Participants: J.B. Rainsberger, Michael Ben-David, Eugene Zozulya, Gil Broza, Beverly Dawe, Ross McKegney, Dan Shields, Charlie, Trainor, Siamack Jabbarzadeh Farshi
We started out informally, chatting about hiring as a practice and the importance of good people, although we didn't agree on "smart." We then got into the meat of the topic.
The expression "great software" is highly personal and definitions varied within the group. Elements of "greatness" identified included:
Practices that historically led to great software varied, including the "Mad Scientist" model, the "Open Source" model, and the "Research Lab" model.
Many of the practices we identified we agreed were either orthogonal to or hygienic to great software. We had some variance based on different ideas of "greatness."
We felt strongly that Agile Values were strong drivers of Great Software, especially communication for teams with two or more creators, however specific practices and implementations (XP, &tc.) were almost all orthogonal.
We ended much where we began informally: by identifying talented people (not necessarily as measured by http://www.techinterview.org/) as the strongest driver of Great Software.
Although talent is no guarantee, we're sure that great software is never produced by a team that's criteria for hire is "x years of foo" and "y years of bar." J
We all want to create great software.
Suggestion: if you decide to take this discussion offline to a newsgroup, please indicate the location of the group here so others can join you.