A smart webdeveloper nowadays has a couple of CMSes up his (or her) sleaves that he can sublicense to customers who want to be able to control (part) of their own website. Work that the webdeveloper used to do himself, is now done by somebody at the customer’s.
This is a good development, because it allows one to put responsibilities where they belong. If customers want to change the news, update prices, add more information, they should not have to go through the developer first.
But of course, conservative webdevelopers might see this move to smarter customers as a threat. It used to be the case once that webbuilding was more a craft than anything else. Only a select circle of initiated knew how to build a cross-browser, cross-platform, cross-everything website.
In order to familiarize myself with CMSes, I downloaded and studied a lot of the open source ones. There wasn’t really much choice; a lot of the commercial CMSes cost more in licensing fees than small and medium businesses would ever want to pay for their website.
The problem with all CMSes that I have encountered in this way is that they are too difficult to set up. Weird dependencies (Zope comes with its own webserver!), badly written or non-existent documentation, and even if you do manage to set up the CMS, that’s no guarantee that you will be able to easily convert a site design to it.
Still, I would not have suspected anything wrong with this situation. I can accept that CMSes are incredibly difficult to program, and that support for gratis CMSes therefore tends to lack due to lack of resources.
However, there are tools that are not CMSes, but that offer similar possibilities and that often are (in my experience) extremely easy to set up and maintain. These tools are wikis, blogs, forums and community software (call them the Nukes).
It seems to me that these easier tools have all the potential to supplant CMSes, at least in the non-commercial space.
And by doing so, they bring the possibility to the websurfer to edit the pages he visits, thereby fulfilling the last unfulfilled promise of the web.
See also: The blog systems that made it as CMSes (2010).