This is something I put some time into a while back, and while I never finished it, I think it is something that needs to be done.
When I first looked at the front page of OpenACS, I dismissed it from consideration. After working through a number of CMS solutions and not finding one that did what I wanted, it became time to select a toolkit that had a framework that wasn't too restrictive.
The one toolkit that was the closest was Midgard. For reasons that I won't go into, there were reasons I reopened my bookmarks and started to look at OpenACS again.
Without feeling pressured to put a new description in quickly, I might suggest that we take some time to put the features together in a list that made each of us choose OpenACS, then, combine those terms and features into an elevator pitch followed by a more descriptive blurb.
My target market is probably different than yours, so of course the features and strengths I see are certainly going to be different. However, I think we can write a compelling statement that will lead casual browsers in to look more deeply at OpenACS's capabilities.
So, with that in mind, I think the way to market OpenACS is to mention the packages that would be put together to make a generic system for a client installation. Allow someone to picture in their mind how the pieces fit together and create a solution. Then, throw in the fact that you can plug in whatever packages that are needed to come up with a total solution. If there isn't a package that does what you need, utilizing the defined framework/API to create a solution is much easier than writing a solution from scratch. (Point well learned this weekend)
There has also been mention on IRC and in these forums that people don't like the fact that this doesn't run under Apache for this or that reason, or that it is in TCL and not PHP, etc. Dispel the rumors or myths. I wouldn't specifically go after the bad parts of Apache, but I would of course illustrate the strengths.
This is just off the top of my head and a very broad stroke:
AOLServer was chosen because it has the scalability and built in parsing ability, etc, etc. AOLServer is not a forking server, allowing it to scale much easier, etc. AOLServer handles the traffic of many of the busiest sites on the internet.
Oracle was originally chosen, and later PostgreSQL because they are ACID-compliant databases with established track records of reliability and rock-solid performance.
OpenACS is a framework that runs on top of AOLServer and Oracle or PostgreSQL which allows you to pick and choose the toolkit components to develop a site solution for your business/school. The OpenACS package library includes packages to handle Press Releases, News Articles, Forum Capabilities with Notifications, Calendaring and Events management, etc. etc. Pick from many other packages to enable your site with a Bookshelf or Photo Album, etc.
A testament to the ability to use the OpenACS framework is the .LRN project which enables Universities and teaching institutions to support course management, e-learning, departmental and schoolwide collaboration, etc, etc.
OpenACS has the tools and capabilities to scale with your organization with its ever growing repository of additional packages. OpenACS is open source and operates with PostgreSQL, reducing your TCO. Because the framework encompasses many standard methods, developers can use defined APIs to handle tasks, dramatically reducing time to market on enhancements. (after a very very steep learning curve haha)
Just some thoughts.
