CMS Decision

Elliot Lee sopwith at redhat.com
Mon Dec 5 20:27:37 UTC 2005


Thanks for spurring this effort along, Patrick!

On Fri, 2 Dec 2005, Patrick Barnes wrote:

> We really need to move forward on selecting a CMS.
>
> It is highly unlikely that we'll see a PHP solution selected.  I've
> heard many calls for Drupal, but let's not count on it.  In particular,
> if a PHP solution is chosen, we'll have to be very, very careful in
> picking it out, and we'll probably need to disable any more dangerous
> functionality.  In the case of Drupal and many others, that would
> include XML-RPC.

Agreed.

> There aren't many Python solutions out there, but we need to scour what
> is available very carefully.  Many people were or would have been
> doubtful of what we could accomplish with MoinMoin, and look at all
> we've managed to do.  The Fedora Project wiki is at the top of the list
> of MoinMoin's list of example wikis[1].  We can accept that there may
> need to be work done on whatever solution we select to get it where we
> want it.  We've got a large pool of talent that will be wiling to help
> with that.

So I've been looking around for a CMS, on and off, for proably four or 
five months. I'll just throw my ideas so far out there as a starting 
point:

	. Zope. We just have to look at it seriously because it's in 
	Python. Its main weaknesses are performance and ease of use.

	. Drupal. PHP. Very cool in general. Main weakness is the security 
	concerns. It also lacks a version history of pages, which is 
	something important to me personally.

	. Typo3 is a CMS-type thing in PHP. Its biggest shortcoming when I 
	tried it was its complexity.

	. Midgard is a PHP CMS that I haven't looked into. Someone should 
	evaluate it, because its features look very attractive overall.

	. Turbogears and Django. Two Python web "frameworks" that make it 
	easier to write web apps, but don't offer an actual CMS 
	themselves.

	. Various Java solutions such as Apache Lenya. Haven't looked into 
	this space much, even though it's probably where the most
	action is.

	. What we've got already. The basic idea is perfectly sound, and 
	everyone knows how to use CVS. This is the baseline that the new 
	solution needs to improve on, so one good question is "what do we
	need that we don't have now?"

I do not consider wikis a viable starting point. They don't seem conducive
to keeping the site organized. Maybe there is one out there that has the
features necessary to do this, but I haven't seen those features in
Moinmoin. :)

http://www.la-grange.net/cms has a list of open source CMS's.

Best,
-- Elliot
Unanswered questions in The Matrix: What happens if you take both the red 
pill AND the blue pill?




More information about the Fedora-websites-list mailing list