[DISCUSS] The need for code reviewers

[Geertjan Wielenga <ge...@apache.org>]

Hi all,

We have a beautiful PR for Payara integration, Java EE 8 integration, and
more.

But we're not going to be able to include those PRs if no one reviews them.

How can we solve this situation? Possibly a tribe concept for reviewing
PRs, similar to the tribe concept in NetCAT? I.e., the tribe name (e.g.,
'jakarta' or 'javascript' or 'profiler' or something) would be a tag in the
PR, created by the author of the PR, signalling to the relevant tribe that
they should review the PR?

And a tribe wouldn't need to be massive, i.e., quality more important than
quantity, probably 3 or 4 per reviewer tribe would be all that's needed --
all we need is a small subset of people who we can ping if no one reviews a
PR and so that reviewers have a way to focus on the PRs that are most
relevant to them.

Thoughts?

Gj

RE: [DISCUSS] The need for code reviewers

[Eric Barboni <sk...@apache.org>]

Sounds a good idea. 
 
 How can we make it "fair" ? 
(maybe tribe member is not skilled enough, not knowing the better pattern and let the pr to be merged compared to another member that will ask for modification) 
 How can we prevent too many people involved in too many tribe ? 
(because if an individual is in every tribe, this will be the same issue having none tribe,no more time to review)

Regards
Eric
-----Message d'origine-----
De : Geertjan Wielenga <ge...@apache.org> 
Envoyé : lundi 17 juin 2019 14:53
À : dev <de...@netbeans.apache.org>
Objet : [DISCUSS] The need for code reviewers

Hi all,

We have a beautiful PR for Payara integration, Java EE 8 integration, and more.

But we're not going to be able to include those PRs if no one reviews them.

How can we solve this situation? Possibly a tribe concept for reviewing PRs, similar to the tribe concept in NetCAT? I.e., the tribe name (e.g., 'jakarta' or 'javascript' or 'profiler' or something) would be a tag in the PR, created by the author of the PR, signalling to the relevant tribe that they should review the PR?

And a tribe wouldn't need to be massive, i.e., quality more important than quantity, probably 3 or 4 per reviewer tribe would be all that's needed -- all we need is a small subset of people who we can ping if no one reviews a PR and so that reviewers have a way to focus on the PRs that are most relevant to them.

Thoughts?

Gj


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@netbeans.apache.org
For additional commands, e-mail: dev-help@netbeans.apache.org

For further information about the NetBeans mailing lists, visit:
https://cwiki.apache.org/confluence/display/NETBEANS/Mailing+lists