So it has come up that we should endevor to replace the existing stale social text wiki with something more useful to the community at large. This is a page describing how that project might work.

The actors so far:

• Chris Prather (perigrin)
• Mark Keating (mdk)
• Jess Robinson (castaway)
• Christian Walde (Mithaldu)
• Matt Trout (mst)

Mandate

==

[[perigrin]]

mdk: out of curiosity what is the mandate for the wiki?

[[perigrin]]

is it simply a resource of information that anyone can update?

[[mdk]]

Update it so it isn't a steaming pile of out dated crap - I guess user resource of all things Perl 5 or Perl 6

[[perigrin]]

How is that mandate different than Perl.org or learn.perl.org?

[[mdk]]

perigrin: it should function like Wikipedia for Perl, a huge user-managed and community-controlled repo of anything Perl related

==

Obviously this mandate is fairly large and overlaps extensively with several other projects. We will need to get buy-in from several projects to merge their efforts with our own (or at the very least allow us to import their work wholesale and provide patches back).

"The Plan"

The following is my [perigrin] rough idea on how to achieve the mandate.

1. Take the existing content from the P5 Wiki and move it into a git repository. The important point is to make sure the repo is publishable by Gollum, Jekyll, Blawd, Ikiwiki or another such tool. At this stage I would suggest using a static site generator without any editor fields in the frontend such as Blawd or Jekyll over a wiki engine like Gollum or Ikiwiki. The advantage to a static generator at this point is that it limits the effort required to keep Spammers at bay until we have built a sufficient community to deal with them. (See also the work I had to do on this wiki when I first logged in to clean up spam links). I (perigrin) would suggest this repository is hosted by a volunteer organization (ShadowCat for example) and mirrored onto GitHub, similar to how the Moose project is done.

2. Work with the other documentation communities (Catalyst, Moose, Perl5 Core, DBIx::Class all come to mind) to integrate their work with the new repository. Either as sub-projects that are merged back and fort (find volunteers to govern this if possible), or by simply folding their projects into the new project.

3. Work to expand the community of contributors beyond those already participating in projects. In my (perigrin) opinion, it's only at this stage that the value of having a wiki really trumps having a static site generator with a liberal commit bit policy.