler Zope 3 to
start from that doesn't have those elements as strange distractions.
I'll ultimately have a Zope 3 where a third party developer might
write a much better behaving through the web Page Template object and
behavior. I'll have a Zope 3 with a simple core and easy-to-install
extensions. I'll have a Zope 3 that I will be able to tailor to the
needs of me and my customers, who have widely varying needs. I'll have
a Zope 3 that I can make lean and mean and fast, with a minimum of
configured objects and options threatening to take up memory and
resources, which is going to be very important in an upcoming project.

> He is probably following the discussions on the lists, and then he is
> publishing his view on them on his blog instead of participating in the
> dicsussion.

It's hard to keep up with all of the discussions, and sometimes it's
good to just get big ideas off the chest and open what I have to say
up to a larger group of people.

> We have probably all seen the Ruby On rails in 15 minutes video, where
> they showcase their take on nineties web technology.
>
> I could probably do one that is a lot more impressive with an UML tool,
> Plone, archetypes and ArchgenXML. And it would most likely last 10
> minutes... if I talked very very slowly.

And I don't know of any UML tool that didn't have me reaching for the
shotgun within 10 minutes. UML is great for sketching. If you want
tools to do all your programming, you don't need Python. I think it's
ridiculous that they hide the programming language and its usability
under so many layers.

> But that is not the point. He rants about Rails. But what are the
> visions of rails? I don't see that anywhere on their website either.

I think it's in pretty big type on the front page. "optimized for
programmer happiness and sustainable productivity... write beautiful
code by following convention over configuration."

One of the best visions I've seen about Rails was saying that J2EE is
Slow'n'Clean. It's totally proper. It's totally abstracted. There are
lots and lots of XML configurations. Lots and lots of separated
objects. And it's a lot of work to do something simple. PHP is the
other end of the spectrum: it's quick, but dirty. It's a lot of work
to do something 'clean'. Rails tries to provide "Quick'n'Clean". After
some quick experiences with it this weekend, it mostly works.

I think Zope could be a "mostly quick'n'clean." But right now, it's
going back towards a "quick'n'mostly confusing".

Ruby's `script/generate scaffold accounts expenses` command is
beautiful. It generates an application based on a model. No huge
layers like ArchGenXML or ArchgenUML. It uses the model in the
database. It then *generates useful and usable starting point code for
an application*. The generated interface is simple, and unlike a lot
of code generation tools it actually feels like code that you can
start modifying and working from. So you have a good starting point
for an application, and you can pretty quickly start making it your
own.

The starting point is really basic, rather bland, but usable. So you
know that you can start modifying it from there.

It gives the experience of immediate results, as well as immediate
"ok, now how do I start making it look/behave like I want?".

> He is just another developer who wants Zope to consists of only those
> elements that he is insterrested in.

I don't mind there being other elements. But I don't want them in the
way. I have different visions for me, for my customers, and so on.
Some of their needs are VERY VERY VERY different than what a full Zope
suite offers. But we still like Zope. As I've said, I've been using it
in some form for nearly a decade now. Why can't I have my mid nineties
experience again? Why can't I save my death march project in a day by
moving to two out of three core pieces?

Give Python programmers an excellent tool base built on simple and
consistent practices. This is what the Component Architecture offers.
Using this, various communities will start building the kick ass tools
that meet other peoples needs.

> In other words. Just like the rails, people he wants the freedom to
> reinvent the wheel. Poorly.

Zope and Bobo have been around for nearly a decade. Why do so many
Python programmers feel the need to reinvent it? I always thought that
if Python was to have any built in web programming concepts beyond
CGI, it should have built on Bobo. I felt that way in the nineties. I
feel that way today. Zope's core ideas are the best out there. Why
does no one use them?

Because there are very smart people out there that find the rest of
Zope intimidating, overwhelming, and getting in the way of their
goals. After getting frustrated with it, they go off and reinvent the
wheel. Or there are people coming in from PHP (like Django's
developers) that are looking for something better than what they came
from. For whatever reason, instead of using Plone/CMF/Zope, they wrote
their own tool, Django, for use in a newspaper style environment.
Funnily enough, newspapers are Zope's roots. Why don't we control this
story?

--
Jeff Shell