I just saw the video from O'Reilly on SAP as a Web 2.0 Company: Your Conversation and I was pretty impressed with its quality and the message that it conveyed. I thought it would be pretty interesting to try and apply the lessons / rules that Tim described in his speech to the SDN and BPX communities. Although you might suggest that Tim's speech referred primarily to commercial Web 2.0 sites, I think it is useful to apply these rules to our communities in order to find ways to improve them.
Tim's Web 2.0 Rules
The main message from the keynote was that data aggregation should harness the collective intelligence and is the main ingredient for a site's success in the marketplace.
The ability to achieve this goal is based on:
- The key to competitive advantage in internet applications is the extent to which users add their own value to that which you provide.
- Only a small percentage of users will go to the trouble of adding value to your application. Therefore: Set inclusive defaults for aggregating user data as a side-effect of their use of the application. Harness self interest, not volunteerism.
Thus, the main question concerns the ability to exploit users' self-interest to contribute information that increases the value for all.
The role of self-interest in SDN and BPX and how to manipulate it
We had a few interesting discussions at Community Day at the TechEd in Munich regarding the current state of affairs in the SDN forums. Based on this discussion, chats with journalists at the Community Day's evening event and From the Grumpy Old (Re)tired Man: The forums are dead, long live the forums, I began to realize that SDN/ BPX points have a value beyond / outside the community itself. When I hear that some companies research their job applicants on SDN to see how many points they have, I began to consider that these points are much more important than just a way to get gifts from SAP (T-shirts, etc.).
Furthermore, there is also a discussion that the quality of forum posts is decreasing in some forums based on the desire to get as many points as possible with the resulting problem of "point-hunters".
What about bringing both topics together? As Tim suggested in his keynote, self-interest is critical to counter the fact that people won't just volunteer information. Although Tim's focus is on the act of contribution itself, I'd like to take things a step further and attempt to use this self-interest to assure that users contribute quality not just quantity. Here are my suggestions:
- As some have suggested, adding a search functionality before posting a forum entry could help eliminate the mass of repeat posts or questions that are already in a FAQ. I assume that many users will just skip this step and continue to form to create forum posts. What about increasing the number of points possible for a correct answer that was first based on a search of existing SDN sources? Likewise if users don't search before posting, then they should have fewer points to provide for correct answers.
- If we are interested in improving the quality of the forum questions, what about giving the moderator of a forum the ability to award 50 points / month to the forum member who posts the "best" questions in the forum. I hope the forum moderators won't hate for this idea which includes more work for them
Default Network Effects in SDN / BPX
One of Tim's main points dealt with the need to set inclusive defaults for aggregating user data as a side-effect of their use of the site. Let's take a closer look to see how this might apply to SDN / BPX. What are some of the ways that users currently create "data":
- Create, read and respond to forum posts
- Create, read and respond to blogs
- Create, read articles
- ...
There is obviously a pattern here dealing with the interaction of users to the content available.
What about tracking this information and creating patterns / profiles of content usage. If you analyze this information, you might be able to create a system like Amazon's "People who have bought this book have also bought the following books..." where the users who read a blog could be provided with links to other blogs that might be of interest. Of course, the weak point in this idea is that a click on a blog doesn't mean that a user's read it. A click on link is not equivalent to the purchase of a book. However, the purchase of a book doesn't mean that I read or like the book. The Amazon "community" is also not equivalent to the SDN / BPX community in that the SDN / BPX community is already selective in nature based on a common interest relating to SAP technology. Thus, it might be assumed that the analysis of clicks on blog links might be more representative of similar interests.
Inasmuch as SDN forum posts are the main source of user content contribution the question is how to better harness this information to provide value to the community.
For example, what about the idea of analyzing the questions to automatically tag the content and create a "tag cloud" that shows which are the hottest topics in SDN. The technology to analyze questions and relate them to content is present in SAP's OSS search UI (Hey, what about a tighter integration between OSS and forum posts. There could be a new tab in the OSS search UI for relevant forum posts,,,,) Of course, users could also set tags when creating forum posts.
Conclusion
There are probably a variety of other ideas that are possible to better use the information that is currently being created by over 900,000 community members. The question is "How can you make these data sources get better automatically as customers use them?" The fact that readers of the CE forum will probably be interested in CE blogs is obvious. What about linking a forum post regarding WebDynpro usage in CE to blogs on the same topic? It is critical to analyze content on an individual basis. This analysis is of course best performed by users themselves; either consciously - for example, via tags- or unconsciously via an analysis of their click behavior.
