After my workshop at Flock about the Fedora Wiki, we discussed many ways to improve and reorganize the wiki, and we sat down with Brian Exelbierd and Peter Travis to discuss it more thoroughly and discuss how the wiki should operate from now on and keep pages and content in general, organized.

So let me sum up the things that we wrote on the pad, and feel free to visit it yourself here[1]:

Wiki consists many things, it's a big data dump, and the first thing we have to do is to agree what it's for. For example, wiki carries information about Fedora's subprojects, documentation, project mission statements, release notes, budgets, meeting minutes, acting like a scratchpad for ideas etc. So at first we need to reconsider what the wiki stands for. We suggest that we should be more selective of what pages exist in the wiki. For example, the docs team is already transfering content to the Docs website, so eventually no documentation will exist on the wiki. Release notes, and mission statements could be moved to the getfedora.org website. Eventually it would be ideal to have the wiki only for information about the subprojects (with links to join.fp.org), working groups, and some more. We still have to figure out what to do with all the pages that act like a "scratchpad", but that's another discussion.

Besides that, old content needs to be archived, categorized, or be deleted. So we came up with a system, that will work with some additional plugins that some people already said that they can write them, so we can have the wiki constantly updated with pages that are categorized, and carry updated content.

We are thinking of developing a plugin, that will test for minimum desired page configuration in a non-blocking manner, and more specifically it will check if the page has a category and if the page is linked from another page within the wiki.
- Pages that have not passed the above 2 tests, will get a "This page needs help" note at the top, with a note that if the page doesn't get a category or is orphaned (not linked from another page within the wiki), it will be archived eg after 3 months. If the page gets a category in the meanwhile, the note will be removed
- Pages that contain this note for more than 1 year, eventually are removed
Apart from that, we can have cronjobs that make various checks and tasks, like:
- Pages that have not been edited in eg 3 months and it does not have a category and a link to it, gets archived.
- Pages that have not been edited in eg 1 year get a note that says "This page has not been edited in a year, it may be out of date."
- Pages that have not been edited in eg 1.5 years get a note that says "This page has not been edited in 1.5 years and is a candidate for archiving"
- Pages that have not been edited in 2 years get archived
  - Archived pages are not searched by wikisearch or external engines by default
  - Archived pages remain available to be accessed via links
  - Archived pages have an "archived" header
  - Archived pages that have no category and no links to them are candidates for manual deletion by wiki-admins
- We can also have a category/tag that will prevent a page from being archived automatically
  - Pages to this category/tag can be added manually
  - Excepted pages are considered "owned" by their last editor.
  - Excepted pages that do not have an active FAS account owner are removed from the exception list and made subject to the archive policy (with a minimum of one year notice).
  - All users are notified of exception pages they own once per year.
  - Excepted pages are subject to review by wiki-admins.

I know that people have already offered their help to develop the necessary plugins and cronjobs in order for something like this to run. Are there any concerns or any additional comments about this? If so I'd like to invite you all and have a discussion, and then move on in to doing this!

[1]http://piratepad.net/1h1o9LNeAC

Zacharias Mitzelos

GPG key ID: B345D18D