[Grok-dev] Re: Skinning/themeing

Martin Aspeli optilude at gmx.net
Sun May 20 12:07:56 EDT 2007


Lennart Regebro wrote:
> On 5/20/07, Martin Aspeli <optilude at gmx.net> wrote:
>> Are you saying that you don't envisage "new" developers picking up Grok
>> and using it to build an application from scratch?
> 
> Nope.

Good. I didn't think so. ;-)

>> This is very different from how I see (and would use) Grok - to me, Grok
>> is something I could use to build a standalone web app (i.e. if I'm not
>> working on Plone) quickly, using the giant toolbox that Zope 3 provides
>> as and when I have a need.
> 
> Sure, but then you have to make these decisions yourself, what
> template language to use, what workflow engine to use, stick with
> Zope3 schemas or use something else, and so on.

Right. But, I don't want to. I want the framework authors to make some 
strong recommendations for me. How the heck should I know whether ZPT or 
Genshi or Kid are more appropriate 
(easy-to-learn/fast/powerful/flexible/integrates well with the rest of 
the stack/etc)? If I had to understand all those kind of decisions, I 
may as well write my own framework.

I'm not saying we should make it impossible to make other choices. I 
also think we agree that having proper separation and "switchability" of 
components produces better internal architecture. But I think that the 
Grok website, the documentation we provide, and the examples we 
recommend that people look at should be promoting one "recommended" or 
"best of breed" set of tools, as far as is practical.

There are a few examples out there where this is not the case, and half 
the documentation you read seems irrelevant because it's using a 
different base technology to do something conceptually similar to what 
you want to do. You need to understand the framework pretty well to be 
able to make even simple leaps between different base technologies. Most 
people start to code via copy-and-paste. ;)

And, even if you do know how to make the leap, you're hardly going to 
know how to make an informed decision between a number of competing 
solutions to the same problem if you're just starting out.

Martin



More information about the Grok-dev mailing list