There's a really interesting paper posted on the MIT website, Understanding knowledge sharing activities in free/open
source software projects: An empirical study, that did an in depth analysis of the Debian mailing lists.  They came up with all sorts of interesting statistics like:

• 2% of the knowledge providers were responsible for about 50% of the answers to questions posted on the help system
• 50% of the questions were provided by 24% of the knowledge providers

They also considered the developer list separate from the user list:

• 32.8% (1224) of the participants in the Developer list posted one email message and 19.3% (721) contributed one reply
• For the User list 16.1% (963) of the participants posted one email message and 22.9% (1369) contributed one reply

And who was asking versus who was replying:

• in the Developer list, the participant who posted the most emails (523) contributed 574 replies
• While the participant with most replies (1517) contributed 79 posts

These can basically be summed up to mean that a very small portion of the community is doing most of the talking/sharing on the mailing lists.  (It will be interesting to see if this holds for other communities besides Debian – the authors plan to do that next.)   I don't think really comes as a big surprise – the open source community is a primarily a volunteer community.  Even those that get paid for their work, don't usually get paid to answer questions on the mailing list.  (At least not in the Debian community.)  If they get paid to answer questions, it's the paid questions coming in through the company's support line.  So, people are on the list are answering questions because they like to.  And I think people that like to answer newbie questions are a small subset of the overall developer population.  What percentage of the overall software developer population loves to do support?   A very small percentage.  (Just ask my parents how reluctant I am to help them trouble shoot their problems.  I love helping them and I hate walking them through all the possible things that might have gone wrong.)  

Now, I think a large percentage of the community is willing to fix bugs – but that's different than answering the hot line and walking a newbie through the process of uncovering the bug. 

Matt Asay used the statistics to show that if you are not a developer you are better off paying for open source support than relying on the mailing lists.  I agree.  You either have to have developers in house that can help trouble shoot your problems and interact with the community or you need to buy commercial support from someone.  (Like OpenLogic!)

I look forward to seeing what the authors, Sulayman K. Sowe, Ioannis Stamelos, and Lefteris Angelis, uncover next!