On a personal level, I'd like to see this in 2.1 because it's a great 
compression algorithm to replace gzip and complement lz4. However, I agree with 
@lindong28 that it's important to have a fair process for all features. As it 
happens, this is a bit of a grey area. The time-based releases wiki page 
(https://cwiki.apache.org/confluence/display/KAFKA/Time+Based+Release+Plan) 
says:

> A month before the release date, the release manager will cut branches and 
> also publish (preferably on the wiki) a list of features that will be 
> included in the release (these will typically be KIPs, but not always). We 
> will leave another week for "minor" features to get in (see below for 
> definitions)
...
> Minor Feature - Feature that takes less than 2 weeks to develop and 
> stabilize. Requires mostly one PR to complete the feature.

This PR seems to be low risk unless zstd is used, it's a single PR, it's medium 
sized (~+600 lines) so it _could_ be classed as a minor feature. The bit where 
it should take less than 2 weeks to develop and stabilize is less clear though 
since development was not all done in one go.

Either way can be justified. @lindong28, I think you should make the call as 
the RM.

[ Full content available at: https://github.com/apache/kafka/pull/2267 ]
This message was relayed via gitbox.apache.org for [email protected]

Reply via email to