Human readable page ID (slug) is important feature that drastically simplify maintenance of complicated multi-space project.  It's easier to maintain consistency of links like  .../PROJECT/Release+Info than ../pages/viewpage.action?pageId=341003791

But I find very confusing, that confluence auto-generate this human-readable slug from the page title and it's not possible to update slug manually or change page title later without breaking links. I.e. I would like to have a page .../PROJECT/Risks+Issues with the title Risks and Issues.

hi

It's a bit late for me to say "me too" to a comment almost 3 years old, but... I agree! Broken bookmarks and links are worse than not being able to tell at a glance where the URL goes. , I change titles all the time (fixing typos, changes of focus, etc.) especially when I'm setting up a new documentation system. Curiously, the pageID is visible on the Edit page. It would be so great to make URL style a setup option.

Suddenly I want to put a DOC: prefix in every title just to force a numbered ID. But that would break too many links.

Coming from a technical background, this sounds crazy to me. 

For instance, we are looking to migrate to Confluence for our documentation needs but it is unclear where everything will reside in the Confluence structure or what titles the pages will have.  This human readable "feature" breaks bookmarks, links and emails during the movement period.

It is crazy that renaming the content page would break someone with a bookmark.

Since there is a already fallback case for the l18n standard and uncodable characters such as punctuation, it wouldn't be that difficult to provide administrators (like myself) with the power to switch to the ?pageId=XXXX thus providing the stability for users to change the names and move content around as they see fit.

Mate,

Sorry about this - we don't like the id=x syntax for URLs either. It's used as little as possible, and only when the page title will not create a nice URL or something conflicts with the link syntax. The big example (as here) is when the page name contains a - character. Another example is a lot of i18n characters will cause a lapse back to the id=x syntax.

When we rewrite the renderer we will be able to get rid of the punctuation character examples, but not the i18n characters (there is no i18n standard that works for URLs).

Cheers,
Mike