Rob Hoff wrote an interesting piece on wiki’s called Learning to Work with Wikis about the Business Week news team using a social text wiki to collaboratively compile their Best of the Web list. They did a number of things right:
1) They waited until he had an actual need, to compile a collaborative list and keep work efficient, and then sought out an appropriate social software tool; a wiki in this case.
2) He limited the authoring on the wiki to a group of interested people, or knowledge-matter-experts.
3) They sound like they approached it with a sense of playfulness. Social software does not work in my opinion without a sense of "what if" and "wouldn’t it be cool if" is behind it, at least initially.
Some observations on what he could have done to improve the use of the wiki.
1) Rob noted that he probably limited the authoring a bit too much. Not enough authors, or editors, and you lose the creativity of a free flowing group. Too many and you wind up with a spawiki (spam wiki).
2) Did not use, or at least did not mention work flow. My personal belief is that wiki implementations need better work flow management. He may have used work flow, but I’d sure love to hear about the work-flow even it the implementation of the work flow was social (verbal, email, offline, whatever).
3) He chose as a wiki test a topic that lends itself to opinion. Not everyone is created equal despite the wisdom of crowds and when compiling a "best of" list you obviously have opinions which prevent the use of NPOV (Neutral Point of View). So from a proof of concept point of view this was a tough selection. Sort of like you really should fly an RC airplane before trying an RC helicopter (different story…)
Overall I’d like to see more journalists working with wiki’s internally or in a controlled but enthusiastic fashion to help move the media past the LA times wiki fiasco.