March 20th, 2009

KDE's brainstorm

I'm still quite wary about the brainstorm concept as being a good use of resources, but maybe upstream projects like KDE can do a better job of leveraging the crowdsourcing concept than downstream integrators can.  Having a big pile of thousands of ideas with millions of votes doesn't mean the good ideas will get implemented. That's not how the meritocracy on which the ecosystem works. 

What is right is not always popular and what is popular is not always right.  It really comes down to whether or not populism translates into good implementable design. If after several months and years the most popular ideas are consistently not being picked up and implemented, what purpose does it serve to make popularity a way to organize the wishlist space? 

This contrasts with how kerneloops.org is suppose to work. The organization of kerneloops showing the most common oops scenarios results directly into developer efficiency as its organizing information that current developers already care about addressing. I'm not sure the crowdsourcing brainstorm concept inherently does that. 

I hope the KDE developers regularly assess the effectiveness of the Brainstorm concept's ability to better organize existing developer time and to effect the development roadmap.  If the ranked list of popularity doesn't correlate well with feature roadmapping and development that could indicate a fundamental problem with the idea behind Brainstorm.  So far I'm not sure any existing Brainstorm implementation has shown its worth as a way of organizing developer output.

-jef