I am currently writing up part II of the blog post "What would science look like if it were invented today?" Part I was focused on knowledge creation in the post-paper era and drafted in a wiki. Part II is focused on knowledge structuring, and as an experiment, I have ported the current draft from the wiki into an Etherpad document which anyone can edit, and embedded it here. Feel free to join in (just one condition: please do not break existing wiki or HTML syntax), or leave me a comment.
This is a response to http://www.wittylama.com/2009/09/wikipedia-journal/ - a new (still hypothetical) initiative consisting of an Open Access journal (with ISSN and CC license) that publishes scholarly reviews that are peer-reviewed and ready to be pasted into Wikipedia.
I like this idea a lot. It is much like Scholarpedia (which contains commissioned but anonymously peer-reviewed articles and which has an ISSN but no coherent license), just has a broader scope and does not confine itself to the top-notch experts in the field. Your proposal, as mentioned above, also bears some resemblance to Citizendium (where the review process usually involves domain generalists rather than topic specialists, and it is non-anonymous; has CC license but no ISSN). Both operate stable versions that can be updated. The former allows for attribution, the latter not.
You also mentioned that a similar journal could be set up for original research (something that the Wikipedias, but also Scholarpedia and Citizendium have avoided so far), and in this regard, it is very close to the journal PLoS ONE (meant to be for all scientific disciplines, though currently with a bias towards the biomedical fields; has ISSN and CC license) and the recently launched PLoS Currents (which, in essence, uses Knol as a preprint server), which I have commented here.
I am also currently drafting a blog post for the Euroscientist on these matters at Wikiversity, to which everzone is welcome to contribute and where a number of related posts is referenced. To quote from just one of them: "science is already a wiki if you look at it a certain way. It's just a really, really inefficient one - the incremental edits are made in papers instead of wikispace, and significant effort is expended to recapitulate the existing knowledge in a paper in order to support the one-to-three new assertions made in any one paper."
"Fantasy Science Funding" is an online game played by people with concrete ideas about science funding who are not currently in a position to put these ideas into practice. There are five rules to the game: 1 - choose a funding body whose funds you are managing in your fantasy, 2 - imagine how their funds could be distributed to the benefit of science, 3 - choose areas of science to be "fired" (i.e. whose funding should be decreased with respect to present state), 4 - choose areas of science to be "hired" (where funding should be increased with respect to now), 5 - blog about it.
Previous shindings that I am aware of were hosted by Duncan Hull, Björn Brembs and Cameron Neylon.