{tag}
(Idea-in-progress)
Tag is an inline macro for attaching metadata to a page. Tags are simple name-value pairs, that may optionally be assigned to a user.
| arg | name | description |
|---|---|---|
| 1 | name | (required) The tag name |
| 2 | value | (optional) The tag value |
| 3 | tag-name | (optional) The username of the person who "owns" the tag |
For example:
{tag|todo|The wording for this could be improved|cmiller}
Tags are rendered in the page as a smallish "note" icon with a mouse-over containing the details of the tag. The rendered tag is also an anchor (<a name="tag-$tagId"></a>) so it can be referred to directly from outside the page.
Tag names and values are case-insensitive (when searching), but case-preserving (when displaying). Leading and trailing whitespace is stripped.
Confluence extracts tags from pages when they are saved, and stores them in the same sort of way that links are stored. A search interface allows queries (and RSS feeds) to be run against this tag metadata. This allows users to come up with their own workflow on an ad-hoc basis, without Confluence imposing any real structure. We could document an example of use that goes something like this:
- Show all "todo" items on the documentation staging site assigned to me
- The user then edits those documents, removes the todo tags, and puts a "needs-review" tag in
- Show all documents with a "needs-review" tag in them
- Someone else reviews the document, and either puts in more todo tags, passing them back to the original author, or replaces the "needs-review" tag with a "publish" tag.
- Show all documents with a "publish" tag in them
- Every so often, all documents marked ready for publishing are copied to the live site.
This would allow a flexible way for users to create their own simple workflow without Confluence imposing too much structure. The more structure we try to impose, the more we end up re-implementing half of Jira inside Confluence to manage pages as issues ![]()
See snipsnap's support for labels
Backpack
Hi guys, you may already know Backpack, if you dont, really check these videos showing the use of metadata, notes, album creation, and more. Its really simple, useful, and a cool demo of the things you can accomplish using ajax: http://www.backpackit.com/.

Comments (3)
Jul 04, 2005
Jed Wesley-Smith says:
1. Personally speaking I can see a lot of value in adding semantics to pages. I...+1.
Personally speaking I can see a lot of value in adding semantics to pages. It is not simply a matter of replicating JIRA inside Confluence (although if meta-data functions are added intelligently then that should be possible...) as JIRA starts from a fairly solid semantic definition. You have Projects that have Components and Versions, that are assigned Issues that may move through various workflows. Page semantics should be a lot more flexible, and potentially allow a much more interesting set of additional application-like functionality to be built on top of Confluence.
Feb 13, 2006
Amit Kumar says:
1 Certainly. This tag would be such an asset for Knowledge Management inside wi...+1 Certainly.
This tag would be such an asset for Knowledge Management inside wiki, think about keywords with type attribute and using title for h1., title[@type="sub"] for h2. and wow if these can be indexed with the tagname as the lucene field. I am thinking about FAQ's, Knowledge Bases I would love to help in this process if you are all still seriously considering implementing this macro/plugin. It would be an ideal thing for my employer Library Information Sciences school to have.
Feb 13, 2006
Andy Armstrong says:
I've been thinking a great deal about metadata in Confluence, so I decided to go...I've been thinking a great deal about metadata in Confluence, so I decided to go ahead and implement a plugin. I've submitted an early version to the Confluence Codegeist competition so I'd love to get any feedback.
You can find my entry here:
http://confluence.atlassian.com/display/CODEGEIST/Metadata+Plugin