Ambrosia Software's Snapz Pro X is the standard Mac tool for advanced screen capture stuff.
Though most Mac folk use Quicktime movies for this kind of thing, and I think most of the other mentioned options for other platforms produce Flash video as output.
I thought the idea was that, as it was a live spreadsheet, you could tailor your own weightings and see the results change.
Surely putting out misleading evaluations without making it 100% clear is poor practice. I'm sure the projects that did worse would be particularly unhappy with it.
I'd think a 'web application' category would be a good fit for Moodle as many of the questions assume you either have a local installation of a 'program' or a 'website' and Moodle (and WebCT and Blackboard, ATutor etc.) is somewhere in between.
I've added some stuff to the wiki about their accessability requirements, I hope it's useful, but because of the mismatch talked about above it's a bit of an square peg / round hole situation.
Apparently the big cheese at BECTA used to be a bit of a loyal Microsoftie, but has been recently replaced with someone who has an actual interest in getting value for our (taxpayer's!) money. (That's all second hand hearsay, so take with a grain of salt)
BECTA's recent study on Open Source software was very good reading and certainly suggests a change of heart:
I'm sorry to come late to this discussion but I thought that this set of guidelines was more appropriate (and more in tune with how Moodle.org is currently):
http://www.css-discuss.org/policies.html#ask-answer
I like it because it focuses first on encouraging the person to actually ask a question, because the only thing worse than a 'dumb' question is one that goes unasked through intimidation. It then tackles the other end of the equation by explicitly prohibiting negative responses (which thankfully, and almost uniquely, are not a problem currently on Moodle.org).
It's also worth pondering what has to occur in the life cycle of a Question before it deserves the title of Frequently Asked Question.
Length of contract and version details are very important too.
WebCT Vista for example seems to be around an order of magnitude more expensive than their legacy codebase WebCT Campus. They are currently backporting a great deal of functionality from Vista with the aim of totally replacing the current Campus technology. After their customers panicked they issued a statement that the Campus price-point would still remain even after they have totally replaced the codebase. However the rather radical changes to the underlying codebase mean you will now need, for WebCT version 6 an Oracle or MS SQL server licence and associated hardware and technical support.
The length of contract is important too as they generally try to lock you in for long periods but the price can rise dramatically when you come to renew the contract (I've been told recently of 30%-40% increases, not including the newer database licences)
I've kind of lost track of what Blackboard is doing but I believe they still do the same thing as WebCT whereby they limit the amount of integration you can do unless you pay a fair bit extra. When I last looked in detail at this the price would double or triple if you wanted the system to integrate with your student records database (i.e. that's not counting the consulting and development costs, just a cost for even being permitted to do it).
So I think to be truly useful the table should include the version (in as much detail as possible), how long the contract was for, and when the deal was done.