How can I create two pages with the same name but different parents

[John]

I have a list of topics inside a space and I want each topic to have the same set of sub topics.

e.g.

Topic A

- Subtopic A

- Subtopic B

Topic B

- Subtopic A

- Subtopic B

However, this is not possible with Confluence. How can I achieve the desired organization or work around the oversight?

[Nic Brough -Adaptavist-]

There are some add-ons which effectively mask the different names and make them look the same, but that's about it really.

It's a bad idea to duplicate names, it ruins search for the users.

[John]

Why do duplicate names ruin search? It would be very convenient if this was supported by Atlassian without an add-on. It's a very natural expectation and I was suprised when it wasn't supported.

[Nic Brough -Adaptavist-]

Very simple.  When I search for a specific thing like "subtopic B" in a space.

Results:

Subtopic B

Subtopic B

Which one do I want?  There's no way to tell. 

Also, imagine structured documentation.  I've got 900 applications, each one documented in a single dedicated space.  I'm in operations, and I want to find a run book for application X.  With duplicated names, which https://confluence/display/<application>/runbook do I need?  (And how would the same url give you different pages?)

[John]

When you search for subtopic B, the results will say

Topic A / Subtopic B

Topic B / Subtopic B

...

which is exactly what i want to see. I want to see all instances of subtopic B, when I search for subtopic B.

If you want to find the runbook for application X, you will search for "application X runbook", and the results will say

Application X / runbook

If you want to see all runbooks, you will search for runbook, then pick the runbook of the application that interests you.

Obviously, two different pages will need to have different URLs. One example is

baseurl/topic_A/subtopic_B

baseurl/topic_B/subtopic_B

Both pages are called "subtopic_B", but they have two different URLs because they have different parents.

[Nic Brough -Adaptavist-]

That's not how Confluence does its urls, so that's a non-starter, and the search doesn't give you the tree, so that's wrong as well.

And why should I have to remember the whole structure of the tree just to enter a url?  Knowing that application/runbook should be enough.  application/bits of tree/runbook fails to answer "which runbook do I want"?

You'll need to rewrite the whole of Confluence to do it that way.

[John]

That's not how Confluence does its urls, so that's a non-starter

I'm in favor of Confluence changing its URL system to support this feature. Of course supporting a new feature will require changes. Why is this a "non-starter"?

the search doesn't give you the tree, so that's wrong as well

The search should give you the tree. This can't be difficult to add.

And why should I have to remember the whole structure of the tree just to enter a url?

If duplicate page names are not allowed, I am forced to make longer page names to disambiguiate pages. Besides, how many navigate by typing out full URLs? Is this a cherished feature? Those that do, (and I do this a lot) probably rely on browser autocomplete to find the correct URL.

Knowing that application/runbook should be enough.  application/bits of tree/runbook fails to answer "which runbook do I want"?

As it stands, duplicates are not allowed, so we have pages named "runbook Application X" and "runbook Application Y". Let's say they are at the root. When I search for runbook, I will get both of those results. I don't see how this changes when two pages named "runbook" are allowed. If you know the application whose runbook you want, you can filter by the application. If you don't know which runbook you want, then yes, you can't pick a unique runbook page... I don't understand your point here.

You'll need to rewrite the whole of Confluence to do it that way.

I'm not sure how true this is, but if so, sounds like a design oversight on Confluence's part (and I would still encourage this feature to be included).

[Admin2]

One workaround could be: if Topic A and Topic B were in separate spaces, then both could have a set of sub-topics with identical names.  The new search feature insists on defaulting to searching all spaces so a search for 'sub-topic B' would find both and show which was which by highlighting the relevant space.

Not quite sure why duplicate page names (but under different parents) is not allowed in any case as every page seems to have a unique number in its URL ... move the page as often as you like, even between spaces, and this number seems to stay the same.

Confused.com ...

[John]

Uh, Confluence should also support deleting of answers. I accidentally posted my reply here initially.

[John]

Hm, I also think my comment reply got deleted.

[Nic Brough -Adaptavist-]

It's a non-starter because you'd need to rewrite Confluence to do it.

As Confluence was designed in this way, including a deliberate decision to not to allow duplicate names to avoid all the problems they cause, it's very unlikely to change.

And yes, we do type out urls , because it's a lot quicker than searching, especially when they're usefully short enough to do so.

[James Brown]

I personally have confidence that this could be done in a future version of Confluence because they have done a "rewrite" already for new features.

So the page tree and breadcrumbs are for aesthetics and not truly a page tree stored in the database that maps to a space structure that is rendered into a url...and further more how does confluence track the page when a page is moved under a different page.

If by "we" you mean yourself type out urls, that is your preference unless you are talking on behalf of the entire Confluence user base. I personally use a combination of bookmarks, hyperlinks from other pages/page tree/breadcrumbs and search results.

[Nic Brough -Adaptavist-]

Actually, there's a hint on a recent CR that they're thinking of binning the structure completely, which would make unique names completely mandatory.