So ... I'm seeing the comments fly around that replacing Moin Moin is
the solution to solving our Wiki woes. For your consideration, I submit
the following details. Thus we can all possess the same set of facts to
work from. I'm not arguing for or against a tooling change. I just
don't want Fedora Documentation to end up in a worse position. :)
1. Because of the long-standing Python bias of the Fedora Project, and
the fact that Moin Moin is a popular and active Python-based wiki,
Fedora Docs has long accepted this tool and adapted tools and practices
to fit. Changing out this wiki engine is more than a technical
exercise, and it affects content creation across the Project in the same
way that replacing the buildsystem did for packaging.
* Wiki formatting that can convert to DocBook -- processes
* Wiki to DocBook conversions -- two GSoC projects; small tooling;
* Release note creation process has small ties to Moin Moin way of
* Other WikiEditing processes that are tied to Moin Moin way of doing
2. We have committed two Summer of Code internships to improving facets
of Moin Moin for Fedora Docs, all this work happening upstream.
However, this work has not been picked up entirely in the upstream tree,
so hasn't made it into any releases. We have been suffering from not
having a Python coder who can maintain the Moin Moin DocBook XML code.
If we replace Moin Moin with another wiki, can we be guaranteed of
Fedora Infrastructure support to replace those toolings? And do that in
whatever language the replacement is written in?
* Granted, some wiki communities are very active, so we might be able
to grow a group of contributors around a useful plugin or two, where
maybe we aren't seeing that in Moin Moin
I guess the bottom line is around the challenges we've had in not being
able to get code in the upstream, or having enough knowledge of the
upstream to know when it is safe or proper to run the beta of the next
release. I'd rather see Fedora Infrastructure or Websites commit to
being involved more in the upstream, and fixing our problems there. If
you decide that switching upstreams would do that, I reckon we'll need
to get a deep requirements doc out of Fedora Documentation.
Karsten Wade, Developer Community Mgr.
Dev Fu : http://developer.redhatmagazine.com
Fedora : http://quaid.fedorapeople.org
gpg key : AD0E0C41