[BangPypers] [pythonpune] Re: To use Pinax or just Django

[Navin Kabra]

I have a different perspective. I think Django is a good choice, especially
if you don't know what you want.

On Mon, Nov 21, 2011 at 1:40 PM, Noufal Ibrahim <noufal at gmail.com> wrote:

> I've done only one non trivial project with Django and my impression was
>  that you are bound to do everything the Django way. You cannot select
> the ORM, templating engine etc. and expect the rest of the things to
> work as smoothly as before.
>

Maybe because none of my projects have had "extreme" requirements, I am
happy to go with Django as far as the core components - i.e. I've not felt
the need to go out and get a different ORM or Templating Engine or user
management module.

For the other requirements (like tagging, social authentication,
autocomplete) I've been very happy with how easy it is to find existing
third party apps to do, and also how easy ti is to extend django with your
own mechanisms of doing thing. And, all of that has been possible without
having to modify any django code (unlike Pinax).


> Also, some parts are opaque and I
> like to know what's going on inside.


I've found Django code to be very accessible - so a few minutes spent
reading the code has always been good enough for me to understand what's
going on; so I don't agree with the opaque comment.

If you're doing everything which the core team has thought of in the way
> they want to do it, well and good. If you want to do something
> differently, there's a good chance you'll hit a wall.
>

I think this statement is a bit too strong. Yes, you will run into problems
if you're replacing something as fundamental as the ORM or template
langauge or forms or user management. But for other things, Django is
surprisingly flexible. If you're using signals, context_processors,
middleware, and other things smartly, django definitely does not give the
impression that "you have to do everything in the way the core teham wants
to do it".