Some questions and thoughts on the ID of a document

The ID of a document is - I assume - a hash from the “fullname”, as it changes when you rename or move a document.
If a document is created again at the same place with the same name, it has the same ID.

Correct ?

If I now want to keep track of the complete history of a document, I have the problem that renaming the original document and creating a document with the same name restarts the history and the fresh document has the same ID.

I would then have to find out if the “fullname” has already been used in the history of any other document. If the original document has been renamed - legally - several times, this can all become very complicated.

First of all, one needs an identification feature of a document that is immune to renaming and moving. (Interestingly, the Like application somehow manages this.) A “free running counter” or a “unique id” that would be assigned each time a document is created would probably be sufficient.

Keeping track of changes to the ID of a document with a timestamp would complete the matter.

Well, I think that this complete form of keeping track of all changes is certainly only needed in practice for special documents. (Tracking all changes of “Test” in the “Sandbox” would probably have entertainment value!)

Well, in summary: For certain workflows, especially in administrations, this tracking of changes would be very practical, which could be activated, for example, when a document is first created.

I welcome all comments, especially where I am wrong.


Close enough, it’s a hash from the “fullName” and the “locale” (since each document translation has it’s own entry in the document table).

If you’re curious, see:

Not sure if this is of any help (you would need to integrate this into the template to use and it will not update the existing pages automatically)

or any external function / service that will generate unique ID’s. I use something like this for a database like app in xWiki.