One capability that may be specific to the type of communities I support (though I expect applies more widely) is the ability for a community member to easily extract their contributions (and possibly also the contributions of others) so that they can use them in other parts of their online lives. Alternatively, the ability for a community member to easily contribute materials that have been developed elsewhere.
The majority of online communities I participate in aren’t organizational in nature – the cross institutional and organizational boundaries. As such, it is likely that they are not the only community that any one of the participants is a part of. Software that acts as a restrictive ‘container’ where a community member can make a deposit but not a withdrawal or a transfer is of less and less interest to me. Software, or the models we set up, needs to recognize that most of us are a part of multiple communities and thus must help (instead of hinder) in participating in as many of them as possible given our limited time and resources.
I think this is important not only because we are members of multiple communities, but also because we are taking more responsibility for our own learning and we need traces of our thinking to reflect and to learn (recent example: Circadian Blog Rhythms). This need also explains why weblog can take place of participation in forums.
See also related post: Who owns narrated experiences?
Btw, definition of social software in the same comment:
This is why I think the whole ‘social software’ movement is in fact different from many of the collaborative technologies we’ve seen before – it’s software that is centered around individuals (instead of the community ‘site’ or server) but that creates conjunctions of these individuals by accepting various interfaces, feeds and formats from those individuals and coalescing them.