The Hadoop project, including ZooKeeper, is currently using the Apache wiki. There are no plans at this time to use confluence:


Hiram Chirino wrote:
On Tue, Jul 22, 2008 at 2:54 PM, Patrick Hunt <[EMAIL PROTECTED]> wrote:
I cloned core/hbase subprojects, so we do what's currently done by other
hadoop subs. The issue has come up in the past, might be a good idea at some
point but might also add even more complexity to already pretty complex
process (see howtorelease page). I think we should get things setup similar
to existing subs (still alot of work to be done converting the wiki over to
apache wiki/docs) and review at some point in the future.

Howtocontrib/howtorelease/faq is already on the wiki:

mailinglists/news/etc... are part of the docs (not wiki), probably more
convention than anything. mailinglist never changes and I believe news (and
prolly some others) is part of docs in order to be versioned, but also to
enable users who have checked out the code to have direct access .

That's cool too.. If you are going to use a Wiki it might be better to
use the confluence one just cause it can easily be skinned to have the
main site's look and feel.  Plus it has a bunch of cool macros
available to import code examples from svn or bring in a table of
issues from JIRA (for example ) etc.



Hiram Chirino wrote:
On Tue, Jul 22, 2008 at 1:24 PM, Patrick Hunt <[EMAIL PROTECTED]> wrote:
Hi Hiram, see ZOOKEEPER-70 & 72. We'll be following Hadoop core/hbase wrt
the docs/site/wiki. Specifically using Apache Forrest for the
which also generates the site. We're in the process of moving the
sourceforge wiki content over from SF to apache.

I'm been working with a couple of the hadoop core team members getting
input re the site, I had actually planned to push something live today.

Additional complexity is due to the fact that on SF we had all docs on
wiki and were not including the api docs in the release. Apache requires
to version our documentation proper along with the code (so has to be in
and not on the wiki) and we also have to include the api docs (as well as
regular docs) along with each release which we are not doing on SF.
Yep.  BTW you can still take a hybrid approach where the general site
information (stuff like how to contirbute, mailing lists, news, etc.
etc.) is wiki driven which then links to release documentation
directories which are generated from svn sources during a release.
But either way is good.  Good to know the site is coming soon!


Hiram Chirino wrote:
Is the ZooKeeper website getting moved over to Apache?  I think a good
website is critical for building a community around the project.

Do you guys want to generate it from wiki content?  For example, the site is generated from the wiki content at

If so I think I can create the confluence space for you guys so that
we can get started with that.

Reply via email to