Greetings
I cannot tell you how delighted I am to see enthusiasm for the idea
of creating a Stata Wiki. Having seen what 4-5 people could create via
the UCLA ATS Stata web site, I have wondered how much more could be
created if the skill and enthusiasm of the hundreds (or thousands) of
members of the StataList community could be channeled into creating a
Wiki for all to share and enjoy. While I believe that the Stata
documentation is truly excellent, a Stata Wiki could be an adjunct
(not a replacement) for the documentation, providing additional
information much as the Stata FAQs and the UCLA ATS resources. Having
been the architect of the UCLA ATS site, may I offer some thoughts?
First, I think the biggest challenge is coming up with the
conceptual model for the site. This is one issue that I struggled with
very hard with respect to the UCLA Stata site, and honestly I am not
sure I made the right decision. If I could go back 7 years in time, I
think I would tell myself to organize the site by topic, not by
function. At first organizing the information by function (e.g. FAQ,
Learning Module, Class Notes) was a useful structure, but over time as
the number of pages blossomed, it became harder to find information
because it was not organized by "topic". Yet, at the same time, trying
to organize the information strictly by "topic" is very difficult
because many pages fall under multiple topics. While I applaud the
idea of moving forward on creating a Stata Wiki, at the same time I
think it would be very useful to create a kind of "mission statement"
for the Stata Wiki and create a conceptual model for the Wiki that
provides the proper hooks or nooks for the kind of information that
will be stored there. For example, the SPSS Wiki has a clear
conceptual model, it is organized according to SPSS Procedures.
However, while the conceptual model is exceedingly clear, I think that
it was very UNinviting because after a year it is still 99% empty. I
believe this structure, while clear, is very inhibiting of user
contribution, making it appear that it merely duplicates the manual.
Second, when creating the UCLA web site, one of the greatest
challenges was "seeding" the site with enough information to make it
useful. I think this is even more important for a Wiki, given that
those who visit the site also are those who create the site. If the
site is merely a shell, then there may be no incentive to visit, and
hence no incentive to create. When getting the UCLA web site started,
many gave us permission to "copy" their pages (with credit) from other
sites and place them on the UCLA site. This helped us get enough pages
to reach a critical mass to generate traffic for the site and get the
project off the ground. The Stata FAQs seem to be an excellent
resource for seeding such a Wiki (especially since the folks at Stata
Corp have offered such pages). Another possible resource is the UCLA
ATS Stata site (please note that I am no longer with that group, so
you would need to contact them to see if they are amenable, but this
would seem consistent with their copyright policy). However, it may be
tempting to create a structure (with respect to the first point) that
is oriented towards the "seed" pages, and does not sufficiently look
towards new contributions. The structure should not only accommodate
the seed pages, but also include good hooks for new kinds of
contributions.
May I suggest that an important hook be to try to capture some of
the "best of" contributions to the StataList. For example, some Stata
FAQs have been based on the contributions from StataList. Some of the
contributions here would form an excellent starting point for a Wiki
entry. Then, others would be able to edit, contribute, and extend the
thoughts.
It is difficult to express how much I admire the expertise,
knowledge, and generosity of the StataList community and, because of
that how very strongly I believe that a Wiki is an *excellent* way to
create an enduring knowledgebase that would be wonderful and enduring
resource for the Stata community.
I hope this is helpful.
Best regards,
Michael Mitchell
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
