[
https://issues.apache.org/jira/browse/HADOOP-1881?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12532716
]
Doug Cutting commented on HADOOP-1881:
--------------------------------------
> posted on the website[ ...] not in the context of any specific release
That can be fixed. I think things will be much simpler if we keep the
documentation that's maintained with the code in sync with the code, and not
maintain lots of old information there.
One way to fix this would be to restructure the website, splitting the
top-level and project stuff from the documentation, then link to documentation
directories checked out per release. Only the documentation need be bundled
with releases. This is akin to the way, e.g., HTTPD, Tomcat, log4j, etc. do
things.
An easier way would be to simply note the version clearly in the documentation,
and suggest that if you're using a different release then you should refer the
documentation included with that release. Then we need to be careful to only
update that portion of the website when a release becomes the stable release.
This is the way we handle javadoc on the website today (it's only updated when
we make new releases, and clearly states what version it's for). This is also
the way, e.g., Apache Axis handle things.
Documentation changes are permitted after the release freeze (today!) so we
could probably still squeeze a site restructuring into 0.15 even.
> Update documentation for hadoop's configuration post HADOOP-785
> ---------------------------------------------------------------
>
> Key: HADOOP-1881
> URL: https://issues.apache.org/jira/browse/HADOOP-1881
> Project: Hadoop
> Issue Type: Improvement
> Components: documentation
> Affects Versions: 0.15.0
> Reporter: Arun C Murthy
> Assignee: Arun C Murthy
> Fix For: 0.15.0
>
> Attachments: HADOOP-1881-2.patch, HADOOP-1881_1_20071004.patch,
> HADOOP-1881_3_20071005.patch
>
>
> With significant changes to hadoop's configuration since HADOOP-785 the
> documentation for it needs to be completely overhauled:
> a) Exhaustive and accurate javadocs, including some specific examples.
> b) Update the wiki: http://wiki.apache.org/lucene-hadoop/HowToConfigure to
> c) Importantly: Put up a page describing hadoop's configuration on the hadoop
> website (via forrest).
> Any thing else folks can think of?
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
