[development] D7 contrib module development

[Marcel Partap]

> We've always had stuff that falls off the face of earth. So what? The
> caravan continues on ...
Sorry but i don't consider that a logically valid argument.

> The very first module I contributed (feedback) illustrates a point: It
> was started in 2002 by someone called "barry". I had it working I took
> it over in 2004 with totally new code for Drupal 4.5. Then "fago"
> overhauled it a lot in 2006. Over time, the contact module in core came
> along, and I stopped using feedback. Then in 2008 "sun" took it over and
> repurposed it with new code.
Well good that process worked out than, btw i am actually using the 
module ;)
But from the experiences gained, how would you design such a module 
from scratch? How could it be integrated with other modules, what 
could be factored out into 'frameworks' (/library modules) ?
That is the stuff each of the drupal code contributors needs to think 
about as early as possible in the D7 cycle, which would be right 
about.. _now_!


> The "too many modules in contrib syndrome" can be taken as confusing,
> excessive, ...etc. but can also be taken as a sign of a healthy and
> vibrant community.
Sure. But what is the prior goal of the Drupal project? To create a 
healthy vibrant community? Or to create the best open source CMS CODE 
out there (best accomplished through a healthy vibrant community)?

> So what if we have a few extra gigabytes of code? So
> what if they become unmaintained?
Uhhm sorry i don't share that so-what mentality. I want to have *all* 
the functionality spread over slightly different modules with the same 
purpose, combined into one flexible solution. And i am quite sure i am 
not the only one.

> If we raise the barrier or block new entries we will be shutting
> ourselves off from being the platform for the new chx or the new
> merlinofchaos.
Woow stop for a minute please.. so you're saying people like chx or 
merlinofchaos are not capable of adapting to coding standards and 
qualified to contribute code which can be worked on to raise over the 
ready-and-useful barrier to be included in the official Drupal repository?

> It does not matter ... if it is the wild wild west, then let it be. It
> is a small price to pay for innovation and the power of the masses.
But the thing is, sometimes innovation needs channeling. Granted, in 
the early phases of the industrial revolution it would have been 
counterproductive to regulate machines and processes in any manner, 
simply because not enough practical experience had been gained. But 
once a certain level of sophistication is reached, there is no way you 
want to live without agreed standard interfaces and common norms. 
Imagine todays industries without all the ISO standards for screws, 
bearings, gears, quality management.. What do you think which kind of 
a car would you drive if none of them would exist?
In my opinion with D7 we have reached that level of wild life 
experience to benefit from a more ordered development process. To say 
again, why trade in quality for quantity/speed? Most people will be 
quite happy with D6 for years to come.

rgds marcel.



-- 
  "Obstacles are those frightful things you see when you take
   your eyes off your goal."         -- Henry Ford (1863-1947)

   Change the world! Vote: http://hfopi.org/vote-future