Skip to content
GitLab
Explore
Sign in
Primary navigation
Search or go to…
Project
R
Release
Manage
Activity
Members
Labels
Plan
Issues
Issue boards
Milestones
Wiki
Code
Repository
Branches
Commits
Tags
Repository graph
Compare revisions
Snippets
Deploy
Releases
Model registry
Monitor
Incidents
Service Desk
Analyze
Value stream analytics
Contributor analytics
Repository analytics
Model experiments
Help
Help
Support
GitLab documentation
Compare GitLab plans
Community forum
Contribute to GitLab
Provide feedback
Keyboard shortcuts
?
Snippets
Groups
Projects
Show more breadcrumbs
Development
Release
Commits
5f6683be
Commit
5f6683be
authored
6 years ago
by
totten
Browse files
Options
Downloads
Patches
Plain Diff
Update merger-intro.md
parent
c4af4267
No related branches found
Branches containing commit
No related tags found
No related merge requests found
Changes
1
Hide whitespace changes
Inline
Side-by-side
Showing
1 changed file
doc/merger-intro.md
+9
-7
9 additions, 7 deletions
doc/merger-intro.md
with
9 additions
and
7 deletions
doc/merger-intro.md
+
9
−
7
View file @
5f6683be
...
...
@@ -117,18 +117,20 @@ Secondly, there's the permutations of how CiviCRM users build on the software:
*
Specialize in data migration/ETL or mobile apps or Python or NodeJS -- and rely heavily on REST integration.
Our current practices are shaped by experiences in previous eras. Generally, there has been a lot of feedback that
upgrading can be difficult, and the processes have matured in response. I look at this a gradual transition
between
upgrading can be difficult, and the processes have matured in response. I look at this
as
a gradual transition
among
three eras:
1.
First Era: A large number of developers had direct commit access. Reviews were not done on a patch-by-patch basis,
and patches could accumulate over a period of 3-9 months. Instead, there'd be a subsequent 3 month period of
alpha/beta testing -- followed by another 3-6 month period of post-release stabilization.
1.
*First Era (SVN)*
: A large number of developers had direct commit access. Changes were committed continuously to the trunk.
Review/QA was asynchronous and post-commit. We had some very good QA'ers, but we also had a structural problem: once
something was committed, we lost a lot of leverage for getting QA done -- which led to a long stabilization
period (several months of alpha/beta plus several months of post-release fixup).
2.
Second Era: A small
er
number of developers had merge rights. There were reviews, but they varied widely depending
on the reviewer. Reviews were always performed by the merger. Some mergers would give a light skim
and approve
2.
*
Second Era
(Early Github)*
: A small number of developers had merge rights. There were reviews, but they varied widely depending
on the reviewer. Reviews were always performed by the merger. Some mergers would give a light skim
(because they'd expect a subsequent period of alpha/beta testing) while other mergers were meticulous about testing
related use-cases.
3.
Third Era: A small
er
number of developers had merge rights. Reviews become more distributed among a broader
3.
*
Third Era
(Later Github)*
: A small number of developers had merge rights. Reviews become more distributed among a broader
set of contributors -- mergers act as facilitators and seek to enforce more particular norms.
We're currently in the third era -- where mergers are facilitators who provide a final check.
This diff is collapsed.
Click to expand it.
Preview
0%
Loading
Try again
or
attach a new file
.
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Save comment
Cancel
Please
register
or
sign in
to comment