incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Steve Loughran <>
Subject Re: Flume Graduation (was Re: June reports in two weeks)
Date Sun, 27 May 2012 13:53:13 GMT
On 27 May 2012 00:43, Jukka Zitting <> wrote:

>    arvind  - 116 commits - Cloudera
>    prasadm -  22 commits - Cloudera
>    brock   -  16 commits - Cloudera
>    esammer -   4 commits - Cloudera
>    jarcec  -   1 commit  - AVG Technologies
>    juhanic -   1 commit  - CyberAgent
> The only two non-Cloudera commits were pretty trivial changes (a
> plugin version upgrade [3] and a spelling fix [4]). My earlier
> classification of Flume as "ready to graduate" [5] was partially based
> on these new committers from outside Cloudera, but the above data
> suggests that they haven't been as active as I would have hoped.
> IIUC Flume operates under an RTC model where people are not supposed
> to commit their own changes, which obviously makes the above data less
> relevant for evaluating the true diversity of the community. However,
> seeing only a single trivial commit by both jarcec and juhanic even
> though they became committers already over three months ago does seem
> to suggest that they may not be as comfortable in their committer role
> as people from Cloudera are.
> Do you think this is a problem for the community? If yes, how do you
> plan to fix it? If not, why?
I think RTC is a barrier to expansion/participation. I understand why it is
viewed as critical in the core codebase of Hadoop -it's a key piece of code
for the big web companies- but do you need RTC on any project in

RTC creates a bias towards commits from organisations with >1
developer/committer, as its easier to ask a colleage over the desk to say
"review this" than it is to stick something up and hope that it gets
noticed by others.

  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message