Content Management System
Oct. 31, 2006 by ravishan
We have had many discussions about this topic over the years and the difficulty has been capturing the questions that we are trying to answer. Mike Roy assembled a group consisting of ITS and Library staff members earleir this year to address this issue. You can read more about this group’s work at http://www.wesleyan.edu/atr/content_management/.During the last couple of meetings, it became obvious that the issues brought to the table by the various groups still remained complex and diverse and our attempt to possibly solve many of the needs with a single solution may not be feasible. What became apparent was that every single group had a plan that required a lot of disk space!
In addition, the issue of trying to manage the fragmented space that is currently available to our community seemed to be one on which we could take some action. For example, today, a faculty member has space on Dragon, condor (for both personal web space, which is different from course web space) and a fileshare space for collaboration with colleagues here and elsewhere. The proposal is to bring all of this together into one space that is easy to understand. The solution also has to be a cross-platform solution so Windows suers, Mac users and the Linux users can all work with it.
The easy part is the acquisition of very large disk space… It is back end technology that a user does not see directly, so the decision involves mainly the technology support services staff. We have already looked at one very good solution from a company called Compellent and are waiting to hear from HP. The decision on this will be made soon. We are talking about some serious disk space here. We may initially acquire 20-25 TB, but we are sizing units which can easily be expanded.
We want to proceed on “logically” consolidating the space available to faculty. The two major requirements here are:
- It must be easy to use. One of the suggestions is that it appears as a drive on their desktop with a very simple hierarchy.
- The user must be able to control access easily and doesn’t have to rely on ITS staff every time they need to do this.
We looked at a cross-platform software solution from Xythos. It is a very impressive software and is entirely based on the WebDAV protocol. Basically all file management from the desktop takes place over the http:// or https:// protocol and the software hides many of this in a way that the user may feel that they are working with a mounted drive. Access control can be done through the web and notifications of sharing is handled via email. A tomcat server manages all requests and security is maintained in database tables. They support most of the common database platforms.
We plan to look at a couple of more solutions before making a decision.
Here are some of my observations:
- These solutions are very expensive.
- They are proprietary, so any commitments we make are for the long haul.
- A clear plan to migrate the existing content to the new system, preserving all the security settings must be discussed before we commit to any software.
- It is not clear how many members of the community will actually take advantage of all the feature set of these software – this is certainly important in deciding if we want to commit valuable dollars to such a software.
- Having spent all of these resources, we would want to convert all users to use this. Whereas this may look trivial, in the end it may not. We need to explore this carefully by including representative users from various departments in the evaluation process. I know that several administrative departments have elaborate ways in which they use Dragon today.
- Whereas giving the power of access control to the end user is very appealing (and I support it whole heartedly), we should also be aware of the pitfalls. The file access may be inadvertently be opened up to users that should not have access. This can especially be dangerous if it happens in the administrative departments.
- This would require changes to the way we currently do business – all file management MUST go through this system for things to work. Some users cannot say that they prefer to sftp their files, for example.
- So, while we go through this process, we should ask ourselves if there is something we can do to address the questions on this topic, or is there a public domain software that can do the job.
I invite you to chime in… Don’t be shy. This is very important and is likely to affect all of us, so it is better that we discuss the issues and find the right solution.

As usual, I’d recommend looking at Open Source tools that do a similar job as Xythos (I know, big shock there – what can I say, I’m an alumni who’s sensitive about Wesleyan spending money). I don’t know what’s out there to fill that gap out, since I’m not sure.. er.. what Xythos *really* does having not seen it in action, but there’s many free “content management systems” being used by people far bigger than us, so it’s not an impossible pipe dream. Also, this lets us test the waters in a far less intruisve way to the clients with minimal cost – we can try something like Plone with a few users without a significant commitment of resources (to be fair, this could be possible with Xythos and a demo, I would hope so, but I’ve no idea).
I was actually on one of these groups at one point, but I certainly made no serious investigation into Xythos-like possibilities since they (like Xythos) did not meet the requirements as I understood them.
Nevertheless, at least identifying one “open source”/”free” tool to toss into the ring seems valuable; if someone wants to point me to a Xythos demo of sorts, I can probably come up with something close to it…
try
http://www.opencms.org
-Henk
Any interest in me installing this somewhere?
It seems to serve a different purpose than Xythos (as explained to me, Xythos focuses on file sharing not necessarily web creation, whereas OpenCMS seems to tout creation of web pages as a major feature). Like I said, there’s tons out there but I’m completley lost as to what we’re looking for and am simply not qualified to identify what’s “user friendly”.