When I first started working for my current employer, I took a look at their CMS. Clunky, quirky, but works like a charm for what they do (they’re a magazine publishing company with over 200K hits a day). Too many CMS developers out there think that their CMS has to look cutting edge to be respectable. If you’re AJAXing up your CMS, or creating a bunch of cute little icons to dress them up, then you’re probably forgetting the real reason that it’s there in the first place.

I’ve added newsletter administration, added a WYSIWYG editor to speed up the conversion of magazine articles to web based articles, created workflows to streamline the process of pushing out the content of our shopper magazines to the web, and added many other features that we actually use. I don’t know how hard it is to add a Google Maps interface to a boxed CMS, but it took about an hour to write one and roll it out on the one we use.

That said, I’d like to take a look at EZpublish, but it still doesn’t support php 5. I think it’s time everything supported php 5, as it has been out for 2.5 years.

The advantages can be summed in that there’s already been a significant amout of design, planning and programming gone into them. You’re getting the benefit of lots of experience and feedback already. Many things that go to make up a CMS are already there. Templating, workflow, rollback, integrated security, and a simple edit interface are taken for granted when choosing an out of the box system.

You have to be sure you’re choosing the right CMS. More specificially, the right size CMS. You’ll usually have to accept, that without modification, there can be differences between requirements and expectations (for instance the way security, groups and workflow are managed).

A bespoke system will have greatly reduced functionality it all has to be written. It will take more specialised local knowledge. If the guy who designed and implemented it walks out the door, that knowledge is going to be hard to replace.
Evolutionary growth can be a problem too - without careful planning, a bespoke system which keeps getting modifed and extended can exceed its original architecture.

Take any application. I use Word but probably have never used it for more than typing out my resume. Someone else, on the other hand, has probably developed wizz-bang apps, using macros and VB, to develop cross-office suite masterpieces (who knows? Maybe even a CMS!). My point being, just because the functionality is there doesnt mean that it has to be used. Just work with what you feel you need to best manage your site.

Where was this post when I had to coax a client off the roof because I told them “Hey, once we get this up and running you can add content yourself!”? Where was this post when I joined the client up there on the roof?

probably has a half-dozen more features than you need.

I agree with many of the problems with a boxed CMS, but I think the modular approach of Drupal helps deal with this issue.

And then the controlling party of your boxed CMS —be it a vendor or the open source project—releases an update. And, more often than not, this update breaks your changes, rendering your implementation of the application useless.

It helps to use version control e.g. subversion.

For sites that need limited content management I have been using CakePHP.